Text Generation
Transformers
Safetensors
English
surjo
feature-extraction
recurrent-transformer
linear-attention
gated-deltanet
custom-code
slm
conversational
custom_code
Instructions to use SurjoLabs/Surjo-100m with libraries, inference providers, notebooks, and local apps. Follow these links to get started.
- Libraries
- Transformers
How to use SurjoLabs/Surjo-100m with Transformers:
# Use a pipeline as a high-level helper from transformers import pipeline pipe = pipeline("text-generation", model="SurjoLabs/Surjo-100m", trust_remote_code=True) messages = [ {"role": "user", "content": "Who are you?"}, ] pipe(messages)# pip install -U transformers accelerate # Load model directly from transformers import AutoModel model = AutoModel.from_pretrained("SurjoLabs/Surjo-100m", trust_remote_code=True, device_map="auto") - Notebooks
- Google Colab
- Kaggle
- Local Apps Settings
- vLLM
How to use SurjoLabs/Surjo-100m with vLLM:
Install from pip and serve model
# Install vLLM from pip: pip install vllm # Start the vLLM server: vllm serve "SurjoLabs/Surjo-100m" # Call the server using curl (OpenAI-compatible API): curl -X POST "http://localhost:8000/v1/chat/completions" \ -H "Content-Type: application/json" \ --data '{ "model": "SurjoLabs/Surjo-100m", "messages": [ { "role": "user", "content": "What is the capital of France?" } ] }'Use Docker
docker model run hf.co/SurjoLabs/Surjo-100m
- SGLang
How to use SurjoLabs/Surjo-100m with SGLang:
Install from pip and serve model
# Install SGLang from pip: pip install sglang # Start the SGLang server: python3 -m sglang.launch_server \ --model-path "SurjoLabs/Surjo-100m" \ --host 0.0.0.0 \ --port 30000 # Call the server using curl (OpenAI-compatible API): curl -X POST "http://localhost:30000/v1/chat/completions" \ -H "Content-Type: application/json" \ --data '{ "model": "SurjoLabs/Surjo-100m", "messages": [ { "role": "user", "content": "What is the capital of France?" } ] }'Use Docker images
docker run --gpus all \ --shm-size 32g \ -p 30000:30000 \ -v ~/.cache/huggingface:/root/.cache/huggingface \ --env "HF_TOKEN=<secret>" \ --ipc=host \ lmsysorg/sglang:latest \ python3 -m sglang.launch_server \ --model-path "SurjoLabs/Surjo-100m" \ --host 0.0.0.0 \ --port 30000 # Call the server using curl (OpenAI-compatible API): curl -X POST "http://localhost:30000/v1/chat/completions" \ -H "Content-Type: application/json" \ --data '{ "model": "SurjoLabs/Surjo-100m", "messages": [ { "role": "user", "content": "What is the capital of France?" } ] }' - Docker Model Runner
How to use SurjoLabs/Surjo-100m with Docker Model Runner:
docker model run hf.co/SurjoLabs/Surjo-100m
Download train_logs.json from SurjoLabs/Surjo-100m: direct link, hf CLI and curl.
- Browser
- Download file 120 kB
-
https://huggingface.co/SurjoLabs/Surjo-100m/resolve/main/train_logs.json
- Command line
-
hf download hf://SurjoLabs/Surjo-100m/train_logs.json
-
curl -L -o train_logs.json https://huggingface.co/SurjoLabs/Surjo-100m/resolve/main/train_logs.json
120 kB
| [ | |
| { | |
| "step": 0, | |
| "train_loss": 10.51517790555954, | |
| "val_loss": 10.160391330718994, | |
| "lr": 0.0, | |
| "tokens_per_sec": 2158.630169134002, | |
| "mfu": 0.6606935809350966, | |
| "total_flops": 1161476270194688.0 | |
| }, | |
| { | |
| "step": 50, | |
| "train_loss": 6.364298492670059, | |
| "val_loss": 6.596065902709961, | |
| "lr": 0.01, | |
| "tokens_per_sec": 70222.7689057813, | |
| "mfu": 21.49313639499045, | |
| "total_flops": 5.923528977992909e+16 | |
| }, | |
| { | |
| "step": 100, | |
| "train_loss": 5.122354328632355, | |
| "val_loss": 5.2908248662948605, | |
| "lr": 0.02, | |
| "tokens_per_sec": 70112.78707176603, | |
| "mfu": 21.459474171237453, | |
| "total_flops": 1.1730910328966349e+17 | |
| }, | |
| { | |
| "step": 150, | |
| "train_loss": 4.422913044691086, | |
| "val_loss": 4.639911162853241, | |
| "lr": 0.02, | |
| "tokens_per_sec": 70142.59368173171, | |
| "mfu": 21.46859710306494, | |
| "total_flops": 1.753829167993979e+17 | |
| }, | |
| { | |
| "step": 200, | |
| "train_loss": 3.8527721911668777, | |
| "val_loss": 4.023697471618652, | |
| "lr": 0.02, | |
| "tokens_per_sec": 70024.24338660217, | |
| "mfu": 21.432373538041187, | |
| "total_flops": 2.334567303091323e+17 | |
| }, | |
| { | |
| "step": 250, | |
| "train_loss": 3.522311434149742, | |
| "val_loss": 3.7592015504837035, | |
| "lr": 0.02, | |
| "tokens_per_sec": 69977.49930725315, | |
| "mfu": 21.41806654205168, | |
| "total_flops": 2.915305438188667e+17 | |
| }, | |
| { | |
| "step": 300, | |
| "train_loss": 3.396108627319336, | |
| "val_loss": 3.6023362636566163, | |
| "lr": 0.02, | |
| "tokens_per_sec": 69978.57065144836, | |
| "mfu": 21.418394449185968, | |
| "total_flops": 3.496043573286011e+17 | |
| }, | |
| { | |
| "step": 350, | |
| "train_loss": 3.4941859245300293, | |
| "val_loss": 3.5021621108055117, | |
| "lr": 0.02, | |
| "tokens_per_sec": 70027.76531890646, | |
| "mfu": 21.433451498545583, | |
| "total_flops": 4.076781708383355e+17 | |
| }, | |
| { | |
| "step": 400, | |
| "train_loss": 3.309720605611801, | |
| "val_loss": 3.464106559753418, | |
| "lr": 0.02, | |
| "tokens_per_sec": 70089.33947406986, | |
| "mfu": 21.452297547141427, | |
| "total_flops": 4.657519843480699e+17 | |
| }, | |
| { | |
| "step": 450, | |
| "train_loss": 3.178159549832344, | |
| "val_loss": 3.4234487891197203, | |
| "lr": 0.02, | |
| "tokens_per_sec": 69999.7253376068, | |
| "mfu": 21.424869280100232, | |
| "total_flops": 5.238257978578043e+17 | |
| }, | |
| { | |
| "step": 500, | |
| "train_loss": 3.371193677186966, | |
| "val_loss": 3.3580533504486083, | |
| "lr": 0.02, | |
| "tokens_per_sec": 69956.24699933741, | |
| "mfu": 21.411561831971763, | |
| "total_flops": 5.818996113675387e+17 | |
| }, | |
| { | |
| "step": 550, | |
| "train_loss": 3.199240431189537, | |
| "val_loss": 3.290653383731842, | |
| "lr": 0.02, | |
| "tokens_per_sec": 70012.44780624029, | |
| "mfu": 21.42876325577054, | |
| "total_flops": 6.399734248772731e+17 | |
| }, | |
| { | |
| "step": 600, | |
| "train_loss": 3.006466805934906, | |
| "val_loss": 3.247062921524048, | |
| "lr": 0.02, | |
| "tokens_per_sec": 70007.14506416151, | |
| "mfu": 21.42714024146133, | |
| "total_flops": 6.980472383870075e+17 | |
| }, | |
| { | |
| "step": 650, | |
| "train_loss": 3.0513384342193604, | |
| "val_loss": 3.206310844421387, | |
| "lr": 0.02, | |
| "tokens_per_sec": 70062.60969625264, | |
| "mfu": 21.444116343674445, | |
| "total_flops": 7.561210518967419e+17 | |
| }, | |
| { | |
| "step": 700, | |
| "train_loss": 3.1359871327877045, | |
| "val_loss": 3.163962686061859, | |
| "lr": 0.02, | |
| "tokens_per_sec": 70051.95890783441, | |
| "mfu": 21.44085644874643, | |
| "total_flops": 8.141948654064763e+17 | |
| }, | |
| { | |
| "step": 750, | |
| "train_loss": 3.0369881242513657, | |
| "val_loss": 3.144151282310486, | |
| "lr": 0.02, | |
| "tokens_per_sec": 69949.8472232128, | |
| "mfu": 21.409603047616102, | |
| "total_flops": 8.722686789162107e+17 | |
| }, | |
| { | |
| "step": 800, | |
| "train_loss": 3.091953292489052, | |
| "val_loss": 3.1556016325950624, | |
| "lr": 0.02, | |
| "tokens_per_sec": 69977.76396697006, | |
| "mfu": 21.41814754665293, | |
| "total_flops": 9.303424924259451e+17 | |
| }, | |
| { | |
| "step": 850, | |
| "train_loss": 3.090632662177086, | |
| "val_loss": 3.136023700237274, | |
| "lr": 0.02, | |
| "tokens_per_sec": 70045.18964450127, | |
| "mfu": 21.43878457515927, | |
| "total_flops": 9.884163059356795e+17 | |
| }, | |
| { | |
| "step": 900, | |
| "train_loss": 3.0360343009233475, | |
| "val_loss": 3.1322978258132936, | |
| "lr": 0.02, | |
| "tokens_per_sec": 70022.76171204478, | |
| "mfu": 21.431920040780238, | |
| "total_flops": 1.0464901194454139e+18 | |
| }, | |
| { | |
| "step": 950, | |
| "train_loss": 2.9815044552087784, | |
| "val_loss": 3.1037992238998413, | |
| "lr": 0.02, | |
| "tokens_per_sec": 70034.69091604456, | |
| "mfu": 21.435571221339565, | |
| "total_flops": 1.1045639329551483e+18 | |
| }, | |
| { | |
| "step": 1000, | |
| "train_loss": 3.0359542667865753, | |
| "val_loss": 3.1057387828826903, | |
| "lr": 0.02, | |
| "tokens_per_sec": 70007.97170503181, | |
| "mfu": 21.427393252062473, | |
| "total_flops": 1.1626377464648827e+18 | |
| }, | |
| { | |
| "step": 1050, | |
| "train_loss": 2.9832030683755875, | |
| "val_loss": 3.0936940908432007, | |
| "lr": 0.02, | |
| "tokens_per_sec": 69997.46563636261, | |
| "mfu": 21.424177651618333, | |
| "total_flops": 1.220711559974617e+18 | |
| }, | |
| { | |
| "step": 1100, | |
| "train_loss": 2.9214714765548706, | |
| "val_loss": 3.0613293051719666, | |
| "lr": 0.02, | |
| "tokens_per_sec": 70055.72019986459, | |
| "mfu": 21.442007670264545, | |
| "total_flops": 1.2787853734843515e+18 | |
| }, | |
| { | |
| "step": 1150, | |
| "train_loss": 2.9381714314222336, | |
| "val_loss": 3.0642426133155825, | |
| "lr": 0.02, | |
| "tokens_per_sec": 69989.99293893101, | |
| "mfu": 21.421890477420586, | |
| "total_flops": 1.336859186994086e+18 | |
| }, | |
| { | |
| "step": 1200, | |
| "train_loss": 3.0071898102760315, | |
| "val_loss": 3.063199543952942, | |
| "lr": 0.02, | |
| "tokens_per_sec": 69989.67483553122, | |
| "mfu": 21.421793115270585, | |
| "total_flops": 1.3949330005038203e+18 | |
| }, | |
| { | |
| "step": 1250, | |
| "train_loss": 3.078396648168564, | |
| "val_loss": 3.0594762563705444, | |
| "lr": 0.02, | |
| "tokens_per_sec": 70000.17087793698, | |
| "mfu": 21.425005646968636, | |
| "total_flops": 1.4530068140135547e+18 | |
| }, | |
| { | |
| "step": 1300, | |
| "train_loss": 3.049118548631668, | |
| "val_loss": 3.040031540393829, | |
| "lr": 0.02, | |
| "tokens_per_sec": 69999.04751218345, | |
| "mfu": 21.42466181755635, | |
| "total_flops": 1.511080627523289e+18 | |
| }, | |
| { | |
| "step": 1350, | |
| "train_loss": 3.1371348053216934, | |
| "val_loss": 3.052434301376343, | |
| "lr": 0.02, | |
| "tokens_per_sec": 70121.85376955378, | |
| "mfu": 21.46224922233898, | |
| "total_flops": 1.5691544410330235e+18 | |
| }, | |
| { | |
| "step": 1400, | |
| "train_loss": 2.9777035415172577, | |
| "val_loss": 3.0453516125679014, | |
| "lr": 0.02, | |
| "tokens_per_sec": 70000.47080475416, | |
| "mfu": 21.42509744579814, | |
| "total_flops": 1.627228254542758e+18 | |
| }, | |
| { | |
| "step": 1450, | |
| "train_loss": 2.9373037219047546, | |
| "val_loss": 3.0371802687644958, | |
| "lr": 0.02, | |
| "tokens_per_sec": 69993.52379124102, | |
| "mfu": 21.422971168077932, | |
| "total_flops": 1.6853020680524923e+18 | |
| }, | |
| { | |
| "step": 1500, | |
| "train_loss": 2.8891674876213074, | |
| "val_loss": 3.0314683616161346, | |
| "lr": 0.02, | |
| "tokens_per_sec": 70034.81448436435, | |
| "mfu": 21.43560904198937, | |
| "total_flops": 1.7433758815622267e+18 | |
| }, | |
| { | |
| "step": 1550, | |
| "train_loss": 2.8753224909305573, | |
| "val_loss": 3.0194230318069457, | |
| "lr": 0.02, | |
| "tokens_per_sec": 70032.54236910073, | |
| "mfu": 21.43491361393908, | |
| "total_flops": 1.801449695071961e+18 | |
| }, | |
| { | |
| "step": 1600, | |
| "train_loss": 2.873562440276146, | |
| "val_loss": 3.0389770805835723, | |
| "lr": 0.02, | |
| "tokens_per_sec": 70093.54067527145, | |
| "mfu": 21.453583411994888, | |
| "total_flops": 1.8595235085816955e+18 | |
| }, | |
| { | |
| "step": 1650, | |
| "train_loss": 2.8681159019470215, | |
| "val_loss": 3.0404816269874573, | |
| "lr": 0.02, | |
| "tokens_per_sec": 70092.78506979671, | |
| "mfu": 21.4533521432514, | |
| "total_flops": 1.91759732209143e+18 | |
| }, | |
| { | |
| "step": 1700, | |
| "train_loss": 2.8477375507354736, | |
| "val_loss": 3.03014994263649, | |
| "lr": 0.02, | |
| "tokens_per_sec": 70043.1170883897, | |
| "mfu": 21.438150226330734, | |
| "total_flops": 1.9756711356011643e+18 | |
| }, | |
| { | |
| "step": 1750, | |
| "train_loss": 2.924525633454323, | |
| "val_loss": 3.0115107893943787, | |
| "lr": 0.02, | |
| "tokens_per_sec": 70089.97079499098, | |
| "mfu": 21.45249077601688, | |
| "total_flops": 2.0337449491108987e+18 | |
| }, | |
| { | |
| "step": 1800, | |
| "train_loss": 2.845635160803795, | |
| "val_loss": 3.021808397769928, | |
| "lr": 0.02, | |
| "tokens_per_sec": 70073.76133559164, | |
| "mfu": 21.447529534425474, | |
| "total_flops": 2.091818762620633e+18 | |
| }, | |
| { | |
| "step": 1850, | |
| "train_loss": 2.7774328142404556, | |
| "val_loss": 3.0103733837604523, | |
| "lr": 0.02, | |
| "tokens_per_sec": 70157.2173610318, | |
| "mfu": 21.47307298373286, | |
| "total_flops": 2.1498925761303675e+18 | |
| }, | |
| { | |
| "step": 1900, | |
| "train_loss": 2.7959036976099014, | |
| "val_loss": 3.010297340154648, | |
| "lr": 0.02, | |
| "tokens_per_sec": 70144.7278424267, | |
| "mfu": 21.469250307255273, | |
| "total_flops": 2.207966389640102e+18 | |
| }, | |
| { | |
| "step": 1950, | |
| "train_loss": 2.8740807473659515, | |
| "val_loss": 3.0071927487850187, | |
| "lr": 0.02, | |
| "tokens_per_sec": 70076.6894273708, | |
| "mfu": 21.448425737707787, | |
| "total_flops": 2.2660402031498363e+18 | |
| }, | |
| { | |
| "step": 2000, | |
| "train_loss": 2.964038208127022, | |
| "val_loss": 3.0082029044628142, | |
| "lr": 0.02, | |
| "tokens_per_sec": 69929.21289313334, | |
| "mfu": 21.403287482483492, | |
| "total_flops": 2.3241140166595707e+18 | |
| }, | |
| { | |
| "step": 2050, | |
| "train_loss": 2.812971353530884, | |
| "val_loss": 3.002509093284607, | |
| "lr": 0.02, | |
| "tokens_per_sec": 70071.29176833555, | |
| "mfu": 21.446773672093368, | |
| "total_flops": 2.382187830169305e+18 | |
| }, | |
| { | |
| "step": 2100, | |
| "train_loss": 2.9027664810419083, | |
| "val_loss": 2.985026216506958, | |
| "lr": 0.02, | |
| "tokens_per_sec": 70077.55031615285, | |
| "mfu": 21.44868923059338, | |
| "total_flops": 2.4402616436790395e+18 | |
| }, | |
| { | |
| "step": 2150, | |
| "train_loss": 2.7932936400175095, | |
| "val_loss": 2.9898294627666475, | |
| "lr": 0.02, | |
| "tokens_per_sec": 70081.02769582007, | |
| "mfu": 21.449753554838203, | |
| "total_flops": 2.498335457188774e+18 | |
| }, | |
| { | |
| "step": 2200, | |
| "train_loss": 2.8083430528640747, | |
| "val_loss": 2.979103463888168, | |
| "lr": 0.02, | |
| "tokens_per_sec": 70060.1254019016, | |
| "mfu": 21.443355973809172, | |
| "total_flops": 2.5564092706985083e+18 | |
| }, | |
| { | |
| "step": 2250, | |
| "train_loss": 2.8303453028202057, | |
| "val_loss": 2.9749884247779845, | |
| "lr": 0.02, | |
| "tokens_per_sec": 69815.61494395022, | |
| "mfu": 21.368518471605263, | |
| "total_flops": 2.6144830842082427e+18 | |
| }, | |
| { | |
| "step": 2300, | |
| "train_loss": 2.7061598151922226, | |
| "val_loss": 2.9586526334285734, | |
| "lr": 0.02, | |
| "tokens_per_sec": 70126.97363792777, | |
| "mfu": 21.463816264354005, | |
| "total_flops": 2.672556897717977e+18 | |
| }, | |
| { | |
| "step": 2350, | |
| "train_loss": 2.920316368341446, | |
| "val_loss": 2.9535837829113007, | |
| "lr": 0.02, | |
| "tokens_per_sec": 70137.85909871584, | |
| "mfu": 21.467147985633094, | |
| "total_flops": 2.7306307112277115e+18 | |
| }, | |
| { | |
| "step": 2400, | |
| "train_loss": 2.9253035485744476, | |
| "val_loss": 2.944886934757233, | |
| "lr": 0.02, | |
| "tokens_per_sec": 70066.0209632928, | |
| "mfu": 21.445160432777115, | |
| "total_flops": 2.788704524737446e+18 | |
| }, | |
| { | |
| "step": 2450, | |
| "train_loss": 2.8674067705869675, | |
| "val_loss": 2.940703195333481, | |
| "lr": 0.02, | |
| "tokens_per_sec": 70068.62992160743, | |
| "mfu": 21.44595895863673, | |
| "total_flops": 2.8467783382471803e+18 | |
| }, | |
| { | |
| "step": 2500, | |
| "train_loss": 2.8450111150741577, | |
| "val_loss": 2.9333001136779786, | |
| "lr": 0.02, | |
| "tokens_per_sec": 69838.27776156714, | |
| "mfu": 21.375454897464362, | |
| "total_flops": 2.9048521517569147e+18 | |
| }, | |
| { | |
| "step": 2550, | |
| "train_loss": 2.759008914232254, | |
| "val_loss": 2.9429529547691344, | |
| "lr": 0.02, | |
| "tokens_per_sec": 70104.37553002524, | |
| "mfu": 21.45689964424651, | |
| "total_flops": 2.962925965266649e+18 | |
| }, | |
| { | |
| "step": 2600, | |
| "train_loss": 2.565167874097824, | |
| "val_loss": 2.940195286273956, | |
| "lr": 0.02, | |
| "tokens_per_sec": 70102.36405519309, | |
| "mfu": 21.45628399061171, | |
| "total_flops": 3.0209997787763835e+18 | |
| }, | |
| { | |
| "step": 2650, | |
| "train_loss": 2.830553784966469, | |
| "val_loss": 2.9388512551784514, | |
| "lr": 0.02, | |
| "tokens_per_sec": 70080.79229376302, | |
| "mfu": 21.4496815051512, | |
| "total_flops": 3.079073592286118e+18 | |
| }, | |
| { | |
| "step": 2700, | |
| "train_loss": 2.769469901919365, | |
| "val_loss": 2.9488451719284057, | |
| "lr": 0.02, | |
| "tokens_per_sec": 70073.8045434036, | |
| "mfu": 21.44754275907341, | |
| "total_flops": 3.1371474057958523e+18 | |
| }, | |
| { | |
| "step": 2750, | |
| "train_loss": 2.7408286333084106, | |
| "val_loss": 2.943127530813217, | |
| "lr": 0.02, | |
| "tokens_per_sec": 69843.7630299741, | |
| "mfu": 21.37713377774594, | |
| "total_flops": 3.1952212193055867e+18 | |
| }, | |
| { | |
| "step": 2550, | |
| "train_loss": 2.690155401825905, | |
| "val_loss": 2.753552335500717, | |
| "lr": 0.02, | |
| "tokens_per_sec": 70145.36765726679, | |
| "mfu": 21.499149298759562, | |
| "total_flops": 3.253295032815321e+18 | |
| }, | |
| { | |
| "step": 2600, | |
| "train_loss": 2.7558372765779495, | |
| "val_loss": 2.7880488693714143, | |
| "lr": 0.02, | |
| "tokens_per_sec": 70354.6993534577, | |
| "mfu": 21.563308252368028, | |
| "total_flops": 3.3113688463250555e+18 | |
| }, | |
| { | |
| "step": 2650, | |
| "train_loss": 2.748625800013542, | |
| "val_loss": 2.8070174753665924, | |
| "lr": 0.02, | |
| "tokens_per_sec": 70307.24311361065, | |
| "mfu": 21.548763189455038, | |
| "total_flops": 3.36944265983479e+18 | |
| }, | |
| { | |
| "step": 2700, | |
| "train_loss": 2.657764434814453, | |
| "val_loss": 2.6618025243282317, | |
| "lr": 0.02, | |
| "tokens_per_sec": 70264.88750243967, | |
| "mfu": 21.53578143971706, | |
| "total_flops": 3.4275164733445243e+18 | |
| }, | |
| { | |
| "step": 2750, | |
| "train_loss": 2.597847282886505, | |
| "val_loss": 2.733213084936142, | |
| "lr": 0.02, | |
| "tokens_per_sec": 70309.20634173382, | |
| "mfu": 21.54936490751479, | |
| "total_flops": 3.4855902868542587e+18 | |
| }, | |
| { | |
| "step": 2800, | |
| "train_loss": 2.6393312364816666, | |
| "val_loss": 2.76564337015152, | |
| "lr": 0.02, | |
| "tokens_per_sec": 70335.78134187244, | |
| "mfu": 21.557509991284064, | |
| "total_flops": 3.543664100363993e+18 | |
| }, | |
| { | |
| "step": 2850, | |
| "train_loss": 2.7604094594717026, | |
| "val_loss": 2.7764769434928893, | |
| "lr": 0.02, | |
| "tokens_per_sec": 70224.05195676716, | |
| "mfu": 21.52326558125867, | |
| "total_flops": 3.6017379138737275e+18 | |
| }, | |
| { | |
| "step": 2900, | |
| "train_loss": 2.8098009526729584, | |
| "val_loss": 2.7969186305999756, | |
| "lr": 0.02, | |
| "tokens_per_sec": 70314.45950237487, | |
| "mfu": 21.55097497085446, | |
| "total_flops": 3.659811727383462e+18 | |
| }, | |
| { | |
| "step": 2950, | |
| "train_loss": 2.8605630844831467, | |
| "val_loss": 2.800990068912506, | |
| "lr": 0.02, | |
| "tokens_per_sec": 70267.142723665, | |
| "mfu": 21.536472651973043, | |
| "total_flops": 3.7178855408931963e+18 | |
| }, | |
| { | |
| "step": 3000, | |
| "train_loss": 2.826046869158745, | |
| "val_loss": 2.7932212591171264, | |
| "lr": 0.02, | |
| "tokens_per_sec": 70312.75564909614, | |
| "mfu": 21.550452749683682, | |
| "total_flops": 3.7759593544029307e+18 | |
| }, | |
| { | |
| "step": 3050, | |
| "train_loss": 2.7675381749868393, | |
| "val_loss": 2.809084439277649, | |
| "lr": 0.02, | |
| "tokens_per_sec": 70366.1560541514, | |
| "mfu": 21.566819664837748, | |
| "total_flops": 3.834033167912665e+18 | |
| }, | |
| { | |
| "step": 3100, | |
| "train_loss": 2.6846775263547897, | |
| "val_loss": 2.8106513619422913, | |
| "lr": 0.02, | |
| "tokens_per_sec": 70376.63593211089, | |
| "mfu": 21.570031686791708, | |
| "total_flops": 3.8921069814223995e+18 | |
| }, | |
| { | |
| "step": 3150, | |
| "train_loss": 2.8498146533966064, | |
| "val_loss": 2.8211086273193358, | |
| "lr": 0.02, | |
| "tokens_per_sec": 70318.95745570271, | |
| "mfu": 21.55235356752254, | |
| "total_flops": 3.950180794932134e+18 | |
| }, | |
| { | |
| "step": 3200, | |
| "train_loss": 2.882045492529869, | |
| "val_loss": 2.8087339460849763, | |
| "lr": 0.02, | |
| "tokens_per_sec": 70306.22248511235, | |
| "mfu": 21.54845037272606, | |
| "total_flops": 4.0082546084418683e+18 | |
| }, | |
| { | |
| "step": 3250, | |
| "train_loss": 2.7344588190317154, | |
| "val_loss": 2.8052581429481505, | |
| "lr": 0.02, | |
| "tokens_per_sec": 70309.31255946922, | |
| "mfu": 21.549397462636076, | |
| "total_flops": 4.0663284219516027e+18 | |
| }, | |
| { | |
| "step": 3300, | |
| "train_loss": 2.6739814579486847, | |
| "val_loss": 2.8148413836956023, | |
| "lr": 0.02, | |
| "tokens_per_sec": 70365.00773946103, | |
| "mfu": 21.56646771302974, | |
| "total_flops": 4.124402235461337e+18 | |
| }, | |
| { | |
| "step": 3350, | |
| "train_loss": 2.687133699655533, | |
| "val_loss": 2.7819845259189604, | |
| "lr": 0.02, | |
| "tokens_per_sec": 70311.11920932027, | |
| "mfu": 21.549951190361448, | |
| "total_flops": 4.1824760489710715e+18 | |
| }, | |
| { | |
| "step": 3400, | |
| "train_loss": 2.738806650042534, | |
| "val_loss": 2.777050405740738, | |
| "lr": 0.02, | |
| "tokens_per_sec": 70313.55276381515, | |
| "mfu": 21.550697060732645, | |
| "total_flops": 4.240549862480806e+18 | |
| }, | |
| { | |
| "step": 3450, | |
| "train_loss": 2.8085870444774628, | |
| "val_loss": 2.7993926048278808, | |
| "lr": 0.02, | |
| "tokens_per_sec": 70308.41404623268, | |
| "mfu": 21.549122073528217, | |
| "total_flops": 4.2986236759905403e+18 | |
| }, | |
| { | |
| "step": 3500, | |
| "train_loss": 2.7735735923051834, | |
| "val_loss": 2.822412884235382, | |
| "lr": 0.02, | |
| "tokens_per_sec": 70330.90176386677, | |
| "mfu": 21.556014428860447, | |
| "total_flops": 4.3566974895002747e+18 | |
| }, | |
| { | |
| "step": 3550, | |
| "train_loss": 2.7676771581172943, | |
| "val_loss": 2.816472613811493, | |
| "lr": 0.02, | |
| "tokens_per_sec": 70321.96971640864, | |
| "mfu": 21.553276807998806, | |
| "total_flops": 4.414771303010009e+18 | |
| }, | |
| { | |
| "step": 3600, | |
| "train_loss": 2.7102452218532562, | |
| "val_loss": 2.818382143974304, | |
| "lr": 0.02, | |
| "tokens_per_sec": 70311.3801038398, | |
| "mfu": 21.550031153022044, | |
| "total_flops": 4.4728451165197435e+18 | |
| }, | |
| { | |
| "step": 3650, | |
| "train_loss": 2.7262541949748993, | |
| "val_loss": 2.8222990155220034, | |
| "lr": 0.02, | |
| "tokens_per_sec": 70288.59055530882, | |
| "mfu": 21.543046288266435, | |
| "total_flops": 4.530918930029478e+18 | |
| }, | |
| { | |
| "step": 3700, | |
| "train_loss": 2.655830502510071, | |
| "val_loss": 2.8140450358390807, | |
| "lr": 0.02, | |
| "tokens_per_sec": 70323.33015866883, | |
| "mfu": 21.553693775679513, | |
| "total_flops": 4.5889927435392123e+18 | |
| }, | |
| { | |
| "step": 3750, | |
| "train_loss": 2.7190513610839844, | |
| "val_loss": 2.8193438947200775, | |
| "lr": 0.02, | |
| "tokens_per_sec": 70344.47533367388, | |
| "mfu": 21.560174649464408, | |
| "total_flops": 4.647066557048947e+18 | |
| }, | |
| { | |
| "step": 3800, | |
| "train_loss": 2.7415073215961456, | |
| "val_loss": 2.8139825463294983, | |
| "lr": 0.02, | |
| "tokens_per_sec": 70199.24064193484, | |
| "mfu": 21.51566105683042, | |
| "total_flops": 4.705140370558681e+18 | |
| }, | |
| { | |
| "step": 3850, | |
| "train_loss": 2.731362007558346, | |
| "val_loss": 2.8108940064907073, | |
| "lr": 0.02, | |
| "tokens_per_sec": 70298.58135692739, | |
| "mfu": 21.546108411151888, | |
| "total_flops": 4.763214184068415e+18 | |
| }, | |
| { | |
| "step": 3900, | |
| "train_loss": 2.796496495604515, | |
| "val_loss": 2.810216724872589, | |
| "lr": 0.02, | |
| "tokens_per_sec": 70265.2313500534, | |
| "mfu": 21.535886827021027, | |
| "total_flops": 4.82128799757815e+18 | |
| }, | |
| { | |
| "step": 3950, | |
| "train_loss": 2.636091724038124, | |
| "val_loss": 2.803800421953201, | |
| "lr": 0.02, | |
| "tokens_per_sec": 70308.34795684878, | |
| "mfu": 21.549101817514472, | |
| "total_flops": 4.879361811087884e+18 | |
| }, | |
| { | |
| "step": 4000, | |
| "train_loss": 2.8011103123426437, | |
| "val_loss": 2.829415625333786, | |
| "lr": 0.02, | |
| "tokens_per_sec": 70348.76681625859, | |
| "mfu": 21.56148996404445, | |
| "total_flops": 4.937435624597619e+18 | |
| }, | |
| { | |
| "step": 4050, | |
| "train_loss": 2.7537961602211, | |
| "val_loss": 2.828208714723587, | |
| "lr": 0.02, | |
| "tokens_per_sec": 70338.02334258889, | |
| "mfu": 21.55819715153057, | |
| "total_flops": 4.995509438107353e+18 | |
| }, | |
| { | |
| "step": 4100, | |
| "train_loss": 2.7435563504695892, | |
| "val_loss": 2.8136156499385834, | |
| "lr": 0.02, | |
| "tokens_per_sec": 70324.44382873985, | |
| "mfu": 21.554035109112153, | |
| "total_flops": 5.053583251617087e+18 | |
| }, | |
| { | |
| "step": 4150, | |
| "train_loss": 2.708776831626892, | |
| "val_loss": 2.8181759417057037, | |
| "lr": 0.02, | |
| "tokens_per_sec": 70153.27003630337, | |
| "mfu": 21.50157133220805, | |
| "total_flops": 5.111657065126822e+18 | |
| }, | |
| { | |
| "step": 4200, | |
| "train_loss": 2.6885371059179306, | |
| "val_loss": 2.8155935287475584, | |
| "lr": 0.02, | |
| "tokens_per_sec": 70219.75953179927, | |
| "mfu": 21.521949977843647, | |
| "total_flops": 5.169730878636556e+18 | |
| }, | |
| { | |
| "step": 4250, | |
| "train_loss": 2.6093505173921585, | |
| "val_loss": 2.84355810880661, | |
| "lr": 0.02, | |
| "tokens_per_sec": 70308.2900725415, | |
| "mfu": 21.54908407630923, | |
| "total_flops": 5.227804692146291e+18 | |
| }, | |
| { | |
| "step": 4300, | |
| "train_loss": 2.75183042883873, | |
| "val_loss": 2.8319630563259124, | |
| "lr": 0.02, | |
| "tokens_per_sec": 70308.23432386186, | |
| "mfu": 21.549066989661508, | |
| "total_flops": 5.285878505656025e+18 | |
| }, | |
| { | |
| "step": 4350, | |
| "train_loss": 2.743694692850113, | |
| "val_loss": 2.8407095193862917, | |
| "lr": 0.02, | |
| "tokens_per_sec": 70338.29039891162, | |
| "mfu": 21.558279002748197, | |
| "total_flops": 5.34395231916576e+18 | |
| }, | |
| { | |
| "step": 4400, | |
| "train_loss": 2.6878776401281357, | |
| "val_loss": 2.8353062510490417, | |
| "lr": 0.02, | |
| "tokens_per_sec": 70268.21217729135, | |
| "mfu": 21.536800433321247, | |
| "total_flops": 5.402026132675494e+18 | |
| }, | |
| { | |
| "step": 4450, | |
| "train_loss": 2.7711153477430344, | |
| "val_loss": 2.8343011856079103, | |
| "lr": 0.02, | |
| "tokens_per_sec": 70268.73984328873, | |
| "mfu": 21.536962159896163, | |
| "total_flops": 5.460099946185228e+18 | |
| }, | |
| { | |
| "step": 4500, | |
| "train_loss": 2.770058751106262, | |
| "val_loss": 2.8305183827877043, | |
| "lr": 0.02, | |
| "tokens_per_sec": 70330.05274618218, | |
| "mfu": 21.555754209852864, | |
| "total_flops": 5.518173759694963e+18 | |
| }, | |
| { | |
| "step": 4550, | |
| "train_loss": 2.7573233991861343, | |
| "val_loss": 2.8480226695537567, | |
| "lr": 0.02, | |
| "tokens_per_sec": 70306.38769881551, | |
| "mfu": 21.54850100977009, | |
| "total_flops": 5.576247573204697e+18 | |
| }, | |
| { | |
| "step": 4600, | |
| "train_loss": 2.683898538351059, | |
| "val_loss": 2.842072516679764, | |
| "lr": 0.02, | |
| "tokens_per_sec": 70336.00878633632, | |
| "mfu": 21.557579701696163, | |
| "total_flops": 5.634321386714431e+18 | |
| }, | |
| { | |
| "step": 4650, | |
| "train_loss": 2.4759643375873566, | |
| "val_loss": 2.2887446224689483, | |
| "lr": 0.02, | |
| "tokens_per_sec": 70300.01955342875, | |
| "mfu": 21.54654921005759, | |
| "total_flops": 5.692395200224166e+18 | |
| }, | |
| { | |
| "step": 4700, | |
| "train_loss": 2.796401672065258, | |
| "val_loss": 2.297064536809921, | |
| "lr": 0.02, | |
| "tokens_per_sec": 70308.00739617708, | |
| "mfu": 21.548997437639215, | |
| "total_flops": 5.7504690137339e+18 | |
| }, | |
| { | |
| "step": 4750, | |
| "train_loss": 2.596481144428253, | |
| "val_loss": 2.4246182203292848, | |
| "lr": 0.02, | |
| "tokens_per_sec": 70007.82326929359, | |
| "mfu": 21.456992739730733, | |
| "total_flops": 5.808542827243635e+18 | |
| }, | |
| { | |
| "step": 4800, | |
| "train_loss": 2.595515549182892, | |
| "val_loss": 2.3040037989616393, | |
| "lr": 0.02, | |
| "tokens_per_sec": 70316.84112334541, | |
| "mfu": 21.55170492390095, | |
| "total_flops": 5.866616640753369e+18 | |
| }, | |
| { | |
| "step": 4850, | |
| "train_loss": 2.62885881960392, | |
| "val_loss": 2.126067227125168, | |
| "lr": 0.02, | |
| "tokens_per_sec": 70333.82472782901, | |
| "mfu": 21.556910300401476, | |
| "total_flops": 5.924690454263103e+18 | |
| }, | |
| { | |
| "step": 4900, | |
| "train_loss": 2.7220041751861572, | |
| "val_loss": 2.335285782814026, | |
| "lr": 0.02, | |
| "tokens_per_sec": 70141.45779781595, | |
| "mfu": 21.497950949461835, | |
| "total_flops": 5.982764267772838e+18 | |
| }, | |
| { | |
| "step": 4950, | |
| "train_loss": 2.6748989075422287, | |
| "val_loss": 2.4292496860027315, | |
| "lr": 0.02, | |
| "tokens_per_sec": 70252.59415868955, | |
| "mfu": 21.532013600991682, | |
| "total_flops": 6.040838081282572e+18 | |
| }, | |
| { | |
| "step": 5000, | |
| "train_loss": 2.4526360854506493, | |
| "val_loss": 2.277197653055191, | |
| "lr": 0.02, | |
| "tokens_per_sec": 70309.10619387723, | |
| "mfu": 21.549334212776245, | |
| "total_flops": 6.098911894792307e+18 | |
| }, | |
| { | |
| "step": 5050, | |
| "train_loss": 2.559838131070137, | |
| "val_loss": 2.317918872833252, | |
| "lr": 0.02, | |
| "tokens_per_sec": 70267.75883994007, | |
| "mfu": 21.632433190049806, | |
| "total_flops": 6.156985708302041e+18 | |
| }, | |
| { | |
| "step": 5100, | |
| "train_loss": 2.591684401035309, | |
| "val_loss": 2.399251621961594, | |
| "lr": 0.02, | |
| "tokens_per_sec": 70248.37588972368, | |
| "mfu": 21.626466009901975, | |
| "total_flops": 6.215059521811775e+18 | |
| }, | |
| { | |
| "step": 5150, | |
| "train_loss": 2.69405797123909, | |
| "val_loss": 2.473741942644119, | |
| "lr": 0.02, | |
| "tokens_per_sec": 70290.93863159792, | |
| "mfu": 21.639569255048546, | |
| "total_flops": 6.27313333532151e+18 | |
| }, | |
| { | |
| "step": 5200, | |
| "train_loss": 2.7576390206813812, | |
| "val_loss": 2.5225872576236723, | |
| "lr": 0.02, | |
| "tokens_per_sec": 70268.60523157175, | |
| "mfu": 21.632693757779972, | |
| "total_flops": 6.331207148831244e+18 | |
| }, | |
| { | |
| "step": 5250, | |
| "train_loss": 2.6569458171725273, | |
| "val_loss": 2.5516360282897947, | |
| "lr": 0.02, | |
| "tokens_per_sec": 70275.72239653025, | |
| "mfu": 21.634884828023413, | |
| "total_flops": 6.389280962340979e+18 | |
| }, | |
| { | |
| "step": 5300, | |
| "train_loss": 2.737651005387306, | |
| "val_loss": 2.5737712025642394, | |
| "lr": 0.02, | |
| "tokens_per_sec": 70077.68129369298, | |
| "mfu": 21.573916455091958, | |
| "total_flops": 6.447354775850713e+18 | |
| }, | |
| { | |
| "step": 5350, | |
| "train_loss": 2.610097751021385, | |
| "val_loss": 2.5789273977279663, | |
| "lr": 0.02, | |
| "tokens_per_sec": 70291.56662424024, | |
| "mfu": 21.639762587084505, | |
| "total_flops": 6.505428589360447e+18 | |
| }, | |
| { | |
| "step": 5400, | |
| "train_loss": 2.6154174357652664, | |
| "val_loss": 2.39971661567688, | |
| "lr": 0.02, | |
| "tokens_per_sec": 70280.81837590337, | |
| "mfu": 21.636453661797916, | |
| "total_flops": 6.563502402870182e+18 | |
| }, | |
| { | |
| "step": 5450, | |
| "train_loss": 2.560743674635887, | |
| "val_loss": 2.471098029613495, | |
| "lr": 0.02, | |
| "tokens_per_sec": 70284.76569050449, | |
| "mfu": 21.637668870889517, | |
| "total_flops": 6.621576216379916e+18 | |
| }, | |
| { | |
| "step": 5500, | |
| "train_loss": 2.5454889237880707, | |
| "val_loss": 2.506662356853485, | |
| "lr": 0.02, | |
| "tokens_per_sec": 70268.84998053168, | |
| "mfu": 21.63276910550148, | |
| "total_flops": 6.679650029889651e+18 | |
| }, | |
| { | |
| "step": 5550, | |
| "train_loss": 2.656815826892853, | |
| "val_loss": 2.5483492255210876, | |
| "lr": 0.02, | |
| "tokens_per_sec": 70284.1575875986, | |
| "mfu": 21.637481662051478, | |
| "total_flops": 6.737723843399385e+18 | |
| }, | |
| { | |
| "step": 5600, | |
| "train_loss": 2.691612422466278, | |
| "val_loss": 2.5813338339328764, | |
| "lr": 0.02, | |
| "tokens_per_sec": 70268.87950770035, | |
| "mfu": 21.632778195651987, | |
| "total_flops": 6.79579765690912e+18 | |
| }, | |
| { | |
| "step": 5650, | |
| "train_loss": 2.6659550815820694, | |
| "val_loss": 2.594164401292801, | |
| "lr": 0.02, | |
| "tokens_per_sec": 70075.86473585301, | |
| "mfu": 21.57335721474108, | |
| "total_flops": 6.853871470418854e+18 | |
| }, | |
| { | |
| "step": 5700, | |
| "train_loss": 2.722677171230316, | |
| "val_loss": 2.6006647884845733, | |
| "lr": 0.02, | |
| "tokens_per_sec": 70276.25938345322, | |
| "mfu": 21.635050143296404, | |
| "total_flops": 6.911945283928588e+18 | |
| }, | |
| { | |
| "step": 5750, | |
| "train_loss": 2.780638873577118, | |
| "val_loss": 2.627731019258499, | |
| "lr": 0.02, | |
| "tokens_per_sec": 70283.08787318766, | |
| "mfu": 21.637152342803084, | |
| "total_flops": 6.970019097438323e+18 | |
| }, | |
| { | |
| "step": 5800, | |
| "train_loss": 2.6606620252132416, | |
| "val_loss": 2.639297807216644, | |
| "lr": 0.02, | |
| "tokens_per_sec": 70288.14931152055, | |
| "mfu": 21.638710542870793, | |
| "total_flops": 7.028092910948057e+18 | |
| }, | |
| { | |
| "step": 5850, | |
| "train_loss": 2.961484670639038, | |
| "val_loss": 2.6399263024330137, | |
| "lr": 0.02, | |
| "tokens_per_sec": 70300.75165277069, | |
| "mfu": 21.6425902639496, | |
| "total_flops": 7.086166724457791e+18 | |
| }, | |
| { | |
| "step": 5900, | |
| "train_loss": 2.790245294570923, | |
| "val_loss": 2.6511899530887604, | |
| "lr": 0.02, | |
| "tokens_per_sec": 70263.6710969525, | |
| "mfu": 21.63117474907294, | |
| "total_flops": 7.144240537967526e+18 | |
| }, | |
| { | |
| "step": 5950, | |
| "train_loss": 2.71671961247921, | |
| "val_loss": 2.6397358894348146, | |
| "lr": 0.02, | |
| "tokens_per_sec": 70225.60293793224, | |
| "mfu": 21.61945519347206, | |
| "total_flops": 7.20231435147726e+18 | |
| }, | |
| { | |
| "step": 6000, | |
| "train_loss": 2.6487374305725098, | |
| "val_loss": 2.6443570375442507, | |
| "lr": 0.02, | |
| "tokens_per_sec": 70068.37071382166, | |
| "mfu": 21.57105012634671, | |
| "total_flops": 7.260388164986995e+18 | |
| }, | |
| { | |
| "step": 6050, | |
| "train_loss": 2.54835969209671, | |
| "val_loss": 2.635010451078415, | |
| "lr": 0.02, | |
| "tokens_per_sec": 70263.49811395227, | |
| "mfu": 21.631121495016494, | |
| "total_flops": 7.318461978496729e+18 | |
| }, | |
| { | |
| "step": 6100, | |
| "train_loss": 2.5841805785894394, | |
| "val_loss": 2.635376179218292, | |
| "lr": 0.02, | |
| "tokens_per_sec": 70280.29288693903, | |
| "mfu": 21.6362918862539, | |
| "total_flops": 7.376535792006463e+18 | |
| }, | |
| { | |
| "step": 6150, | |
| "train_loss": 2.6273926347494125, | |
| "val_loss": 2.6592205107212066, | |
| "lr": 0.02, | |
| "tokens_per_sec": 70289.42059841492, | |
| "mfu": 21.639101917652937, | |
| "total_flops": 7.434609605516198e+18 | |
| }, | |
| { | |
| "step": 6200, | |
| "train_loss": 2.7345776110887527, | |
| "val_loss": 2.68223152756691, | |
| "lr": 0.02, | |
| "tokens_per_sec": 70245.30586681265, | |
| "mfu": 21.625520881344993, | |
| "total_flops": 7.492683419025932e+18 | |
| }, | |
| { | |
| "step": 6250, | |
| "train_loss": 2.643840879201889, | |
| "val_loss": 2.6727379858493805, | |
| "lr": 0.02, | |
| "tokens_per_sec": 70289.36971010863, | |
| "mfu": 21.639086251323114, | |
| "total_flops": 7.550757232535667e+18 | |
| }, | |
| { | |
| "step": 6300, | |
| "train_loss": 2.755268767476082, | |
| "val_loss": 2.6929265022277833, | |
| "lr": 0.02, | |
| "tokens_per_sec": 70267.55979098973, | |
| "mfu": 21.632371911403798, | |
| "total_flops": 7.608831046045401e+18 | |
| }, | |
| { | |
| "step": 6350, | |
| "train_loss": 2.768963858485222, | |
| "val_loss": 2.694007235765457, | |
| "lr": 0.02, | |
| "tokens_per_sec": 70111.05374796665, | |
| "mfu": 21.584190404331004, | |
| "total_flops": 7.666904859555135e+18 | |
| }, | |
| { | |
| "step": 6400, | |
| "train_loss": 2.6468516141176224, | |
| "val_loss": 2.6841990649700165, | |
| "lr": 0.02, | |
| "tokens_per_sec": 70281.87847469341, | |
| "mfu": 21.63678002080849, | |
| "total_flops": 7.72497867306487e+18 | |
| }, | |
| { | |
| "step": 6450, | |
| "train_loss": 2.712806150317192, | |
| "val_loss": 2.684369647502899, | |
| "lr": 0.02, | |
| "tokens_per_sec": 70267.07985526483, | |
| "mfu": 21.632224159751033, | |
| "total_flops": 7.783052486574604e+18 | |
| }, | |
| { | |
| "step": 6500, | |
| "train_loss": 2.5837599337100983, | |
| "val_loss": 2.6865688979625704, | |
| "lr": 0.02, | |
| "tokens_per_sec": 70286.26060807702, | |
| "mfu": 21.63812909197877, | |
| "total_flops": 7.841126300084339e+18 | |
| }, | |
| { | |
| "step": 6550, | |
| "train_loss": 2.7502972185611725, | |
| "val_loss": 2.696410584449768, | |
| "lr": 0.02, | |
| "tokens_per_sec": 70281.19023232057, | |
| "mfu": 21.63656814046116, | |
| "total_flops": 7.899200113594073e+18 | |
| }, | |
| { | |
| "step": 6600, | |
| "train_loss": 2.7288148403167725, | |
| "val_loss": 2.686836302280426, | |
| "lr": 0.02, | |
| "tokens_per_sec": 70292.68692806634, | |
| "mfu": 21.64010748064694, | |
| "total_flops": 7.957273927103807e+18 | |
| }, | |
| { | |
| "step": 6650, | |
| "train_loss": 2.747691661119461, | |
| "val_loss": 2.6874350547790526, | |
| "lr": 0.02, | |
| "tokens_per_sec": 70281.46943607075, | |
| "mfu": 21.636654095336798, | |
| "total_flops": 8.015347740613542e+18 | |
| }, | |
| { | |
| "step": 6700, | |
| "train_loss": 2.7368868738412857, | |
| "val_loss": 2.711217999458313, | |
| "lr": 0.02, | |
| "tokens_per_sec": 70153.36962880565, | |
| "mfu": 21.597217651538543, | |
| "total_flops": 8.073421554123276e+18 | |
| }, | |
| { | |
| "step": 6750, | |
| "train_loss": 2.686244487762451, | |
| "val_loss": 2.7106106102466585, | |
| "lr": 0.02, | |
| "tokens_per_sec": 70274.80138537045, | |
| "mfu": 21.63460128813669, | |
| "total_flops": 8.131495367633011e+18 | |
| }, | |
| { | |
| "step": 6800, | |
| "train_loss": 2.6627790927886963, | |
| "val_loss": 2.703348255157471, | |
| "lr": 0.02, | |
| "tokens_per_sec": 70283.19165337789, | |
| "mfu": 21.637184292278473, | |
| "total_flops": 8.189569181142745e+18 | |
| }, | |
| { | |
| "step": 6850, | |
| "train_loss": 2.628435805439949, | |
| "val_loss": 2.7075068950653076, | |
| "lr": 0.02, | |
| "tokens_per_sec": 70271.51706881971, | |
| "mfu": 21.63359018774681, | |
| "total_flops": 8.24764299465248e+18 | |
| }, | |
| { | |
| "step": 6900, | |
| "train_loss": 2.694452792406082, | |
| "val_loss": 2.7081696808338167, | |
| "lr": 0.02, | |
| "tokens_per_sec": 70299.27567071345, | |
| "mfu": 21.642135872294453, | |
| "total_flops": 8.305716808162214e+18 | |
| }, | |
| { | |
| "step": 6950, | |
| "train_loss": 2.6705697178840637, | |
| "val_loss": 2.7355406820774077, | |
| "lr": 0.02, | |
| "tokens_per_sec": 70283.07765242794, | |
| "mfu": 21.63714919626895, | |
| "total_flops": 8.363790621671948e+18 | |
| }, | |
| { | |
| "step": 7000, | |
| "train_loss": 2.6670849472284317, | |
| "val_loss": 2.714768886566162, | |
| "lr": 0.02, | |
| "tokens_per_sec": 70282.50900099597, | |
| "mfu": 21.636974132849918, | |
| "total_flops": 8.421864435181683e+18 | |
| }, | |
| { | |
| "step": 7050, | |
| "train_loss": 2.727104902267456, | |
| "val_loss": 2.735373377799988, | |
| "lr": 0.02, | |
| "tokens_per_sec": 70280.4880772634, | |
| "mfu": 21.63635197699425, | |
| "total_flops": 8.479938248691417e+18 | |
| }, | |
| { | |
| "step": 7100, | |
| "train_loss": 2.7122743129730225, | |
| "val_loss": 2.729275506734848, | |
| "lr": 0.02, | |
| "tokens_per_sec": 70293.89287865201, | |
| "mfu": 21.640478741178168, | |
| "total_flops": 8.538012062201151e+18 | |
| }, | |
| { | |
| "step": 7150, | |
| "train_loss": 2.6882054656744003, | |
| "val_loss": 2.7395458936691286, | |
| "lr": 0.02, | |
| "tokens_per_sec": 70289.38835791209, | |
| "mfu": 21.63909199218306, | |
| "total_flops": 8.596085875710886e+18 | |
| }, | |
| { | |
| "step": 7200, | |
| "train_loss": 2.770738333463669, | |
| "val_loss": 2.7353258550167086, | |
| "lr": 0.02, | |
| "tokens_per_sec": 70304.98790253063, | |
| "mfu": 21.64389442380073, | |
| "total_flops": 8.65415968922062e+18 | |
| }, | |
| { | |
| "step": 7250, | |
| "train_loss": 2.681736469268799, | |
| "val_loss": 2.7411834478378294, | |
| "lr": 0.02, | |
| "tokens_per_sec": 70285.81164313164, | |
| "mfu": 21.63799087490246, | |
| "total_flops": 8.712233502730355e+18 | |
| }, | |
| { | |
| "step": 7300, | |
| "train_loss": 2.6936238557100296, | |
| "val_loss": 2.7522403955459596, | |
| "lr": 0.02, | |
| "tokens_per_sec": 70295.82648556051, | |
| "mfu": 21.64107401592372, | |
| "total_flops": 8.770307316240089e+18 | |
| }, | |
| { | |
| "step": 7350, | |
| "train_loss": 2.725588798522949, | |
| "val_loss": 2.7442231357097624, | |
| "lr": 0.02, | |
| "tokens_per_sec": 70298.27954427102, | |
| "mfu": 21.641829207629595, | |
| "total_flops": 8.828381129749823e+18 | |
| }, | |
| { | |
| "step": 7400, | |
| "train_loss": 2.646110162138939, | |
| "val_loss": 2.7503407299518585, | |
| "lr": 0.02, | |
| "tokens_per_sec": 70284.60641909202, | |
| "mfu": 21.63761983804364, | |
| "total_flops": 8.886454943259558e+18 | |
| }, | |
| { | |
| "step": 7450, | |
| "train_loss": 2.7466748505830765, | |
| "val_loss": 2.7323903262615206, | |
| "lr": 0.02, | |
| "tokens_per_sec": 70288.42486357277, | |
| "mfu": 21.63879537354504, | |
| "total_flops": 8.944528756769292e+18 | |
| }, | |
| { | |
| "step": 7500, | |
| "train_loss": 2.725738510489464, | |
| "val_loss": 2.746686214208603, | |
| "lr": 0.02, | |
| "tokens_per_sec": 70294.84079339539, | |
| "mfu": 21.64077056352592, | |
| "total_flops": 9.002602570279027e+18 | |
| }, | |
| { | |
| "step": 7550, | |
| "train_loss": 2.6565753519535065, | |
| "val_loss": 2.7430707275867463, | |
| "lr": 0.02, | |
| "tokens_per_sec": 70298.68934041717, | |
| "mfu": 21.641955366310324, | |
| "total_flops": 9.060676383788761e+18 | |
| }, | |
| { | |
| "step": 7600, | |
| "train_loss": 2.7076889276504517, | |
| "val_loss": 2.7611879467964173, | |
| "lr": 0.02, | |
| "tokens_per_sec": 70256.34739863472, | |
| "mfu": 21.628920096054557, | |
| "total_flops": 9.118750197298495e+18 | |
| }, | |
| { | |
| "step": 7650, | |
| "train_loss": 2.7396631836891174, | |
| "val_loss": 2.7371096432209017, | |
| "lr": 0.02, | |
| "tokens_per_sec": 70277.87960201211, | |
| "mfu": 21.635548939192372, | |
| "total_flops": 9.17682401080823e+18 | |
| }, | |
| { | |
| "step": 7700, | |
| "train_loss": 2.6412112563848495, | |
| "val_loss": 2.7318264484405517, | |
| "lr": 0.02, | |
| "tokens_per_sec": 70300.82907752205, | |
| "mfu": 21.642614099714233, | |
| "total_flops": 9.234897824317964e+18 | |
| }, | |
| { | |
| "step": 7750, | |
| "train_loss": 2.54968099296093, | |
| "val_loss": 2.738618016242981, | |
| "lr": 0.02, | |
| "tokens_per_sec": 70302.56022262946, | |
| "mfu": 21.64314704514321, | |
| "total_flops": 9.292971637827699e+18 | |
| }, | |
| { | |
| "step": 7800, | |
| "train_loss": 2.649760976433754, | |
| "val_loss": 2.7415043830871584, | |
| "lr": 0.02, | |
| "tokens_per_sec": 70278.13295039837, | |
| "mfu": 21.635626934308913, | |
| "total_flops": 9.351045451337433e+18 | |
| }, | |
| { | |
| "step": 7550, | |
| "train_loss": 2.5435707718133926, | |
| "val_loss": 2.573974919319153, | |
| "lr": 0.02, | |
| "tokens_per_sec": 69895.20805720806, | |
| "mfu": 21.52970835915071, | |
| "total_flops": 9.409119264847167e+18 | |
| }, | |
| { | |
| "step": 7600, | |
| "train_loss": 2.6072546392679214, | |
| "val_loss": 2.596907156705856, | |
| "lr": 0.02, | |
| "tokens_per_sec": 70064.23548152526, | |
| "mfu": 21.5817736044143, | |
| "total_flops": 9.467193078356902e+18 | |
| }, | |
| { | |
| "step": 7650, | |
| "train_loss": 2.609854079782963, | |
| "val_loss": 2.6193448185920714, | |
| "lr": 0.02, | |
| "tokens_per_sec": 70083.55278385278, | |
| "mfu": 21.587723881936274, | |
| "total_flops": 9.525266891866636e+18 | |
| }, | |
| { | |
| "step": 7700, | |
| "train_loss": 2.6742303520441055, | |
| "val_loss": 2.633003169298172, | |
| "lr": 0.02, | |
| "tokens_per_sec": 70080.20580557312, | |
| "mfu": 21.586692917607696, | |
| "total_flops": 9.58334070537637e+18 | |
| }, | |
| { | |
| "step": 7750, | |
| "train_loss": 2.5816686004400253, | |
| "val_loss": 2.6430088937282563, | |
| "lr": 0.02, | |
| "tokens_per_sec": 70082.89633950195, | |
| "mfu": 21.58752167844021, | |
| "total_flops": 9.641414518886105e+18 | |
| }, | |
| { | |
| "step": 7800, | |
| "train_loss": 2.672006607055664, | |
| "val_loss": 2.6543741524219513, | |
| "lr": 0.02, | |
| "tokens_per_sec": 70083.62760901512, | |
| "mfu": 21.587746930210518, | |
| "total_flops": 9.69948833239584e+18 | |
| }, | |
| { | |
| "step": 7850, | |
| "train_loss": 2.549602121114731, | |
| "val_loss": 2.6550877690315247, | |
| "lr": 0.02, | |
| "tokens_per_sec": 69909.72719958337, | |
| "mfu": 21.53418066719091, | |
| "total_flops": 9.757562145905574e+18 | |
| }, | |
| { | |
| "step": 7900, | |
| "train_loss": 2.7753456830978394, | |
| "val_loss": 2.6652530431747437, | |
| "lr": 0.02, | |
| "tokens_per_sec": 70079.33580661588, | |
| "mfu": 21.58642493323019, | |
| "total_flops": 9.815635959415308e+18 | |
| }, | |
| { | |
| "step": 7950, | |
| "train_loss": 2.656064510345459, | |
| "val_loss": 2.684378969669342, | |
| "lr": 0.02, | |
| "tokens_per_sec": 70077.42447544802, | |
| "mfu": 21.585836188969093, | |
| "total_flops": 9.873709772925043e+18 | |
| }, | |
| { | |
| "step": 8000, | |
| "train_loss": 2.650539129972458, | |
| "val_loss": 2.679304671287537, | |
| "lr": 0.02, | |
| "tokens_per_sec": 70074.83105163992, | |
| "mfu": 21.585037340810658, | |
| "total_flops": 9.931783586434777e+18 | |
| }, | |
| { | |
| "step": 8050, | |
| "train_loss": 2.6973265260457993, | |
| "val_loss": 2.67431520819664, | |
| "lr": 0.02, | |
| "tokens_per_sec": 70097.39491881663, | |
| "mfu": 21.591987652473964, | |
| "total_flops": 9.989857399944511e+18 | |
| }, | |
| { | |
| "step": 8100, | |
| "train_loss": 2.7974444180727005, | |
| "val_loss": 2.6660505414009092, | |
| "lr": 0.02, | |
| "tokens_per_sec": 70063.2442190119, | |
| "mfu": 21.581468267418913, | |
| "total_flops": 1.0047931213454246e+19 | |
| }, | |
| { | |
| "step": 8150, | |
| "train_loss": 2.6419722586870193, | |
| "val_loss": 2.6528017103672026, | |
| "lr": 0.02, | |
| "tokens_per_sec": 70082.9707165531, | |
| "mfu": 21.58754458868348, | |
| "total_flops": 1.010600502696398e+19 | |
| }, | |
| { | |
| "step": 8200, | |
| "train_loss": 2.627345636487007, | |
| "val_loss": 2.649439036846161, | |
| "lr": 0.02, | |
| "tokens_per_sec": 70074.56811350217, | |
| "mfu": 21.584956348399597, | |
| "total_flops": 1.0164078840473715e+19 | |
| }, | |
| { | |
| "step": 8250, | |
| "train_loss": 2.7267237305641174, | |
| "val_loss": 2.6591224551200865, | |
| "lr": 0.02, | |
| "tokens_per_sec": 70057.47656743563, | |
| "mfu": 21.579691666993703, | |
| "total_flops": 1.022215265398345e+19 | |
| }, | |
| { | |
| "step": 8300, | |
| "train_loss": 2.440946727991104, | |
| "val_loss": 2.672474372386932, | |
| "lr": 0.02, | |
| "tokens_per_sec": 70054.52809839167, | |
| "mfu": 21.57878345482313, | |
| "total_flops": 1.0280226467493183e+19 | |
| }, | |
| { | |
| "step": 8350, | |
| "train_loss": 2.662572145462036, | |
| "val_loss": 2.6874975621700288, | |
| "lr": 0.02, | |
| "tokens_per_sec": 70080.70251085235, | |
| "mfu": 21.586845916934827, | |
| "total_flops": 1.0338300281002918e+19 | |
| }, | |
| { | |
| "step": 8400, | |
| "train_loss": 2.7091101929545403, | |
| "val_loss": 2.686986118555069, | |
| "lr": 0.02, | |
| "tokens_per_sec": 70085.49862505833, | |
| "mfu": 21.588323256268698, | |
| "total_flops": 1.0396374094512652e+19 | |
| }, | |
| { | |
| "step": 8450, | |
| "train_loss": 2.6898171603679657, | |
| "val_loss": 2.699655681848526, | |
| "lr": 0.02, | |
| "tokens_per_sec": 69901.0587853534, | |
| "mfu": 21.531510549517677, | |
| "total_flops": 1.0454447908022387e+19 | |
| }, | |
| { | |
| "step": 8500, | |
| "train_loss": 2.6142636835575104, | |
| "val_loss": 2.7020849883556366, | |
| "lr": 0.02, | |
| "tokens_per_sec": 70093.13326739804, | |
| "mfu": 21.590674942851646, | |
| "total_flops": 1.0512521721532121e+19 | |
| }, | |
| { | |
| "step": 8550, | |
| "train_loss": 2.5702943354845047, | |
| "val_loss": 2.6926130056381226, | |
| "lr": 0.02, | |
| "tokens_per_sec": 70079.20247791831, | |
| "mfu": 21.586383864206226, | |
| "total_flops": 1.0570595535041855e+19 | |
| }, | |
| { | |
| "step": 8600, | |
| "train_loss": 2.730316862463951, | |
| "val_loss": 2.6773874521255494, | |
| "lr": 0.02, | |
| "tokens_per_sec": 70071.69010267203, | |
| "mfu": 21.584069839359135, | |
| "total_flops": 1.062866934855159e+19 | |
| }, | |
| { | |
| "step": 8650, | |
| "train_loss": 2.7265526801347733, | |
| "val_loss": 2.6893074452877044, | |
| "lr": 0.02, | |
| "tokens_per_sec": 70073.10473971444, | |
| "mfu": 21.584505587728785, | |
| "total_flops": 1.0686743162061324e+19 | |
| }, | |
| { | |
| "step": 8700, | |
| "train_loss": 2.5364869236946106, | |
| "val_loss": 2.690558123588562, | |
| "lr": 0.02, | |
| "tokens_per_sec": 70063.07769055446, | |
| "mfu": 21.58141697192639, | |
| "total_flops": 1.0744816975571059e+19 | |
| }, | |
| { | |
| "step": 8750, | |
| "train_loss": 2.694303020834923, | |
| "val_loss": 2.6988609194755555, | |
| "lr": 0.02, | |
| "tokens_per_sec": 70084.14993777053, | |
| "mfu": 21.58790782229574, | |
| "total_flops": 1.0802890789080793e+19 | |
| }, | |
| { | |
| "step": 8800, | |
| "train_loss": 2.6549243479967117, | |
| "val_loss": 2.688143104314804, | |
| "lr": 0.02, | |
| "tokens_per_sec": 69908.60717930882, | |
| "mfu": 21.53383566914973, | |
| "total_flops": 1.0860964602590527e+19 | |
| }, | |
| { | |
| "step": 8850, | |
| "train_loss": 2.6796169131994247, | |
| "val_loss": 2.703415608406067, | |
| "lr": 0.02, | |
| "tokens_per_sec": 70085.21784901444, | |
| "mfu": 21.58823676927614, | |
| "total_flops": 1.0919038416100262e+19 | |
| }, | |
| { | |
| "step": 8900, | |
| "train_loss": 2.7315694838762283, | |
| "val_loss": 2.706169080734253, | |
| "lr": 0.02, | |
| "tokens_per_sec": 70079.59185720404, | |
| "mfu": 21.586503804080444, | |
| "total_flops": 1.0977112229609996e+19 | |
| }, | |
| { | |
| "step": 8950, | |
| "train_loss": 2.6746401339769363, | |
| "val_loss": 2.696428966522217, | |
| "lr": 0.02, | |
| "tokens_per_sec": 70065.1631482729, | |
| "mfu": 21.5820593521084, | |
| "total_flops": 1.103518604311973e+19 | |
| }, | |
| { | |
| "step": 9000, | |
| "train_loss": 2.6321981251239777, | |
| "val_loss": 2.696945106983185, | |
| "lr": 0.02, | |
| "tokens_per_sec": 70080.09558887687, | |
| "mfu": 21.586658967736277, | |
| "total_flops": 1.1093259856629465e+19 | |
| }, | |
| { | |
| "step": 9050, | |
| "train_loss": 2.574547991156578, | |
| "val_loss": 2.6979831516742707, | |
| "lr": 0.02, | |
| "tokens_per_sec": 70080.95175949985, | |
| "mfu": 21.586922692594257, | |
| "total_flops": 1.11513336701392e+19 | |
| }, | |
| { | |
| "step": 9100, | |
| "train_loss": 2.6053851395845413, | |
| "val_loss": 2.729407620429993, | |
| "lr": 0.02, | |
| "tokens_per_sec": 70049.16029463499, | |
| "mfu": 21.577130018877764, | |
| "total_flops": 1.1209407483648934e+19 | |
| }, | |
| { | |
| "step": 9150, | |
| "train_loss": 2.6026515066623688, | |
| "val_loss": 2.7150147020816804, | |
| "lr": 0.02, | |
| "tokens_per_sec": 70077.26122935086, | |
| "mfu": 21.58578590453672, | |
| "total_flops": 1.1267481297158668e+19 | |
| }, | |
| { | |
| "step": 9200, | |
| "train_loss": 2.638631984591484, | |
| "val_loss": 2.724787563085556, | |
| "lr": 0.02, | |
| "tokens_per_sec": 70069.194224259, | |
| "mfu": 21.583301037951628, | |
| "total_flops": 1.1325555110668403e+19 | |
| }, | |
| { | |
| "step": 9250, | |
| "train_loss": 2.687884122133255, | |
| "val_loss": 2.7205813467502593, | |
| "lr": 0.02, | |
| "tokens_per_sec": 70071.53235369503, | |
| "mfu": 21.584021248195835, | |
| "total_flops": 1.1383628924178137e+19 | |
| }, | |
| { | |
| "step": 9300, | |
| "train_loss": 2.6440293043851852, | |
| "val_loss": 2.7422130703926086, | |
| "lr": 0.02, | |
| "tokens_per_sec": 70093.24296655692, | |
| "mfu": 21.590708733306865, | |
| "total_flops": 1.1441702737687871e+19 | |
| }, | |
| { | |
| "step": 9350, | |
| "train_loss": 2.6162674874067307, | |
| "val_loss": 2.7284474194049837, | |
| "lr": 0.02, | |
| "tokens_per_sec": 70091.7371465895, | |
| "mfu": 21.59024489800769, | |
| "total_flops": 1.1499776551197606e+19 | |
| }, | |
| { | |
| "step": 9400, | |
| "train_loss": 2.61313958466053, | |
| "val_loss": 2.727823382616043, | |
| "lr": 0.02, | |
| "tokens_per_sec": 69939.90379685092, | |
| "mfu": 21.543475915842432, | |
| "total_flops": 1.155785036470734e+19 | |
| }, | |
| { | |
| "step": 9450, | |
| "train_loss": 2.645725265145302, | |
| "val_loss": 2.740615874528885, | |
| "lr": 0.02, | |
| "tokens_per_sec": 70059.97998144061, | |
| "mfu": 21.580462789577464, | |
| "total_flops": 1.1615924178217075e+19 | |
| }, | |
| { | |
| "step": 9500, | |
| "train_loss": 2.6479314118623734, | |
| "val_loss": 2.734837234020233, | |
| "lr": 0.02, | |
| "tokens_per_sec": 70060.40698137914, | |
| "mfu": 21.58059431768077, | |
| "total_flops": 1.167399799172681e+19 | |
| }, | |
| { | |
| "step": 9550, | |
| "train_loss": 2.748686507344246, | |
| "val_loss": 2.728809928894043, | |
| "lr": 0.02, | |
| "tokens_per_sec": 70073.24083636962, | |
| "mfu": 21.58454750936227, | |
| "total_flops": 1.1732071805236543e+19 | |
| }, | |
| { | |
| "step": 9600, | |
| "train_loss": 2.582173064351082, | |
| "val_loss": 2.719840705394745, | |
| "lr": 0.02, | |
| "tokens_per_sec": 70062.49574008441, | |
| "mfu": 21.581237714660436, | |
| "total_flops": 1.1790145618746278e+19 | |
| }, | |
| { | |
| "step": 9650, | |
| "train_loss": 2.7906376272439957, | |
| "val_loss": 2.7412230372428894, | |
| "lr": 0.02, | |
| "tokens_per_sec": 70085.55469118859, | |
| "mfu": 21.588340526228496, | |
| "total_flops": 1.1848219432256012e+19 | |
| }, | |
| { | |
| "step": 9700, | |
| "train_loss": 2.62669138610363, | |
| "val_loss": 2.728373169898987, | |
| "lr": 0.02, | |
| "tokens_per_sec": 70065.22197239265, | |
| "mfu": 21.58207747160726, | |
| "total_flops": 1.1906293245765747e+19 | |
| }, | |
| { | |
| "step": 9750, | |
| "train_loss": 2.624654397368431, | |
| "val_loss": 2.7269852578639986, | |
| "lr": 0.02, | |
| "tokens_per_sec": 69933.18341549518, | |
| "mfu": 21.5414058475978, | |
| "total_flops": 1.1964367059275481e+19 | |
| }, | |
| { | |
| "step": 9800, | |
| "train_loss": 2.648192882537842, | |
| "val_loss": 2.7176962196826935, | |
| "lr": 0.02, | |
| "tokens_per_sec": 70090.91467002199, | |
| "mfu": 21.589991551875347, | |
| "total_flops": 1.2022440872785215e+19 | |
| }, | |
| { | |
| "step": 9850, | |
| "train_loss": 2.6563764810562134, | |
| "val_loss": 2.724771249294281, | |
| "lr": 0.02, | |
| "tokens_per_sec": 70081.3885072608, | |
| "mfu": 21.587057223303603, | |
| "total_flops": 1.208051468629495e+19 | |
| }, | |
| { | |
| "step": 9900, | |
| "train_loss": 2.681455746293068, | |
| "val_loss": 2.7229500114917755, | |
| "lr": 0.02, | |
| "tokens_per_sec": 70072.09994073665, | |
| "mfu": 21.584196081118066, | |
| "total_flops": 1.2138588499804684e+19 | |
| }, | |
| { | |
| "step": 9950, | |
| "train_loss": 2.62647445499897, | |
| "val_loss": 2.7153716802597048, | |
| "lr": 0.02, | |
| "tokens_per_sec": 70062.32050948143, | |
| "mfu": 21.58118373866004, | |
| "total_flops": 1.2196662313314419e+19 | |
| }, | |
| { | |
| "step": 10000, | |
| "train_loss": 2.5326335802674294, | |
| "val_loss": 2.717075729370117, | |
| "lr": 0.02, | |
| "tokens_per_sec": 70096.4264025015, | |
| "mfu": 21.591689321953382, | |
| "total_flops": 1.2254736126824153e+19 | |
| }, | |
| { | |
| "step": 10050, | |
| "train_loss": 2.5265747159719467, | |
| "val_loss": 2.550719881057739, | |
| "lr": 0.02, | |
| "tokens_per_sec": 70016.48820081342, | |
| "mfu": 21.406339952207, | |
| "total_flops": 1.2312809940333887e+19 | |
| }, | |
| { | |
| "step": 10100, | |
| "train_loss": 2.5934030413627625, | |
| "val_loss": 2.584617280960083, | |
| "lr": 0.02, | |
| "tokens_per_sec": 70021.87329482104, | |
| "mfu": 21.407986352304516, | |
| "total_flops": 1.2370883753843622e+19 | |
| }, | |
| { | |
| "step": 10150, | |
| "train_loss": 2.5912967920303345, | |
| "val_loss": 2.610349267721176, | |
| "lr": 0.02, | |
| "tokens_per_sec": 70013.94999834869, | |
| "mfu": 21.405563940352884, | |
| "total_flops": 1.2428957567353356e+19 | |
| }, | |
| { | |
| "step": 10200, | |
| "train_loss": 2.6515150517225266, | |
| "val_loss": 2.622386109828949, | |
| "lr": 0.02, | |
| "tokens_per_sec": 70030.94537464414, | |
| "mfu": 21.410759985055275, | |
| "total_flops": 1.248703138086309e+19 | |
| }, | |
| { | |
| "step": 10050, | |
| "train_loss": 2.526685908436775, | |
| "val_loss": 2.552409863471985, | |
| "lr": 0.02, | |
| "tokens_per_sec": 70125.69793352022, | |
| "mfu": 21.60070578450605, | |
| "total_flops": 1.2545105194372825e+19 | |
| }, | |
| { | |
| "step": 10100, | |
| "train_loss": 2.593015134334564, | |
| "val_loss": 2.583539479970932, | |
| "lr": 0.02, | |
| "tokens_per_sec": 70113.93309309517, | |
| "mfu": 21.5970818796593, | |
| "total_flops": 1.260317900788256e+19 | |
| }, | |
| { | |
| "step": 10150, | |
| "train_loss": 2.5912905409932137, | |
| "val_loss": 2.61044899225235, | |
| "lr": 0.02, | |
| "tokens_per_sec": 70132.29410189865, | |
| "mfu": 21.60273759162735, | |
| "total_flops": 1.2661252821392294e+19 | |
| }, | |
| { | |
| "step": 10200, | |
| "train_loss": 2.6520476937294006, | |
| "val_loss": 2.6223703742027284, | |
| "lr": 0.02, | |
| "tokens_per_sec": 70178.88626107713, | |
| "mfu": 21.617089299374108, | |
| "total_flops": 1.2719326634902028e+19 | |
| }, | |
| { | |
| "step": 10050, | |
| "train_loss": 2.52651284635067, | |
| "val_loss": 2.5529006838798525, | |
| "lr": 0.02, | |
| "tokens_per_sec": 70204.14387890541, | |
| "mfu": 21.487435818820614, | |
| "total_flops": 1.2777400448411763e+19 | |
| }, | |
| { | |
| "step": 10100, | |
| "train_loss": 2.5928536653518677, | |
| "val_loss": 2.582450878620148, | |
| "lr": 0.02, | |
| "tokens_per_sec": 70192.73068313688, | |
| "mfu": 21.483942573293874, | |
| "total_flops": 1.2835474261921497e+19 | |
| }, | |
| { | |
| "step": 10150, | |
| "train_loss": 2.5910468697547913, | |
| "val_loss": 2.6105341851711272, | |
| "lr": 0.02, | |
| "tokens_per_sec": 70218.74469285099, | |
| "mfu": 21.49190470107206, | |
| "total_flops": 1.2893548075431231e+19 | |
| }, | |
| { | |
| "step": 10200, | |
| "train_loss": 2.6514677554368973, | |
| "val_loss": 2.624534994363785, | |
| "lr": 0.02, | |
| "tokens_per_sec": 70219.55436424627, | |
| "mfu": 21.492152517813057, | |
| "total_flops": 1.2951621888940966e+19 | |
| }, | |
| { | |
| "step": 10250, | |
| "train_loss": 2.564076505601406, | |
| "val_loss": 2.6290538609027863, | |
| "lr": 0.02, | |
| "tokens_per_sec": 70220.83740988419, | |
| "mfu": 21.492545220569248, | |
| "total_flops": 1.30096957024507e+19 | |
| }, | |
| { | |
| "step": 10300, | |
| "train_loss": 2.652680352330208, | |
| "val_loss": 2.6442084670066834, | |
| "lr": 0.02, | |
| "tokens_per_sec": 70235.22701111587, | |
| "mfu": 21.496949456784385, | |
| "total_flops": 1.3067769515960435e+19 | |
| }, | |
| { | |
| "step": 10350, | |
| "train_loss": 2.5286182314157486, | |
| "val_loss": 2.6446537137031556, | |
| "lr": 0.02, | |
| "tokens_per_sec": 70221.15829073807, | |
| "mfu": 21.492643432816735, | |
| "total_flops": 1.312584332947017e+19 | |
| }, | |
| { | |
| "step": 10400, | |
| "train_loss": 2.7530941367149353, | |
| "val_loss": 2.656357008218765, | |
| "lr": 0.02, | |
| "tokens_per_sec": 70236.16616565398, | |
| "mfu": 21.497236904529586, | |
| "total_flops": 1.3183917142979903e+19 | |
| }, | |
| { | |
| "step": 10450, | |
| "train_loss": 2.6332227289676666, | |
| "val_loss": 2.673243999481201, | |
| "lr": 0.02, | |
| "tokens_per_sec": 70229.65653406357, | |
| "mfu": 21.495244496627787, | |
| "total_flops": 1.3241990956489638e+19 | |
| }, | |
| { | |
| "step": 10500, | |
| "train_loss": 2.6284239292144775, | |
| "val_loss": 2.662519484758377, | |
| "lr": 0.02, | |
| "tokens_per_sec": 70214.50167879055, | |
| "mfu": 21.49060603852503, | |
| "total_flops": 1.3300064769999372e+19 | |
| }, | |
| { | |
| "step": 10550, | |
| "train_loss": 2.6753184646368027, | |
| "val_loss": 2.6621902704238893, | |
| "lr": 0.02, | |
| "tokens_per_sec": 70230.3363657566, | |
| "mfu": 21.495452573232164, | |
| "total_flops": 1.3358138583509107e+19 | |
| }, | |
| { | |
| "step": 10600, | |
| "train_loss": 2.775514140725136, | |
| "val_loss": 2.6512930035591125, | |
| "lr": 0.02, | |
| "tokens_per_sec": 70222.59029373217, | |
| "mfu": 21.49308172706445, | |
| "total_flops": 1.3416212397018841e+19 | |
| }, | |
| { | |
| "step": 10650, | |
| "train_loss": 2.621285066008568, | |
| "val_loss": 2.647852545976639, | |
| "lr": 0.02, | |
| "tokens_per_sec": 70225.17930426326, | |
| "mfu": 21.49387414749076, | |
| "total_flops": 1.3474286210528575e+19 | |
| }, | |
| { | |
| "step": 10700, | |
| "train_loss": 2.6071013510227203, | |
| "val_loss": 2.6399790406227113, | |
| "lr": 0.02, | |
| "tokens_per_sec": 70208.71694363162, | |
| "mfu": 21.488835500226454, | |
| "total_flops": 1.353236002403831e+19 | |
| }, | |
| { | |
| "step": 10750, | |
| "train_loss": 2.705596625804901, | |
| "val_loss": 2.649714803695679, | |
| "lr": 0.02, | |
| "tokens_per_sec": 70220.28075246631, | |
| "mfu": 21.492374844009134, | |
| "total_flops": 1.3590433837548044e+19 | |
| }, | |
| { | |
| "step": 10800, | |
| "train_loss": 2.4201852083206177, | |
| "val_loss": 2.6591944634914397, | |
| "lr": 0.02, | |
| "tokens_per_sec": 70213.37512246182, | |
| "mfu": 21.49026123257096, | |
| "total_flops": 1.3648507651057779e+19 | |
| }, | |
| { | |
| "step": 10850, | |
| "train_loss": 2.64433017373085, | |
| "val_loss": 2.6790621519088744, | |
| "lr": 0.02, | |
| "tokens_per_sec": 70210.31600380773, | |
| "mfu": 21.48932492579318, | |
| "total_flops": 1.3706581464567513e+19 | |
| }, | |
| { | |
| "step": 10900, | |
| "train_loss": 2.6901479735970497, | |
| "val_loss": 2.6743451356887817, | |
| "lr": 0.02, | |
| "tokens_per_sec": 70049.41735203927, | |
| "mfu": 21.44007855282732, | |
| "total_flops": 1.3764655278077247e+19 | |
| }, | |
| { | |
| "step": 10950, | |
| "train_loss": 2.6701857894659042, | |
| "val_loss": 2.683942812681198, | |
| "lr": 0.02, | |
| "tokens_per_sec": 70224.3843041066, | |
| "mfu": 21.49363082118695, | |
| "total_flops": 1.3822729091586982e+19 | |
| }, | |
| { | |
| "step": 11000, | |
| "train_loss": 2.5939954668283463, | |
| "val_loss": 2.6858055651187898, | |
| "lr": 0.02, | |
| "tokens_per_sec": 70231.74048079467, | |
| "mfu": 21.495882331791933, | |
| "total_flops": 1.3880802905096716e+19 | |
| }, | |
| { | |
| "step": 11050, | |
| "train_loss": 2.549482077360153, | |
| "val_loss": 2.6854387283325196, | |
| "lr": 0.02, | |
| "tokens_per_sec": 70214.96890201059, | |
| "mfu": 21.490749041892048, | |
| "total_flops": 1.393887671860645e+19 | |
| }, | |
| { | |
| "step": 11100, | |
| "train_loss": 2.711370125412941, | |
| "val_loss": 2.6654779374599458, | |
| "lr": 0.02, | |
| "tokens_per_sec": 70197.87995457259, | |
| "mfu": 21.485518614726402, | |
| "total_flops": 1.3996950532116185e+19 | |
| }, | |
| { | |
| "step": 11150, | |
| "train_loss": 2.707639902830124, | |
| "val_loss": 2.6756514489650725, | |
| "lr": 0.02, | |
| "tokens_per_sec": 70224.52323160897, | |
| "mfu": 21.493673342833475, | |
| "total_flops": 1.405502434562592e+19 | |
| }, | |
| { | |
| "step": 11200, | |
| "train_loss": 2.520912617444992, | |
| "val_loss": 2.6752108156681063, | |
| "lr": 0.02, | |
| "tokens_per_sec": 70210.97573402476, | |
| "mfu": 21.489526849923475, | |
| "total_flops": 1.4113098159135654e+19 | |
| }, | |
| { | |
| "step": 11250, | |
| "train_loss": 2.674424484372139, | |
| "val_loss": 2.6870215952396395, | |
| "lr": 0.02, | |
| "tokens_per_sec": 70065.63273939754, | |
| "mfu": 21.445041608793634, | |
| "total_flops": 1.4171171972645388e+19 | |
| }, | |
| { | |
| "step": 11300, | |
| "train_loss": 2.635969787836075, | |
| "val_loss": 2.6746863961219787, | |
| "lr": 0.02, | |
| "tokens_per_sec": 70204.44947845214, | |
| "mfu": 21.48752935390678, | |
| "total_flops": 1.4229245786155123e+19 | |
| }, | |
| { | |
| "step": 11350, | |
| "train_loss": 2.6615608781576157, | |
| "val_loss": 2.6862066984176636, | |
| "lr": 0.02, | |
| "tokens_per_sec": 70219.71524689811, | |
| "mfu": 21.49220175928891, | |
| "total_flops": 1.4287319599664857e+19 | |
| }, | |
| { | |
| "step": 11400, | |
| "train_loss": 2.712399512529373, | |
| "val_loss": 2.690833550691605, | |
| "lr": 0.02, | |
| "tokens_per_sec": 70220.12457604909, | |
| "mfu": 21.492327042974093, | |
| "total_flops": 1.4345393413174591e+19 | |
| }, | |
| { | |
| "step": 11450, | |
| "train_loss": 2.6577775329351425, | |
| "val_loss": 2.6866107404232027, | |
| "lr": 0.02, | |
| "tokens_per_sec": 70229.6213204413, | |
| "mfu": 21.49523371876759, | |
| "total_flops": 1.4403467226684326e+19 | |
| }, | |
| { | |
| "step": 11500, | |
| "train_loss": 2.614283174276352, | |
| "val_loss": 2.688479536771774, | |
| "lr": 0.02, | |
| "tokens_per_sec": 70236.05882282139, | |
| "mfu": 21.497204050027012, | |
| "total_flops": 1.446154104019406e+19 | |
| }, | |
| { | |
| "step": 11550, | |
| "train_loss": 2.5547793954610825, | |
| "val_loss": 2.6836730897426606, | |
| "lr": 0.02, | |
| "tokens_per_sec": 70199.13173122169, | |
| "mfu": 21.48590174695936, | |
| "total_flops": 1.4519614853703795e+19 | |
| }, | |
| { | |
| "step": 11600, | |
| "train_loss": 2.5856354534626007, | |
| "val_loss": 2.7118352591991424, | |
| "lr": 0.02, | |
| "tokens_per_sec": 70029.10581562246, | |
| "mfu": 21.433861785397028, | |
| "total_flops": 1.457768866721353e+19 | |
| }, | |
| { | |
| "step": 11650, | |
| "train_loss": 2.584961473941803, | |
| "val_loss": 2.697824990749359, | |
| "lr": 0.02, | |
| "tokens_per_sec": 70228.01487665919, | |
| "mfu": 21.49474203329494, | |
| "total_flops": 1.4635762480723263e+19 | |
| }, | |
| { | |
| "step": 11700, | |
| "train_loss": 2.6195385307073593, | |
| "val_loss": 2.7136584401130674, | |
| "lr": 0.02, | |
| "tokens_per_sec": 70223.3484761831, | |
| "mfu": 21.493313784545016, | |
| "total_flops": 1.4693836294232998e+19 | |
| }, | |
| { | |
| "step": 11750, | |
| "train_loss": 2.670626640319824, | |
| "val_loss": 2.7112324714660643, | |
| "lr": 0.02, | |
| "tokens_per_sec": 70213.6927964773, | |
| "mfu": 21.49035846329897, | |
| "total_flops": 1.4751910107742732e+19 | |
| }, | |
| { | |
| "step": 11800, | |
| "train_loss": 2.627764016389847, | |
| "val_loss": 2.73016876578331, | |
| "lr": 0.02, | |
| "tokens_per_sec": 70190.65073107202, | |
| "mfu": 21.483305960780275, | |
| "total_flops": 1.4809983921252467e+19 | |
| }, | |
| { | |
| "step": 11850, | |
| "train_loss": 2.5967312157154083, | |
| "val_loss": 2.7091954708099366, | |
| "lr": 0.02, | |
| "tokens_per_sec": 70217.17753509365, | |
| "mfu": 21.49142503990296, | |
| "total_flops": 1.4868057734762201e+19 | |
| }, | |
| { | |
| "step": 11900, | |
| "train_loss": 2.596388280391693, | |
| "val_loss": 2.7094963908195497, | |
| "lr": 0.02, | |
| "tokens_per_sec": 70215.31170073652, | |
| "mfu": 21.49085396255934, | |
| "total_flops": 1.4926131548271935e+19 | |
| }, | |
| { | |
| "step": 11950, | |
| "train_loss": 2.624119371175766, | |
| "val_loss": 2.728128719329834, | |
| "lr": 0.02, | |
| "tokens_per_sec": 70048.5275888958, | |
| "mfu": 21.43980622234392, | |
| "total_flops": 1.498420536178167e+19 | |
| }, | |
| { | |
| "step": 12000, | |
| "train_loss": 2.626932516694069, | |
| "val_loss": 2.712033325433731, | |
| "lr": 0.02, | |
| "tokens_per_sec": 70215.11474239366, | |
| "mfu": 21.49079367936924, | |
| "total_flops": 1.5042279175291404e+19 | |
| }, | |
| { | |
| "step": 12050, | |
| "train_loss": 2.731254890561104, | |
| "val_loss": 2.712426573038101, | |
| "lr": 0.02, | |
| "tokens_per_sec": 70177.50172730487, | |
| "mfu": 21.47928143517655, | |
| "total_flops": 1.5100352988801139e+19 | |
| }, | |
| { | |
| "step": 12100, | |
| "train_loss": 2.562843546271324, | |
| "val_loss": 2.712849426269531, | |
| "lr": 0.02, | |
| "tokens_per_sec": 70220.87609033482, | |
| "mfu": 21.492557059524252, | |
| "total_flops": 1.5158426802310873e+19 | |
| }, | |
| { | |
| "step": 12150, | |
| "train_loss": 2.764897644519806, | |
| "val_loss": 2.728255307674408, | |
| "lr": 0.02, | |
| "tokens_per_sec": 70222.98788337511, | |
| "mfu": 21.49320341762947, | |
| "total_flops": 1.5216500615820607e+19 | |
| }, | |
| { | |
| "step": 12200, | |
| "train_loss": 2.6027358323335648, | |
| "val_loss": 2.718504363298416, | |
| "lr": 0.02, | |
| "tokens_per_sec": 70226.1347859551, | |
| "mfu": 21.494166592500346, | |
| "total_flops": 1.5274574429330342e+19 | |
| }, | |
| { | |
| "step": 12250, | |
| "train_loss": 2.600238487124443, | |
| "val_loss": 2.7240344643592835, | |
| "lr": 0.02, | |
| "tokens_per_sec": 70217.72225994278, | |
| "mfu": 21.49159176425266, | |
| "total_flops": 1.5332648242840076e+19 | |
| }, | |
| { | |
| "step": 12300, | |
| "train_loss": 2.624945491552353, | |
| "val_loss": 2.7019182324409483, | |
| "lr": 0.02, | |
| "tokens_per_sec": 70037.50231473647, | |
| "mfu": 21.43643170827973, | |
| "total_flops": 1.539072205634981e+19 | |
| }, | |
| { | |
| "step": 12350, | |
| "train_loss": 2.6347887068986893, | |
| "val_loss": 2.713118499517441, | |
| "lr": 0.02, | |
| "tokens_per_sec": 70236.29818544973, | |
| "mfu": 21.497277311929082, | |
| "total_flops": 1.5448795869859545e+19 | |
| }, | |
| { | |
| "step": 12400, | |
| "train_loss": 2.6599233150482178, | |
| "val_loss": 2.7061276376247405, | |
| "lr": 0.02, | |
| "tokens_per_sec": 70235.88967720557, | |
| "mfu": 21.497152279499474, | |
| "total_flops": 1.550686968336928e+19 | |
| }, | |
| { | |
| "step": 12450, | |
| "train_loss": 2.60282726585865, | |
| "val_loss": 2.705427312850952, | |
| "lr": 0.02, | |
| "tokens_per_sec": 70221.31604141547, | |
| "mfu": 21.49269171568679, | |
| "total_flops": 1.5564943496879014e+19 | |
| }, | |
| { | |
| "step": 12500, | |
| "train_loss": 2.509654685854912, | |
| "val_loss": 2.7021617352962495, | |
| "lr": 0.02, | |
| "tokens_per_sec": 70214.04645600512, | |
| "mfu": 21.490466708140172, | |
| "total_flops": 1.5623017310388748e+19 | |
| }, | |
| { | |
| "step": 12550, | |
| "train_loss": 2.6215831637382507, | |
| "val_loss": 2.7008169412612917, | |
| "lr": 0.02, | |
| "tokens_per_sec": 70214.42803116795, | |
| "mfu": 21.490583497141053, | |
| "total_flops": 1.5681091123898483e+19 | |
| }, | |
| { | |
| "step": 12600, | |
| "train_loss": 2.5994905531406403, | |
| "val_loss": 2.7037415862083436, | |
| "lr": 0.02, | |
| "tokens_per_sec": 70192.67590175563, | |
| "mfu": 21.483925806314765, | |
| "total_flops": 1.5739164937408217e+19 | |
| }, | |
| { | |
| "step": 12650, | |
| "train_loss": 2.525320991873741, | |
| "val_loss": 2.7214488089084625, | |
| "lr": 0.02, | |
| "tokens_per_sec": 70028.52955854105, | |
| "mfu": 21.43368540995296, | |
| "total_flops": 1.5797238750917951e+19 | |
| }, | |
| { | |
| "step": 12700, | |
| "train_loss": 2.614622578024864, | |
| "val_loss": 2.7287656903266906, | |
| "lr": 0.02, | |
| "tokens_per_sec": 70190.07763727287, | |
| "mfu": 21.483130553524422, | |
| "total_flops": 1.5855312564427686e+19 | |
| }, | |
| { | |
| "step": 12750, | |
| "train_loss": 2.609872132539749, | |
| "val_loss": 2.728152310848236, | |
| "lr": 0.02, | |
| "tokens_per_sec": 70161.95478187368, | |
| "mfu": 21.474522969740267, | |
| "total_flops": 1.591338637793742e+19 | |
| }, | |
| { | |
| "step": 12800, | |
| "train_loss": 2.667870581150055, | |
| "val_loss": 2.7394967794418337, | |
| "lr": 0.02, | |
| "tokens_per_sec": 70205.6370498768, | |
| "mfu": 21.48789283479776, | |
| "total_flops": 1.5971460191447155e+19 | |
| }, | |
| { | |
| "step": 12550, | |
| "train_loss": 2.511006385087967, | |
| "val_loss": 2.526371055841446, | |
| "lr": 0.02, | |
| "tokens_per_sec": 70175.22683880641, | |
| "mfu": 21.56199713951752, | |
| "total_flops": 1.602953400495689e+19 | |
| }, | |
| { | |
| "step": 12600, | |
| "train_loss": 2.5791891515254974, | |
| "val_loss": 2.5669545888900758, | |
| "lr": 0.02, | |
| "tokens_per_sec": 70163.2790416696, | |
| "mfu": 21.558326066700335, | |
| "total_flops": 1.6087607818466623e+19 | |
| }, | |
| { | |
| "step": 12650, | |
| "train_loss": 2.576032876968384, | |
| "val_loss": 2.592360574007034, | |
| "lr": 0.02, | |
| "tokens_per_sec": 70180.9795146999, | |
| "mfu": 21.563764703182752, | |
| "total_flops": 1.6145681631976358e+19 | |
| }, | |
| { | |
| "step": 12700, | |
| "train_loss": 2.639482229948044, | |
| "val_loss": 2.603729647397995, | |
| "lr": 0.02, | |
| "tokens_per_sec": 70206.94939833201, | |
| "mfu": 21.57174419369264, | |
| "total_flops": 1.6203755445486092e+19 | |
| }, | |
| { | |
| "step": 12750, | |
| "train_loss": 2.5492171943187714, | |
| "val_loss": 2.6102789878845214, | |
| "lr": 0.02, | |
| "tokens_per_sec": 70194.73188656349, | |
| "mfu": 21.567990248523213, | |
| "total_flops": 1.6261829258995827e+19 | |
| }, | |
| { | |
| "step": 12800, | |
| "train_loss": 2.636518955230713, | |
| "val_loss": 2.6257937788963317, | |
| "lr": 0.02, | |
| "tokens_per_sec": 70190.43946045496, | |
| "mfu": 21.566671360310906, | |
| "total_flops": 1.6319903072505561e+19 | |
| }, | |
| { | |
| "step": 12850, | |
| "train_loss": 2.5141112208366394, | |
| "val_loss": 2.627405750751495, | |
| "lr": 0.02, | |
| "tokens_per_sec": 70171.50622025452, | |
| "mfu": 21.56085394454421, | |
| "total_flops": 1.6377976886015295e+19 | |
| }, | |
| { | |
| "step": 12900, | |
| "train_loss": 2.7366262525320053, | |
| "val_loss": 2.6409005224704742, | |
| "lr": 0.02, | |
| "tokens_per_sec": 70196.86765263397, | |
| "mfu": 21.568646482695538, | |
| "total_flops": 1.643605069952503e+19 | |
| }, | |
| { | |
| "step": 12950, | |
| "train_loss": 2.619333431124687, | |
| "val_loss": 2.658438378572464, | |
| "lr": 0.02, | |
| "tokens_per_sec": 70251.94520442004, | |
| "mfu": 21.58556957746217, | |
| "total_flops": 1.6494124513034764e+19 | |
| }, | |
| { | |
| "step": 13000, | |
| "train_loss": 2.611935794353485, | |
| "val_loss": 2.658729982376099, | |
| "lr": 0.02, | |
| "tokens_per_sec": 70170.87432532028, | |
| "mfu": 21.56065978889444, | |
| "total_flops": 1.6552198326544499e+19 | |
| }, | |
| { | |
| "step": 13050, | |
| "train_loss": 2.6581060141324997, | |
| "val_loss": 2.649619275331497, | |
| "lr": 0.02, | |
| "tokens_per_sec": 70227.76581436433, | |
| "mfu": 21.57814023290994, | |
| "total_flops": 1.6610272140054233e+19 | |
| }, | |
| { | |
| "step": 13100, | |
| "train_loss": 2.7615827918052673, | |
| "val_loss": 2.641291284561157, | |
| "lr": 0.02, | |
| "tokens_per_sec": 70210.33651511092, | |
| "mfu": 21.57278491711582, | |
| "total_flops": 1.6668345953563967e+19 | |
| }, | |
| { | |
| "step": 13150, | |
| "train_loss": 2.6068419814109802, | |
| "val_loss": 2.628552460670471, | |
| "lr": 0.02, | |
| "tokens_per_sec": 70227.09287280244, | |
| "mfu": 21.577933465298543, | |
| "total_flops": 1.6726419767073702e+19 | |
| }, | |
| { | |
| "step": 13200, | |
| "train_loss": 2.5915939286351204, | |
| "val_loss": 2.6240718185901644, | |
| "lr": 0.02, | |
| "tokens_per_sec": 70242.12482435642, | |
| "mfu": 21.58255217350953, | |
| "total_flops": 1.6784493580583436e+19 | |
| }, | |
| { | |
| "step": 13250, | |
| "train_loss": 2.692144051194191, | |
| "val_loss": 2.6368056356906893, | |
| "lr": 0.02, | |
| "tokens_per_sec": 70234.45645879045, | |
| "mfu": 21.58019599621098, | |
| "total_flops": 1.684256739409317e+19 | |
| }, | |
| { | |
| "step": 13300, | |
| "train_loss": 2.404242806136608, | |
| "val_loss": 2.6440458834171294, | |
| "lr": 0.02, | |
| "tokens_per_sec": 70212.24254943595, | |
| "mfu": 21.573370564053, | |
| "total_flops": 1.6900641207602905e+19 | |
| }, | |
| { | |
| "step": 13350, | |
| "train_loss": 2.6278859823942184, | |
| "val_loss": 2.6582672238349914, | |
| "lr": 0.02, | |
| "tokens_per_sec": 70219.35244915294, | |
| "mfu": 21.575555147477463, | |
| "total_flops": 1.695871502111264e+19 | |
| }, | |
| { | |
| "step": 13400, | |
| "train_loss": 2.6749037951231003, | |
| "val_loss": 2.660851800441742, | |
| "lr": 0.02, | |
| "tokens_per_sec": 70055.26580612546, | |
| "mfu": 21.52513798622313, | |
| "total_flops": 1.7016788834622374e+19 | |
| }, | |
| { | |
| "step": 13450, | |
| "train_loss": 2.6571804583072662, | |
| "val_loss": 2.6723348915576937, | |
| "lr": 0.02, | |
| "tokens_per_sec": 70243.49485641379, | |
| "mfu": 21.582973128718834, | |
| "total_flops": 1.7074862648132108e+19 | |
| }, | |
| { | |
| "step": 13500, | |
| "train_loss": 2.5777985006570816, | |
| "val_loss": 2.6738951981067656, | |
| "lr": 0.02, | |
| "tokens_per_sec": 70221.8123944745, | |
| "mfu": 21.576310988767563, | |
| "total_flops": 1.7132936461641843e+19 | |
| }, | |
| { | |
| "step": 13550, | |
| "train_loss": 2.5361034870147705, | |
| "val_loss": 2.6723001420497896, | |
| "lr": 0.02, | |
| "tokens_per_sec": 70233.72170186242, | |
| "mfu": 21.579970235248112, | |
| "total_flops": 1.7191010275151577e+19 | |
| }, | |
| { | |
| "step": 13600, | |
| "train_loss": 2.6962577551603317, | |
| "val_loss": 2.6532412350177763, | |
| "lr": 0.02, | |
| "tokens_per_sec": 70226.71687890886, | |
| "mfu": 21.57781793764551, | |
| "total_flops": 1.7249084088661311e+19 | |
| }, | |
| { | |
| "step": 13650, | |
| "train_loss": 2.6913505494594574, | |
| "val_loss": 2.66121444106102, | |
| "lr": 0.02, | |
| "tokens_per_sec": 70210.21893955192, | |
| "mfu": 21.572748790921707, | |
| "total_flops": 1.7307157902171046e+19 | |
| }, | |
| { | |
| "step": 13700, | |
| "train_loss": 2.5043933764100075, | |
| "val_loss": 2.6647586822509766, | |
| "lr": 0.02, | |
| "tokens_per_sec": 70221.3220958266, | |
| "mfu": 21.57616033990585, | |
| "total_flops": 1.736523171568078e+19 | |
| }, | |
| { | |
| "step": 13750, | |
| "train_loss": 2.6589005440473557, | |
| "val_loss": 2.674451744556427, | |
| "lr": 0.02, | |
| "tokens_per_sec": 70083.35522385615, | |
| "mfu": 21.533768723479614, | |
| "total_flops": 1.7423305529190515e+19 | |
| }, | |
| { | |
| "step": 13800, | |
| "train_loss": 2.62132728099823, | |
| "val_loss": 2.6628175556659697, | |
| "lr": 0.02, | |
| "tokens_per_sec": 70216.07364959198, | |
| "mfu": 21.574547705536208, | |
| "total_flops": 1.748137934270025e+19 | |
| }, | |
| { | |
| "step": 13850, | |
| "train_loss": 2.6479734033346176, | |
| "val_loss": 2.6767729818820953, | |
| "lr": 0.02, | |
| "tokens_per_sec": 70236.02270545306, | |
| "mfu": 21.5806772401995, | |
| "total_flops": 1.7539453156209983e+19 | |
| }, | |
| { | |
| "step": 13900, | |
| "train_loss": 2.7008413672447205, | |
| "val_loss": 2.6886323153972627, | |
| "lr": 0.02, | |
| "tokens_per_sec": 70175.82813137904, | |
| "mfu": 21.562181892304434, | |
| "total_flops": 1.7597526969719718e+19 | |
| }, | |
| { | |
| "step": 13950, | |
| "train_loss": 2.6451959162950516, | |
| "val_loss": 2.679285502433777, | |
| "lr": 0.02, | |
| "tokens_per_sec": 70229.14493070339, | |
| "mfu": 21.578563979350065, | |
| "total_flops": 1.7655600783229452e+19 | |
| }, | |
| { | |
| "step": 14000, | |
| "train_loss": 2.6014012545347214, | |
| "val_loss": 2.680087924003601, | |
| "lr": 0.02, | |
| "tokens_per_sec": 70238.65353790967, | |
| "mfu": 21.581485588165858, | |
| "total_flops": 1.7713674596739187e+19 | |
| }, | |
| { | |
| "step": 14050, | |
| "train_loss": 2.5424724519252777, | |
| "val_loss": 2.6780451416969298, | |
| "lr": 0.02, | |
| "tokens_per_sec": 70229.17812525907, | |
| "mfu": 21.578574178688875, | |
| "total_flops": 1.7771748410248921e+19 | |
| }, | |
| { | |
| "step": 14100, | |
| "train_loss": 2.5748010873794556, | |
| "val_loss": 2.6970475554466247, | |
| "lr": 0.02, | |
| "tokens_per_sec": 70071.97021212461, | |
| "mfu": 21.53027057176073, | |
| "total_flops": 1.7829822223758655e+19 | |
| }, | |
| { | |
| "step": 14150, | |
| "train_loss": 2.5738344490528107, | |
| "val_loss": 2.6904390811920167, | |
| "lr": 0.02, | |
| "tokens_per_sec": 70275.71666961232, | |
| "mfu": 21.592873583271164, | |
| "total_flops": 1.788789603726839e+19 | |
| }, | |
| { | |
| "step": 14200, | |
| "train_loss": 2.607429787516594, | |
| "val_loss": 2.699114978313446, | |
| "lr": 0.02, | |
| "tokens_per_sec": 70223.35666132403, | |
| "mfu": 21.576785479252198, | |
| "total_flops": 1.7945969850778124e+19 | |
| }, | |
| { | |
| "step": 14250, | |
| "train_loss": 2.658610910177231, | |
| "val_loss": 2.7006686329841614, | |
| "lr": 0.02, | |
| "tokens_per_sec": 70218.85175854548, | |
| "mfu": 21.575401305588635, | |
| "total_flops": 1.8004043664287859e+19 | |
| }, | |
| { | |
| "step": 14300, | |
| "train_loss": 2.613561823964119, | |
| "val_loss": 2.7155104994773867, | |
| "lr": 0.02, | |
| "tokens_per_sec": 70206.51545496765, | |
| "mfu": 21.57161086052077, | |
| "total_flops": 1.8062117477797593e+19 | |
| }, | |
| { | |
| "step": 14350, | |
| "train_loss": 2.5862353146076202, | |
| "val_loss": 2.706381583213806, | |
| "lr": 0.02, | |
| "tokens_per_sec": 70232.6663078057, | |
| "mfu": 21.579645955517833, | |
| "total_flops": 1.8120191291307327e+19 | |
| }, | |
| { | |
| "step": 14400, | |
| "train_loss": 2.584561437368393, | |
| "val_loss": 2.7018906712532043, | |
| "lr": 0.02, | |
| "tokens_per_sec": 70251.8311925962, | |
| "mfu": 21.585534546259137, | |
| "total_flops": 1.8178265104817062e+19 | |
| }, | |
| { | |
| "step": 14450, | |
| "train_loss": 2.610737219452858, | |
| "val_loss": 2.7197303116321563, | |
| "lr": 0.02, | |
| "tokens_per_sec": 70059.37408795145, | |
| "mfu": 21.526400294376174, | |
| "total_flops": 1.8236338918326796e+19 | |
| }, | |
| { | |
| "step": 14500, | |
| "train_loss": 2.6162658035755157, | |
| "val_loss": 2.71315957903862, | |
| "lr": 0.02, | |
| "tokens_per_sec": 70209.70044612157, | |
| "mfu": 21.57258947894844, | |
| "total_flops": 1.829441273183653e+19 | |
| }, | |
| { | |
| "step": 14550, | |
| "train_loss": 2.719397410750389, | |
| "val_loss": 2.7064169228076933, | |
| "lr": 0.02, | |
| "tokens_per_sec": 70227.90610092366, | |
| "mfu": 21.578183337272076, | |
| "total_flops": 1.8352486545346265e+19 | |
| }, | |
| { | |
| "step": 14600, | |
| "train_loss": 2.550754353404045, | |
| "val_loss": 2.704353654384613, | |
| "lr": 0.02, | |
| "tokens_per_sec": 70209.50901122653, | |
| "mfu": 21.57253065878005, | |
| "total_flops": 1.8410560358856e+19 | |
| }, | |
| { | |
| "step": 14650, | |
| "train_loss": 2.751055210828781, | |
| "val_loss": 2.7144021570682524, | |
| "lr": 0.02, | |
| "tokens_per_sec": 70246.63136750608, | |
| "mfu": 21.583936851192515, | |
| "total_flops": 1.8468634172365734e+19 | |
| }, | |
| { | |
| "step": 14700, | |
| "train_loss": 2.588581830263138, | |
| "val_loss": 2.703622704744339, | |
| "lr": 0.02, | |
| "tokens_per_sec": 70220.90636202156, | |
| "mfu": 21.576032601792008, | |
| "total_flops": 1.852670798587547e+19 | |
| }, | |
| { | |
| "step": 14750, | |
| "train_loss": 2.5866953879594803, | |
| "val_loss": 2.7090446412563325, | |
| "lr": 0.02, | |
| "tokens_per_sec": 70219.54752534849, | |
| "mfu": 21.575615086469607, | |
| "total_flops": 1.8584781799385203e+19 | |
| }, | |
| { | |
| "step": 14800, | |
| "train_loss": 2.6120127886533737, | |
| "val_loss": 2.6936957955360414, | |
| "lr": 0.02, | |
| "tokens_per_sec": 70056.30594351621, | |
| "mfu": 21.525457578199592, | |
| "total_flops": 1.8642855612894937e+19 | |
| }, | |
| { | |
| "step": 14850, | |
| "train_loss": 2.6211347430944443, | |
| "val_loss": 2.7043174862861634, | |
| "lr": 0.02, | |
| "tokens_per_sec": 70201.50397715726, | |
| "mfu": 21.570071036923753, | |
| "total_flops": 1.870092942640467e+19 | |
| }, | |
| { | |
| "step": 14900, | |
| "train_loss": 2.646672412753105, | |
| "val_loss": 2.7002520680427553, | |
| "lr": 0.02, | |
| "tokens_per_sec": 70214.55380391362, | |
| "mfu": 21.57408071868564, | |
| "total_flops": 1.8759003239914406e+19 | |
| }, | |
| { | |
| "step": 14950, | |
| "train_loss": 2.5893217772245407, | |
| "val_loss": 2.68537123799324, | |
| "lr": 0.02, | |
| "tokens_per_sec": 70209.82900398185, | |
| "mfu": 21.572628979557706, | |
| "total_flops": 1.881707705342414e+19 | |
| }, | |
| { | |
| "step": 15000, | |
| "train_loss": 2.496798686683178, | |
| "val_loss": 2.6873112499713896, | |
| "lr": 0.02, | |
| "tokens_per_sec": 70225.07861072318, | |
| "mfu": 21.57731456436801, | |
| "total_flops": 1.8875150866933875e+19 | |
| }, | |
| { | |
| "step": 15050, | |
| "train_loss": 2.6120809465646744, | |
| "val_loss": 2.6986561715602875, | |
| "lr": 0.02, | |
| "tokens_per_sec": 70235.34377313984, | |
| "mfu": 21.580468631873526, | |
| "total_flops": 1.893322468044361e+19 | |
| }, | |
| { | |
| "step": 15100, | |
| "train_loss": 2.589638665318489, | |
| "val_loss": 2.6991723120212554, | |
| "lr": 0.02, | |
| "tokens_per_sec": 70226.08421351104, | |
| "mfu": 21.577623545263574, | |
| "total_flops": 1.8991298493953343e+19 | |
| }, | |
| { | |
| "step": 15150, | |
| "train_loss": 2.5153046399354935, | |
| "val_loss": 2.7059680104255674, | |
| "lr": 0.02, | |
| "tokens_per_sec": 70237.30029581553, | |
| "mfu": 21.58106979183035, | |
| "total_flops": 1.9049372307463078e+19 | |
| }, | |
| { | |
| "step": 15200, | |
| "train_loss": 2.603958636522293, | |
| "val_loss": 2.7175380647182465, | |
| "lr": 0.02, | |
| "tokens_per_sec": 70217.51508637612, | |
| "mfu": 21.57499060051805, | |
| "total_flops": 1.9107446120972812e+19 | |
| }, | |
| { | |
| "step": 15250, | |
| "train_loss": 2.599534645676613, | |
| "val_loss": 2.717556869983673, | |
| "lr": 0.02, | |
| "tokens_per_sec": 70212.6102057161, | |
| "mfu": 21.57348352989605, | |
| "total_flops": 1.9165519934482547e+19 | |
| }, | |
| { | |
| "step": 15300, | |
| "train_loss": 2.6571564823389053, | |
| "val_loss": 2.720191490650177, | |
| "lr": 0.02, | |
| "tokens_per_sec": 70215.26484315922, | |
| "mfu": 21.574299192167956, | |
| "total_flops": 1.922359374799228e+19 | |
| }, | |
| { | |
| "step": 15050, | |
| "train_loss": 2.5011045336723328, | |
| "val_loss": 2.5259799718856812, | |
| "lr": 0.02, | |
| "tokens_per_sec": 70822.64640515842, | |
| "mfu": 21.658787127552376, | |
| "total_flops": 1.9281667561502015e+19 | |
| }, | |
| { | |
| "step": 15100, | |
| "train_loss": 2.5692661106586456, | |
| "val_loss": 2.5606980443000795, | |
| "lr": 0.02, | |
| "tokens_per_sec": 70767.62368058978, | |
| "mfu": 21.64196022910816, | |
| "total_flops": 1.933974137501175e+19 | |
| }, | |
| { | |
| "step": 15150, | |
| "train_loss": 2.5663330778479576, | |
| "val_loss": 2.5890700042247774, | |
| "lr": 0.02, | |
| "tokens_per_sec": 70799.76122967034, | |
| "mfu": 21.65178844606514, | |
| "total_flops": 1.9397815188521484e+19 | |
| }, | |
| { | |
| "step": 15200, | |
| "train_loss": 2.6242035180330276, | |
| "val_loss": 2.595632380247116, | |
| "lr": 0.02, | |
| "tokens_per_sec": 70832.7004053005, | |
| "mfu": 21.66186181425092, | |
| "total_flops": 1.945588900203122e+19 | |
| }, | |
| { | |
| "step": 15250, | |
| "train_loss": 2.5373122468590736, | |
| "val_loss": 2.5982221364974976, | |
| "lr": 0.02, | |
| "tokens_per_sec": 70710.97738859533, | |
| "mfu": 21.62463681573479, | |
| "total_flops": 1.9513962815540953e+19 | |
| }, | |
| { | |
| "step": 15300, | |
| "train_loss": 2.626233994960785, | |
| "val_loss": 2.6194027781486513, | |
| "lr": 0.02, | |
| "tokens_per_sec": 70690.91638907796, | |
| "mfu": 21.618501815983056, | |
| "total_flops": 1.9572036629050687e+19 | |
| }, | |
| { | |
| "step": 15350, | |
| "train_loss": 2.5037445425987244, | |
| "val_loss": 2.6230907142162323, | |
| "lr": 0.02, | |
| "tokens_per_sec": 70841.31925088949, | |
| "mfu": 21.664497605927345, | |
| "total_flops": 1.963011044256042e+19 | |
| }, | |
| { | |
| "step": 15400, | |
| "train_loss": 2.725143328309059, | |
| "val_loss": 2.6347573161125184, | |
| "lr": 0.02, | |
| "tokens_per_sec": 70836.69318627186, | |
| "mfu": 21.663082875556803, | |
| "total_flops": 1.9688184256070156e+19 | |
| }, | |
| { | |
| "step": 15450, | |
| "train_loss": 2.607233941555023, | |
| "val_loss": 2.6485621273517608, | |
| "lr": 0.02, | |
| "tokens_per_sec": 70812.53403285159, | |
| "mfu": 21.655694589638795, | |
| "total_flops": 1.974625806957989e+19 | |
| }, | |
| { | |
| "step": 15500, | |
| "train_loss": 2.6106958985328674, | |
| "val_loss": 2.6626220107078553, | |
| "lr": 0.02, | |
| "tokens_per_sec": 70740.52877982834, | |
| "mfu": 21.63367414100469, | |
| "total_flops": 1.9804331883089625e+19 | |
| }, | |
| { | |
| "step": 15550, | |
| "train_loss": 2.654520571231842, | |
| "val_loss": 2.6427935302257537, | |
| "lr": 0.02, | |
| "tokens_per_sec": 70855.78450661922, | |
| "mfu": 21.66892133068911, | |
| "total_flops": 1.986240569659936e+19 | |
| }, | |
| { | |
| "step": 15600, | |
| "train_loss": 2.7503251135349274, | |
| "val_loss": 2.63273109793663, | |
| "lr": 0.02, | |
| "tokens_per_sec": 70792.46802091923, | |
| "mfu": 21.649558057004086, | |
| "total_flops": 1.9920479510109094e+19 | |
| }, | |
| { | |
| "step": 15650, | |
| "train_loss": 2.598275765776634, | |
| "val_loss": 2.621151161193848, | |
| "lr": 0.02, | |
| "tokens_per_sec": 70701.00678300665, | |
| "mfu": 21.621587632529426, | |
| "total_flops": 1.997855332361883e+19 | |
| }, | |
| { | |
| "step": 15700, | |
| "train_loss": 2.5822859704494476, | |
| "val_loss": 2.617991364002228, | |
| "lr": 0.02, | |
| "tokens_per_sec": 70797.39009105547, | |
| "mfu": 21.651063310969004, | |
| "total_flops": 2.0036627137128563e+19 | |
| }, | |
| { | |
| "step": 15750, | |
| "train_loss": 2.6830928325653076, | |
| "val_loss": 2.6261918783187865, | |
| "lr": 0.02, | |
| "tokens_per_sec": 70786.01949477864, | |
| "mfu": 21.647585986458083, | |
| "total_flops": 2.0094700950638297e+19 | |
| }, | |
| { | |
| "step": 15800, | |
| "train_loss": 2.395069792866707, | |
| "val_loss": 2.6370294749736787, | |
| "lr": 0.02, | |
| "tokens_per_sec": 70823.31358734316, | |
| "mfu": 21.658991163374928, | |
| "total_flops": 2.015277476414803e+19 | |
| }, | |
| { | |
| "step": 15850, | |
| "train_loss": 2.6195893734693527, | |
| "val_loss": 2.6514434933662416, | |
| "lr": 0.02, | |
| "tokens_per_sec": 70822.91954916564, | |
| "mfu": 21.658870659702163, | |
| "total_flops": 2.0210848577657766e+19 | |
| }, | |
| { | |
| "step": 15900, | |
| "train_loss": 2.6659241691231728, | |
| "val_loss": 2.6493006229400633, | |
| "lr": 0.02, | |
| "tokens_per_sec": 70844.27349820061, | |
| "mfu": 21.66540106572279, | |
| "total_flops": 2.02689223911675e+19 | |
| }, | |
| { | |
| "step": 15950, | |
| "train_loss": 2.6470499634742737, | |
| "val_loss": 2.6604777455329893, | |
| "lr": 0.02, | |
| "tokens_per_sec": 70789.01379554643, | |
| "mfu": 21.648501695291586, | |
| "total_flops": 2.0326996204677235e+19 | |
| }, | |
| { | |
| "step": 16000, | |
| "train_loss": 2.567369759082794, | |
| "val_loss": 2.66320680975914, | |
| "lr": 0.02, | |
| "tokens_per_sec": 70693.97673794771, | |
| "mfu": 21.619437723465627, | |
| "total_flops": 2.038507001818697e+19 | |
| }, | |
| { | |
| "step": 16050, | |
| "train_loss": 2.5259591192007065, | |
| "val_loss": 2.6549350917339325, | |
| "lr": 0.01943255903154212, | |
| "tokens_per_sec": 70815.6263577521, | |
| "mfu": 21.656640275943236, | |
| "total_flops": 2.0443143831696703e+19 | |
| }, | |
| { | |
| "step": 16100, | |
| "train_loss": 2.6835711896419525, | |
| "val_loss": 2.635953629016876, | |
| "lr": 0.018881217525718468, | |
| "tokens_per_sec": 70834.84756150097, | |
| "mfu": 21.66251845166051, | |
| "total_flops": 2.0501217645206438e+19 | |
| }, | |
| { | |
| "step": 16150, | |
| "train_loss": 2.676758661866188, | |
| "val_loss": 2.644348645210266, | |
| "lr": 0.01834551870779559, | |
| "tokens_per_sec": 70833.7125334854, | |
| "mfu": 21.662171340511456, | |
| "total_flops": 2.0559291458716172e+19 | |
| }, | |
| { | |
| "step": 16200, | |
| "train_loss": 2.491854965686798, | |
| "val_loss": 2.6453607082366943, | |
| "lr": 0.01782501876267491, | |
| "tokens_per_sec": 70804.74951601207, | |
| "mfu": 21.653313950088084, | |
| "total_flops": 2.0617365272225907e+19 | |
| }, | |
| { | |
| "step": 16250, | |
| "train_loss": 2.638686791062355, | |
| "val_loss": 2.6507443487644196, | |
| "lr": 0.01731928646720131, | |
| "tokens_per_sec": 70825.65635173608, | |
| "mfu": 21.65970762114447, | |
| "total_flops": 2.067543908573564e+19 | |
| }, | |
| { | |
| "step": 16300, | |
| "train_loss": 2.600520685315132, | |
| "val_loss": 2.6338987708091737, | |
| "lr": 0.016827902832903904, | |
| "tokens_per_sec": 70849.17663872243, | |
| "mfu": 21.66690053068504, | |
| "total_flops": 2.0733512899245375e+19 | |
| }, | |
| { | |
| "step": 16350, | |
| "train_loss": 2.6237353086471558, | |
| "val_loss": 2.6439998149871826, | |
| "lr": 0.016350460758873, | |
| "tokens_per_sec": 70703.86261251083, | |
| "mfu": 21.622460994461562, | |
| "total_flops": 2.079158671275511e+19 | |
| }, | |
| { | |
| "step": 16400, | |
| "train_loss": 2.673539698123932, | |
| "val_loss": 2.6453349232673644, | |
| "lr": 0.01588656469448563, | |
| "tokens_per_sec": 70845.07186851281, | |
| "mfu": 21.665645221138018, | |
| "total_flops": 2.0849660526264844e+19 | |
| }, | |
| { | |
| "step": 16450, | |
| "train_loss": 2.6117686182260513, | |
| "val_loss": 2.6378572285175323, | |
| "lr": 0.01543583031170025, | |
| "tokens_per_sec": 70741.21808301542, | |
| "mfu": 21.6338849418114, | |
| "total_flops": 2.090773433977458e+19 | |
| }, | |
| { | |
| "step": 16500, | |
| "train_loss": 2.569175586104393, | |
| "val_loss": 2.630873453617096, | |
| "lr": 0.014997884186649118, | |
| "tokens_per_sec": 70848.69488819782, | |
| "mfu": 21.666753203063287, | |
| "total_flops": 2.0965808153284313e+19 | |
| }, | |
| { | |
| "step": 16550, | |
| "train_loss": 2.5066262930631638, | |
| "val_loss": 2.6248888432979585, | |
| "lr": 0.014572363490264556, | |
| "tokens_per_sec": 70831.78663993345, | |
| "mfu": 21.661582369037077, | |
| "total_flops": 2.1023881966794047e+19 | |
| }, | |
| { | |
| "step": 16600, | |
| "train_loss": 2.533449575304985, | |
| "val_loss": 2.643479061126709, | |
| "lr": 0.014158915687682759, | |
| "tokens_per_sec": 70832.01285455696, | |
| "mfu": 21.661651549371683, | |
| "total_flops": 2.108195578030378e+19 | |
| }, | |
| { | |
| "step": 16650, | |
| "train_loss": 2.5322001576423645, | |
| "val_loss": 2.6342837035655977, | |
| "lr": 0.013757198246176152, | |
| "tokens_per_sec": 70853.31213328421, | |
| "mfu": 21.668165236269022, | |
| "total_flops": 2.1140029593813516e+19 | |
| }, | |
| { | |
| "step": 16700, | |
| "train_loss": 2.5647076219320297, | |
| "val_loss": 2.636631351709366, | |
| "lr": 0.013366878351372293, | |
| "tokens_per_sec": 70662.51164198358, | |
| "mfu": 21.60981515427305, | |
| "total_flops": 2.119810340732325e+19 | |
| }, | |
| { | |
| "step": 16750, | |
| "train_loss": 2.6134786158800125, | |
| "val_loss": 2.632091587781906, | |
| "lr": 0.012987632631524226, | |
| "tokens_per_sec": 70829.18487637353, | |
| "mfu": 21.660786704853933, | |
| "total_flops": 2.1256177220832985e+19 | |
| }, | |
| { | |
| "step": 16800, | |
| "train_loss": 2.563320219516754, | |
| "val_loss": 2.6496681392192842, | |
| "lr": 0.012619146889603866, | |
| "tokens_per_sec": 70828.32515462447, | |
| "mfu": 21.66052378711085, | |
| "total_flops": 2.131425103434272e+19 | |
| }, | |
| { | |
| "step": 16850, | |
| "train_loss": 2.5337288677692413, | |
| "val_loss": 2.62899107336998, | |
| "lr": 0.012261115842996415, | |
| "tokens_per_sec": 70843.56255750875, | |
| "mfu": 21.66518364779382, | |
| "total_flops": 2.1372324847852454e+19 | |
| }, | |
| { | |
| "step": 16900, | |
| "train_loss": 2.531308576464653, | |
| "val_loss": 2.624918353557587, | |
| "lr": 0.01191324287058021, | |
| "tokens_per_sec": 70843.99254368903, | |
| "mfu": 21.66531514498602, | |
| "total_flops": 2.143039866136219e+19 | |
| }, | |
| { | |
| "step": 16950, | |
| "train_loss": 2.552768364548683, | |
| "val_loss": 2.6276935577392577, | |
| "lr": 0.011575239766982413, | |
| "tokens_per_sec": 70791.12024610583, | |
| "mfu": 21.649145884213866, | |
| "total_flops": 2.1488472474871923e+19 | |
| }, | |
| { | |
| "step": 17000, | |
| "train_loss": 2.5594132989645004, | |
| "val_loss": 2.6281732857227325, | |
| "lr": 0.011246826503806982, | |
| "tokens_per_sec": 70846.84612722936, | |
| "mfu": 21.66618782006367, | |
| "total_flops": 2.1546546288381657e+19 | |
| }, | |
| { | |
| "step": 17050, | |
| "train_loss": 2.6567023247480392, | |
| "val_loss": 2.620693105459213, | |
| "lr": 0.010927730997637084, | |
| "tokens_per_sec": 70635.0011758644, | |
| "mfu": 21.601401979113696, | |
| "total_flops": 2.160462010189139e+19 | |
| }, | |
| { | |
| "step": 17100, | |
| "train_loss": 2.491407111287117, | |
| "val_loss": 2.6200718581676483, | |
| "lr": 0.010617688884619766, | |
| "tokens_per_sec": 70835.64195443952, | |
| "mfu": 21.66276139072626, | |
| "total_flops": 2.1662693915401126e+19 | |
| }, | |
| { | |
| "step": 17150, | |
| "train_loss": 2.6877047419548035, | |
| "val_loss": 2.6259263932704924, | |
| "lr": 0.010316443301446114, | |
| "tokens_per_sec": 70749.92382525704, | |
| "mfu": 21.636547307983363, | |
| "total_flops": 2.172076772891086e+19 | |
| }, | |
| { | |
| "step": 17200, | |
| "train_loss": 2.525592103600502, | |
| "val_loss": 2.611518979072571, | |
| "lr": 0.010023744672545448, | |
| "tokens_per_sec": 70787.86815875485, | |
| "mfu": 21.648151339795795, | |
| "total_flops": 2.1778841542420595e+19 | |
| }, | |
| { | |
| "step": 17250, | |
| "train_loss": 2.5193743258714676, | |
| "val_loss": 2.610617792606354, | |
| "lr": 0.009739350503317262, | |
| "tokens_per_sec": 70782.25900458822, | |
| "mfu": 21.64643596368057, | |
| "total_flops": 2.183691535593033e+19 | |
| }, | |
| { | |
| "step": 17300, | |
| "train_loss": 2.5447549670934677, | |
| "val_loss": 2.6041156351566315, | |
| "lr": 0.00946302517922961, | |
| "tokens_per_sec": 70798.21395966616, | |
| "mfu": 21.65131526420382, | |
| "total_flops": 2.1894989169440063e+19 | |
| }, | |
| { | |
| "step": 17050, | |
| "train_loss": 2.4430361688137054, | |
| "val_loss": 2.492651104927063, | |
| "lr": 0.010927730997637084, | |
| "tokens_per_sec": 70194.53280299473, | |
| "mfu": 21.56194795535554, | |
| "total_flops": 2.1953062982949798e+19 | |
| }, | |
| { | |
| "step": 17100, | |
| "train_loss": 2.5078289210796356, | |
| "val_loss": 2.5014576375484467, | |
| "lr": 0.010617688884619766, | |
| "tokens_per_sec": 70187.43364078536, | |
| "mfu": 21.559767276035185, | |
| "total_flops": 2.2011136796459532e+19 | |
| }, | |
| { | |
| "step": 17150, | |
| "train_loss": 2.500446632504463, | |
| "val_loss": 2.515454649925232, | |
| "lr": 0.010316443301446114, | |
| "tokens_per_sec": 70188.95186454445, | |
| "mfu": 21.560233635171304, | |
| "total_flops": 2.2069210609969267e+19 | |
| }, | |
| { | |
| "step": 17200, | |
| "train_loss": 2.5545513927936554, | |
| "val_loss": 2.526023209095001, | |
| "lr": 0.010023744672545448, | |
| "tokens_per_sec": 70195.58066519728, | |
| "mfu": 21.56226983156688, | |
| "total_flops": 2.2127284423479e+19 | |
| }, | |
| { | |
| "step": 17250, | |
| "train_loss": 2.4661522805690765, | |
| "val_loss": 2.52562854886055, | |
| "lr": 0.009739350503317262, | |
| "tokens_per_sec": 70194.09453183571, | |
| "mfu": 21.561813329772264, | |
| "total_flops": 2.2185358236988735e+19 | |
| }, | |
| { | |
| "step": 17300, | |
| "train_loss": 2.5579015016555786, | |
| "val_loss": 2.5291117131710052, | |
| "lr": 0.00946302517922961, | |
| "tokens_per_sec": 70178.53306763705, | |
| "mfu": 21.557033249789363, | |
| "total_flops": 2.224343205049847e+19 | |
| }, | |
| { | |
| "step": 17350, | |
| "train_loss": 2.43189400434494, | |
| "val_loss": 2.528818500041962, | |
| "lr": 0.009194539770617445, | |
| "tokens_per_sec": 70201.79005590732, | |
| "mfu": 21.564177196057784, | |
| "total_flops": 2.2301505864008204e+19 | |
| }, | |
| { | |
| "step": 17400, | |
| "train_loss": 2.6497219800949097, | |
| "val_loss": 2.5407229542732237, | |
| "lr": 0.008933671843019263, | |
| "tokens_per_sec": 70196.11978921283, | |
| "mfu": 21.562435436543513, | |
| "total_flops": 2.235957967751794e+19 | |
| }, | |
| { | |
| "step": 17450, | |
| "train_loss": 2.535208210349083, | |
| "val_loss": 2.543413257598877, | |
| "lr": 0.008680205272894878, | |
| "tokens_per_sec": 70208.10410382063, | |
| "mfu": 21.566116708539145, | |
| "total_flops": 2.2417653491027673e+19 | |
| }, | |
| { | |
| "step": 17050, | |
| "train_loss": 2.4430273175239563, | |
| "val_loss": 2.492583030462265, | |
| "lr": 0.010927730997637084, | |
| "tokens_per_sec": 69976.98224120555, | |
| "mfu": 21.590922027044734, | |
| "total_flops": 2.2475727304537407e+19 | |
| }, | |
| { | |
| "step": 17100, | |
| "train_loss": 2.5079168379306793, | |
| "val_loss": 2.5017378509044645, | |
| "lr": 0.010617688884619766, | |
| "tokens_per_sec": 70199.67471947223, | |
| "mfu": 21.65963227690506, | |
| "total_flops": 2.253380111804714e+19 | |
| }, | |
| { | |
| "step": 17150, | |
| "train_loss": 2.5002063512802124, | |
| "val_loss": 2.5166682302951813, | |
| "lr": 0.010316443301446114, | |
| "tokens_per_sec": 70221.68480180511, | |
| "mfu": 21.666423338140316, | |
| "total_flops": 2.2591874931556876e+19 | |
| }, | |
| { | |
| "step": 17200, | |
| "train_loss": 2.5549302101135254, | |
| "val_loss": 2.526432579755783, | |
| "lr": 0.010023744672545448, | |
| "tokens_per_sec": 70243.03028256173, | |
| "mfu": 21.6730093410785, | |
| "total_flops": 2.264994874506661e+19 | |
| }, | |
| { | |
| "step": 17250, | |
| "train_loss": 2.466187208890915, | |
| "val_loss": 2.5247345089912416, | |
| "lr": 0.009739350503317262, | |
| "tokens_per_sec": 70228.35040146303, | |
| "mfu": 21.668479963588688, | |
| "total_flops": 2.2708022558576345e+19 | |
| }, | |
| { | |
| "step": 17300, | |
| "train_loss": 2.557755470275879, | |
| "val_loss": 2.528937244415283, | |
| "lr": 0.00946302517922961, | |
| "tokens_per_sec": 70228.95406261715, | |
| "mfu": 21.668666219132895, | |
| "total_flops": 2.276609637208608e+19 | |
| }, | |
| { | |
| "step": 17350, | |
| "train_loss": 2.432028129696846, | |
| "val_loss": 2.527354818582535, | |
| "lr": 0.009194539770617445, | |
| "tokens_per_sec": 70061.32081891438, | |
| "mfu": 21.616944121694466, | |
| "total_flops": 2.2824170185595814e+19 | |
| }, | |
| { | |
| "step": 17400, | |
| "train_loss": 2.649830251932144, | |
| "val_loss": 2.5389151513576507, | |
| "lr": 0.008933671843019263, | |
| "tokens_per_sec": 70234.34250372445, | |
| "mfu": 21.670328785995153, | |
| "total_flops": 2.288224399910555e+19 | |
| }, | |
| { | |
| "step": 17450, | |
| "train_loss": 2.534702032804489, | |
| "val_loss": 2.542128974199295, | |
| "lr": 0.008680205272894878, | |
| "tokens_per_sec": 70238.1502163951, | |
| "mfu": 21.671503629847237, | |
| "total_flops": 2.2940317812615283e+19 | |
| }, | |
| { | |
| "step": 17500, | |
| "train_loss": 2.5225761979818344, | |
| "val_loss": 2.539433038234711, | |
| "lr": 0.008433930068571646, | |
| "tokens_per_sec": 70230.80548197155, | |
| "mfu": 21.669237461985045, | |
| "total_flops": 2.2998391626125017e+19 | |
| }, | |
| { | |
| "step": 17550, | |
| "train_loss": 2.5620259642601013, | |
| "val_loss": 2.5314638674259187, | |
| "lr": 0.008194642196270831, | |
| "tokens_per_sec": 70189.97693235113, | |
| "mfu": 21.6566400906338, | |
| "total_flops": 2.305646543963475e+19 | |
| }, | |
| { | |
| "step": 17600, | |
| "train_loss": 2.6607564091682434, | |
| "val_loss": 2.5253191769123076, | |
| "lr": 0.007962143411069945, | |
| "tokens_per_sec": 70225.33539103418, | |
| "mfu": 21.66754970262294, | |
| "total_flops": 2.3114539253144486e+19 | |
| }, | |
| { | |
| "step": 17650, | |
| "train_loss": 2.5130698829889297, | |
| "val_loss": 2.5219079494476317, | |
| "lr": 0.007736241092661044, | |
| "tokens_per_sec": 70207.26667798325, | |
| "mfu": 21.66197472407832, | |
| "total_flops": 2.317261306665422e+19 | |
| }, | |
| { | |
| "step": 17700, | |
| "train_loss": 2.4957679137587547, | |
| "val_loss": 2.5156410336494446, | |
| "lr": 0.007516748085768884, | |
| "tokens_per_sec": 70217.80555572028, | |
| "mfu": 21.66522642313971, | |
| "total_flops": 2.3230686880163955e+19 | |
| }, | |
| { | |
| "step": 17750, | |
| "train_loss": 2.5936799198389053, | |
| "val_loss": 2.5133773267269133, | |
| "lr": 0.007303482545096754, | |
| "tokens_per_sec": 70207.28292869608, | |
| "mfu": 21.66197973812526, | |
| "total_flops": 2.328876069367369e+19 | |
| }, | |
| { | |
| "step": 17800, | |
| "train_loss": 2.3071534037590027, | |
| "val_loss": 2.523169732093811, | |
| "lr": 0.00709626778467151, | |
| "tokens_per_sec": 70210.6295030122, | |
| "mfu": 21.663012300873792, | |
| "total_flops": 2.3346834507183423e+19 | |
| }, | |
| { | |
| "step": 17850, | |
| "train_loss": 2.52487476170063, | |
| "val_loss": 2.5232641220092775, | |
| "lr": 0.0068949321314629875, | |
| "tokens_per_sec": 70216.21063847265, | |
| "mfu": 21.66473432226839, | |
| "total_flops": 2.3404908320693158e+19 | |
| }, | |
| { | |
| "step": 17900, | |
| "train_loss": 2.57292353361845, | |
| "val_loss": 2.525147634744644, | |
| "lr": 0.006699308783156554, | |
| "tokens_per_sec": 70208.71739194557, | |
| "mfu": 21.662422332006543, | |
| "total_flops": 2.3462982134202892e+19 | |
| }, | |
| { | |
| "step": 17950, | |
| "train_loss": 2.5538694113492966, | |
| "val_loss": 2.5261887848377227, | |
| "lr": 0.006509235669960918, | |
| "tokens_per_sec": 70191.6308148483, | |
| "mfu": 21.6571503848319, | |
| "total_flops": 2.3521055947712627e+19 | |
| }, | |
| { | |
| "step": 18000, | |
| "train_loss": 2.475142151117325, | |
| "val_loss": 2.5296940565109254, | |
| "lr": 0.006324555320336759, | |
| "tokens_per_sec": 70216.04248525678, | |
| "mfu": 21.66468243973708, | |
| "total_flops": 2.357912976122236e+19 | |
| }, | |
| { | |
| "step": 18050, | |
| "train_loss": 2.4305569678545, | |
| "val_loss": 2.5247761785984038, | |
| "lr": 0.006145114730534894, | |
| "tokens_per_sec": 70213.43430780503, | |
| "mfu": 21.663877704319948, | |
| "total_flops": 2.3637203574732095e+19 | |
| }, | |
| { | |
| "step": 18100, | |
| "train_loss": 2.5882591605186462, | |
| "val_loss": 2.519233852624893, | |
| "lr": 0.005970765237835919, | |
| "tokens_per_sec": 70215.29241962059, | |
| "mfu": 21.664451012085483, | |
| "total_flops": 2.369527738824183e+19 | |
| }, | |
| { | |
| "step": 18150, | |
| "train_loss": 2.579355038702488, | |
| "val_loss": 2.520945113897324, | |
| "lr": 0.005801362397386308, | |
| "tokens_per_sec": 70217.91418806078, | |
| "mfu": 21.665259940909674, | |
| "total_flops": 2.3753351201751564e+19 | |
| }, | |
| { | |
| "step": 18200, | |
| "train_loss": 2.3963413536548615, | |
| "val_loss": 2.521712827682495, | |
| "lr": 0.005636765862528908, | |
| "tokens_per_sec": 70216.60131755349, | |
| "mfu": 21.66485486364217, | |
| "total_flops": 2.38114250152613e+19 | |
| }, | |
| { | |
| "step": 18250, | |
| "train_loss": 2.548183634877205, | |
| "val_loss": 2.524004542827606, | |
| "lr": 0.0054768392685287225, | |
| "tokens_per_sec": 70199.54698313973, | |
| "mfu": 21.659592864728282, | |
| "total_flops": 2.3869498828771033e+19 | |
| }, | |
| { | |
| "step": 18300, | |
| "train_loss": 2.514558032155037, | |
| "val_loss": 2.5144589722156523, | |
| "lr": 0.005321450119597621, | |
| "tokens_per_sec": 70043.88400575175, | |
| "mfu": 21.61156411156346, | |
| "total_flops": 2.3927572642280767e+19 | |
| }, | |
| { | |
| "step": 18350, | |
| "train_loss": 2.5371487140655518, | |
| "val_loss": 2.517710816860199, | |
| "lr": 0.005170469679124381, | |
| "tokens_per_sec": 70222.98205292481, | |
| "mfu": 21.66682359615199, | |
| "total_flops": 2.39856464557905e+19 | |
| }, | |
| { | |
| "step": 18400, | |
| "train_loss": 2.5916066020727158, | |
| "val_loss": 2.5215661764144897, | |
| "lr": 0.005023772863019161, | |
| "tokens_per_sec": 70194.03168204919, | |
| "mfu": 21.657891156080836, | |
| "total_flops": 2.4043720269300236e+19 | |
| }, | |
| { | |
| "step": 18450, | |
| "train_loss": 2.526241347193718, | |
| "val_loss": 2.5169094502925873, | |
| "lr": 0.0048812381360839605, | |
| "tokens_per_sec": 70216.15335418667, | |
| "mfu": 21.664716647591487, | |
| "total_flops": 2.410179408280997e+19 | |
| }, | |
| { | |
| "step": 18500, | |
| "train_loss": 2.487909287214279, | |
| "val_loss": 2.511305093765259, | |
| "lr": 0.004742747411323311, | |
| "tokens_per_sec": 70188.60697085253, | |
| "mfu": 21.656217398329147, | |
| "total_flops": 2.4159867896319705e+19 | |
| }, | |
| { | |
| "step": 18550, | |
| "train_loss": 2.426807537674904, | |
| "val_loss": 2.5122386038303377, | |
| "lr": 0.004608185952111692, | |
| "tokens_per_sec": 70229.14605214057, | |
| "mfu": 21.668725456194814, | |
| "total_flops": 2.421794170982944e+19 | |
| }, | |
| { | |
| "step": 18600, | |
| "train_loss": 2.4519720375537872, | |
| "val_loss": 2.5156042873859406, | |
| "lr": 0.004477442277136679, | |
| "tokens_per_sec": 70199.08175790055, | |
| "mfu": 21.659449322644264, | |
| "total_flops": 2.4276015523339174e+19 | |
| }, | |
| { | |
| "step": 18650, | |
| "train_loss": 2.456761196255684, | |
| "val_loss": 2.5095840215682985, | |
| "lr": 0.004350408068039046, | |
| "tokens_per_sec": 70222.81487638665, | |
| "mfu": 21.666772014967943, | |
| "total_flops": 2.433408933684891e+19 | |
| }, | |
| { | |
| "step": 18700, | |
| "train_loss": 2.487348422408104, | |
| "val_loss": 2.514208322763443, | |
| "lr": 0.004226978079673293, | |
| "tokens_per_sec": 70193.76919298085, | |
| "mfu": 21.65781016686373, | |
| "total_flops": 2.4392163150358643e+19 | |
| }, | |
| { | |
| "step": 18750, | |
| "train_loss": 2.5389831364154816, | |
| "val_loss": 2.5142862021923067, | |
| "lr": 0.004107050052914292, | |
| "tokens_per_sec": 70220.9965045322, | |
| "mfu": 21.66621096869718, | |
| "total_flops": 2.4450236963868377e+19 | |
| }, | |
| { | |
| "step": 18800, | |
| "train_loss": 2.487064689397812, | |
| "val_loss": 2.52103568315506, | |
| "lr": 0.0039905246299377604, | |
| "tokens_per_sec": 70206.9877272651, | |
| "mfu": 21.6618886557311, | |
| "total_flops": 2.450831077737811e+19 | |
| }, | |
| { | |
| "step": 18850, | |
| "train_loss": 2.458393931388855, | |
| "val_loss": 2.5108425080776215, | |
| "lr": 0.0038773052719044145, | |
| "tokens_per_sec": 70196.25076219505, | |
| "mfu": 21.658575838170098, | |
| "total_flops": 2.4566384590887846e+19 | |
| }, | |
| { | |
| "step": 18900, | |
| "train_loss": 2.461804822087288, | |
| "val_loss": 2.5080023407936096, | |
| "lr": 0.003767298178979601, | |
| "tokens_per_sec": 70043.49959678273, | |
| "mfu": 21.61144550479005, | |
| "total_flops": 2.462445840439758e+19 | |
| }, | |
| { | |
| "step": 18950, | |
| "train_loss": 2.4762733578681946, | |
| "val_loss": 2.5128548741340637, | |
| "lr": 0.003660412212622112, | |
| "tokens_per_sec": 70186.46487495826, | |
| "mfu": 21.65555646920722, | |
| "total_flops": 2.4682532217907315e+19 | |
| }, | |
| { | |
| "step": 19000, | |
| "train_loss": 2.492058739066124, | |
| "val_loss": 2.5074475705623627, | |
| "lr": 0.003556558820077846, | |
| "tokens_per_sec": 70216.60468066345, | |
| "mfu": 21.664855901306883, | |
| "total_flops": 2.474060603141705e+19 | |
| }, | |
| { | |
| "step": 19050, | |
| "train_loss": 2.6086084991693497, | |
| "val_loss": 2.509049689769745, | |
| "lr": 0.003455651961015727, | |
| "tokens_per_sec": 70218.33986415529, | |
| "mfu": 21.66539128037404, | |
| "total_flops": 2.4798679844926783e+19 | |
| }, | |
| { | |
| "step": 19100, | |
| "train_loss": 2.4477206766605377, | |
| "val_loss": 2.511517381668091, | |
| "lr": 0.0033576080362451204, | |
| "tokens_per_sec": 70222.40125522732, | |
| "mfu": 21.666644394971854, | |
| "total_flops": 2.4856753658436518e+19 | |
| }, | |
| { | |
| "step": 19150, | |
| "train_loss": 2.6432832330465317, | |
| "val_loss": 2.511579215526581, | |
| "lr": 0.003262345818455677, | |
| "tokens_per_sec": 70208.37241605691, | |
| "mfu": 21.66231589204193, | |
| "total_flops": 2.4914827471946252e+19 | |
| }, | |
| { | |
| "step": 19200, | |
| "train_loss": 2.485611543059349, | |
| "val_loss": 2.5106746435165403, | |
| "lr": 0.003169786384922227, | |
| "tokens_per_sec": 70229.11240904075, | |
| "mfu": 21.668715075845142, | |
| "total_flops": 2.4972901285455987e+19 | |
| }, | |
| { | |
| "step": 19250, | |
| "train_loss": 2.477419838309288, | |
| "val_loss": 2.507012498378754, | |
| "lr": 0.0030798530521189845, | |
| "tokens_per_sec": 70049.52457141867, | |
| "mfu": 21.613304469744147, | |
| "total_flops": 2.503097509896572e+19 | |
| }, | |
| { | |
| "step": 19300, | |
| "train_loss": 2.505843862891197, | |
| "val_loss": 2.4998787224292753, | |
| "lr": 0.0029924713121888677, | |
| "tokens_per_sec": 70222.45843758582, | |
| "mfu": 21.66666203819973, | |
| "total_flops": 2.5089048912475455e+19 | |
| }, | |
| { | |
| "step": 19350, | |
| "train_loss": 2.514888495206833, | |
| "val_loss": 2.5030120730400087, | |
| "lr": 0.0029075687712153235, | |
| "tokens_per_sec": 70215.38904950574, | |
| "mfu": 21.66448082657946, | |
| "total_flops": 2.514712272598519e+19 | |
| }, | |
| { | |
| "step": 19400, | |
| "train_loss": 2.5361353158950806, | |
| "val_loss": 2.502032619714737, | |
| "lr": 0.0028250750892455092, | |
| "tokens_per_sec": 70220.36315738855, | |
| "mfu": 21.666015553742763, | |
| "total_flops": 2.5205196539494924e+19 | |
| }, | |
| { | |
| "step": 19450, | |
| "train_loss": 2.4763139337301254, | |
| "val_loss": 2.5012336313724517, | |
| "lr": 0.002744921922015124, | |
| "tokens_per_sec": 70211.37520972393, | |
| "mfu": 21.663242383608893, | |
| "total_flops": 2.526327035300466e+19 | |
| }, | |
| { | |
| "step": 19500, | |
| "train_loss": 2.384517587721348, | |
| "val_loss": 2.497239035367966, | |
| "lr": 0.002667042864326648, | |
| "tokens_per_sec": 70220.23478517876, | |
| "mfu": 21.66597594537037, | |
| "total_flops": 2.5321344166514393e+19 | |
| }, | |
| { | |
| "step": 19550, | |
| "train_loss": 2.4997721165418625, | |
| "val_loss": 2.4959007024765016, | |
| "lr": 0.0025913733950340394, | |
| "tokens_per_sec": 70220.8249648787, | |
| "mfu": 21.666158041303564, | |
| "total_flops": 2.5379417980024127e+19 | |
| }, | |
| { | |
| "step": 19600, | |
| "train_loss": 2.4810939878225327, | |
| "val_loss": 2.4974619328975676, | |
| "lr": 0.0025178508235883345, | |
| "tokens_per_sec": 70220.9263189256, | |
| "mfu": 21.66618931340553, | |
| "total_flops": 2.543749179353386e+19 | |
| }, | |
| { | |
| "step": 19650, | |
| "train_loss": 2.4052472189068794, | |
| "val_loss": 2.494679808616638, | |
| "lr": 0.0024464142380998632, | |
| "tokens_per_sec": 70213.12111983016, | |
| "mfu": 21.663781072299983, | |
| "total_flops": 2.5495565607043596e+19 | |
| }, | |
| { | |
| "step": 19700, | |
| "train_loss": 2.4900813549757004, | |
| "val_loss": 2.499013751745224, | |
| "lr": 0.0023770044548740367, | |
| "tokens_per_sec": 70217.72730476891, | |
| "mfu": 21.66520227934077, | |
| "total_flops": 2.555363942055333e+19 | |
| }, | |
| { | |
| "step": 19750, | |
| "train_loss": 2.484506592154503, | |
| "val_loss": 2.4978590846061706, | |
| "lr": 0.0023095639693789167, | |
| "tokens_per_sec": 70216.77720863638, | |
| "mfu": 21.664909133639735, | |
| "total_flops": 2.5611713234063065e+19 | |
| }, | |
| { | |
| "step": 19800, | |
| "train_loss": 2.538615956902504, | |
| "val_loss": 2.4995242774486544, | |
| "lr": 0.0022440369086039274, | |
| "tokens_per_sec": 70221.13945488365, | |
| "mfu": 21.66625507505606, | |
| "total_flops": 2.56697870475728e+19 | |
| }, | |
| { | |
| "step": 19550, | |
| "train_loss": 2.369577407836914, | |
| "val_loss": 2.434698057174683, | |
| "lr": 0.0025913733950340394, | |
| "tokens_per_sec": 70228.78181122785, | |
| "mfu": 21.61442648762866, | |
| "total_flops": 2.5727860861082534e+19 | |
| }, | |
| { | |
| "step": 19600, | |
| "train_loss": 2.436405763030052, | |
| "val_loss": 2.435771381855011, | |
| "lr": 0.0025178508235883345, | |
| "tokens_per_sec": 70236.42684073243, | |
| "mfu": 21.616779410802906, | |
| "total_flops": 2.578593467459227e+19 | |
| }, | |
| { | |
| "step": 19650, | |
| "train_loss": 2.4230239391326904, | |
| "val_loss": 2.4384620606899263, | |
| "lr": 0.0024464142380998632, | |
| "tokens_per_sec": 70220.75051917222, | |
| "mfu": 21.61195468380037, | |
| "total_flops": 2.5844008488102003e+19 | |
| }, | |
| { | |
| "step": 19700, | |
| "train_loss": 2.4815607368946075, | |
| "val_loss": 2.44344682097435, | |
| "lr": 0.0023770044548740367, | |
| "tokens_per_sec": 70243.49205167516, | |
| "mfu": 21.61895388227461, | |
| "total_flops": 2.5902082301611737e+19 | |
| }, | |
| { | |
| "step": 19750, | |
| "train_loss": 2.3950591534376144, | |
| "val_loss": 2.4454896986484527, | |
| "lr": 0.0023095639693789167, | |
| "tokens_per_sec": 70258.77042879947, | |
| "mfu": 21.623656133270384, | |
| "total_flops": 2.596015611512147e+19 | |
| }, | |
| { | |
| "step": 19800, | |
| "train_loss": 2.4920027405023575, | |
| "val_loss": 2.448630851507187, | |
| "lr": 0.0022440369086039274, | |
| "tokens_per_sec": 70099.8157121793, | |
| "mfu": 21.57473438140968, | |
| "total_flops": 2.6018229928631206e+19 | |
| }, | |
| { | |
| "step": 19850, | |
| "train_loss": 2.3662567362189293, | |
| "val_loss": 2.4463677644729613, | |
| "lr": 0.0021803689847702556, | |
| "tokens_per_sec": 70249.89910754852, | |
| "mfu": 21.620925792289356, | |
| "total_flops": 2.607630374214094e+19 | |
| }, | |
| { | |
| "step": 19900, | |
| "train_loss": 2.584068328142166, | |
| "val_loss": 2.4525881469249726, | |
| "lr": 0.0021185074503545782, | |
| "tokens_per_sec": 70253.34300150532, | |
| "mfu": 21.621985725137975, | |
| "total_flops": 2.6134377555650675e+19 | |
| }, | |
| { | |
| "step": 19950, | |
| "train_loss": 2.47366926074028, | |
| "val_loss": 2.4520198345184325, | |
| "lr": 0.0020584010543888563, | |
| "tokens_per_sec": 70260.01483742267, | |
| "mfu": 21.624039126937838, | |
| "total_flops": 2.619245136916041e+19 | |
| }, | |
| { | |
| "step": 19999, | |
| "train_loss": 2.5081333369016647, | |
| "val_loss": 2.45241938829422, | |
| "lr": 0.0020011516239787216, | |
| "tokens_per_sec": 70258.21878164218, | |
| "mfu": 21.62348635192688, | |
| "total_flops": 2.624936370639995e+19 | |
| } | |
| ] |