TAMA-vA / trainer_state.json
Naihao Deng
Added model files
59d6370
{
"best_metric": null,
"best_model_checkpoint": null,
"epoch": 2.9897610921501707,
"eval_steps": 146,
"global_step": 438,
"is_hyper_param_search": false,
"is_local_process_zero": true,
"is_world_process_zero": true,
"log_history": [
{
"epoch": 0.06825938566552901,
"grad_norm": 121.90807294112876,
"learning_rate": 9.885844748858448e-07,
"loss": 2.2874,
"step": 10
},
{
"epoch": 0.13651877133105803,
"grad_norm": 50.17053366859769,
"learning_rate": 9.771689497716894e-07,
"loss": 0.9472,
"step": 20
},
{
"epoch": 0.20477815699658702,
"grad_norm": 26.596625414659396,
"learning_rate": 9.657534246575343e-07,
"loss": 0.9087,
"step": 30
},
{
"epoch": 0.27303754266211605,
"grad_norm": 66.69994801604244,
"learning_rate": 9.54337899543379e-07,
"loss": 0.8543,
"step": 40
},
{
"epoch": 0.3412969283276451,
"grad_norm": 48.76266229603266,
"learning_rate": 9.429223744292237e-07,
"loss": 0.6994,
"step": 50
},
{
"epoch": 0.40955631399317405,
"grad_norm": 52.043092597792516,
"learning_rate": 9.315068493150684e-07,
"loss": 0.7693,
"step": 60
},
{
"epoch": 0.4778156996587031,
"grad_norm": 98.55471713268969,
"learning_rate": 9.200913242009132e-07,
"loss": 0.7138,
"step": 70
},
{
"epoch": 0.5460750853242321,
"grad_norm": 78.48196775888992,
"learning_rate": 9.08675799086758e-07,
"loss": 0.7903,
"step": 80
},
{
"epoch": 0.6143344709897611,
"grad_norm": 84.85124588773758,
"learning_rate": 8.972602739726027e-07,
"loss": 0.7586,
"step": 90
},
{
"epoch": 0.6825938566552902,
"grad_norm": 58.99018773695002,
"learning_rate": 8.858447488584474e-07,
"loss": 0.663,
"step": 100
},
{
"epoch": 0.7508532423208191,
"grad_norm": 39.94579058556075,
"learning_rate": 8.744292237442922e-07,
"loss": 0.589,
"step": 110
},
{
"epoch": 0.8191126279863481,
"grad_norm": 58.372841216401504,
"learning_rate": 8.63013698630137e-07,
"loss": 0.699,
"step": 120
},
{
"epoch": 0.8873720136518771,
"grad_norm": 43.360346848406074,
"learning_rate": 8.515981735159817e-07,
"loss": 0.6512,
"step": 130
},
{
"epoch": 0.9556313993174061,
"grad_norm": 61.008846083100586,
"learning_rate": 8.401826484018264e-07,
"loss": 0.7701,
"step": 140
},
{
"epoch": 0.9965870307167235,
"eval_loss": 0.619476318359375,
"eval_runtime": 10.3711,
"eval_samples_per_second": 25.07,
"eval_steps_per_second": 3.182,
"step": 146
},
{
"epoch": 1.023890784982935,
"grad_norm": 36.97894806080412,
"learning_rate": 8.287671232876712e-07,
"loss": 0.5175,
"step": 150
},
{
"epoch": 1.0921501706484642,
"grad_norm": 18.0897726845655,
"learning_rate": 8.173515981735159e-07,
"loss": 0.3988,
"step": 160
},
{
"epoch": 1.1604095563139931,
"grad_norm": 53.788209383918755,
"learning_rate": 8.059360730593608e-07,
"loss": 0.3701,
"step": 170
},
{
"epoch": 1.2286689419795223,
"grad_norm": 34.202912199875364,
"learning_rate": 7.945205479452054e-07,
"loss": 0.4629,
"step": 180
},
{
"epoch": 1.2969283276450512,
"grad_norm": 32.334790445469515,
"learning_rate": 7.831050228310501e-07,
"loss": 0.4946,
"step": 190
},
{
"epoch": 1.36518771331058,
"grad_norm": 34.57788598902016,
"learning_rate": 7.71689497716895e-07,
"loss": 0.5149,
"step": 200
},
{
"epoch": 1.4334470989761092,
"grad_norm": 40.59898497248684,
"learning_rate": 7.602739726027397e-07,
"loss": 0.4565,
"step": 210
},
{
"epoch": 1.5017064846416384,
"grad_norm": 28.000076288707174,
"learning_rate": 7.488584474885844e-07,
"loss": 0.4383,
"step": 220
},
{
"epoch": 1.5699658703071673,
"grad_norm": 28.11781200722797,
"learning_rate": 7.374429223744292e-07,
"loss": 0.3446,
"step": 230
},
{
"epoch": 1.6382252559726962,
"grad_norm": 21.457173259928325,
"learning_rate": 7.260273972602739e-07,
"loss": 0.4206,
"step": 240
},
{
"epoch": 1.7064846416382253,
"grad_norm": 11.954122609487266,
"learning_rate": 7.146118721461188e-07,
"loss": 0.4117,
"step": 250
},
{
"epoch": 1.7747440273037542,
"grad_norm": 25.454890078823407,
"learning_rate": 7.031963470319634e-07,
"loss": 0.3557,
"step": 260
},
{
"epoch": 1.8430034129692832,
"grad_norm": 59.00305506329269,
"learning_rate": 6.917808219178081e-07,
"loss": 0.4312,
"step": 270
},
{
"epoch": 1.9112627986348123,
"grad_norm": 19.693798816454596,
"learning_rate": 6.80365296803653e-07,
"loss": 0.3798,
"step": 280
},
{
"epoch": 1.9795221843003414,
"grad_norm": 31.09835580820732,
"learning_rate": 6.689497716894977e-07,
"loss": 0.4721,
"step": 290
},
{
"epoch": 1.993174061433447,
"eval_loss": 0.6094198226928711,
"eval_runtime": 10.038,
"eval_samples_per_second": 25.901,
"eval_steps_per_second": 3.287,
"step": 292
},
{
"epoch": 2.04778156996587,
"grad_norm": 14.148404582312557,
"learning_rate": 6.575342465753423e-07,
"loss": 0.3175,
"step": 300
},
{
"epoch": 2.1160409556313993,
"grad_norm": 9.302005033170696,
"learning_rate": 6.461187214611872e-07,
"loss": 0.3385,
"step": 310
},
{
"epoch": 2.1843003412969284,
"grad_norm": 23.055786559150018,
"learning_rate": 6.347031963470319e-07,
"loss": 0.2497,
"step": 320
},
{
"epoch": 2.252559726962457,
"grad_norm": 15.965787246539906,
"learning_rate": 6.232876712328768e-07,
"loss": 0.2529,
"step": 330
},
{
"epoch": 2.3208191126279862,
"grad_norm": 19.137778271986814,
"learning_rate": 6.118721461187214e-07,
"loss": 0.2681,
"step": 340
},
{
"epoch": 2.3890784982935154,
"grad_norm": 43.528989020233965,
"learning_rate": 6.004566210045661e-07,
"loss": 0.1969,
"step": 350
},
{
"epoch": 2.4573378839590445,
"grad_norm": 38.488063777446804,
"learning_rate": 5.89041095890411e-07,
"loss": 0.244,
"step": 360
},
{
"epoch": 2.5255972696245736,
"grad_norm": 19.9306913959441,
"learning_rate": 5.776255707762557e-07,
"loss": 0.3198,
"step": 370
},
{
"epoch": 2.5938566552901023,
"grad_norm": 19.0407162824872,
"learning_rate": 5.662100456621004e-07,
"loss": 0.2783,
"step": 380
},
{
"epoch": 2.6621160409556315,
"grad_norm": 16.817106035170273,
"learning_rate": 5.547945205479452e-07,
"loss": 0.2687,
"step": 390
},
{
"epoch": 2.73037542662116,
"grad_norm": 27.440144369018864,
"learning_rate": 5.433789954337899e-07,
"loss": 0.2999,
"step": 400
},
{
"epoch": 2.7986348122866893,
"grad_norm": 26.842832940417683,
"learning_rate": 5.319634703196348e-07,
"loss": 0.3246,
"step": 410
},
{
"epoch": 2.8668941979522184,
"grad_norm": 52.24386186460223,
"learning_rate": 5.205479452054794e-07,
"loss": 0.2099,
"step": 420
},
{
"epoch": 2.9351535836177476,
"grad_norm": 39.20725213578627,
"learning_rate": 5.091324200913241e-07,
"loss": 0.3482,
"step": 430
},
{
"epoch": 2.9897610921501707,
"eval_loss": 0.6506538391113281,
"eval_runtime": 10.0544,
"eval_samples_per_second": 25.859,
"eval_steps_per_second": 3.282,
"step": 438
}
],
"logging_steps": 10,
"max_steps": 876,
"num_input_tokens_seen": 0,
"num_train_epochs": 6,
"save_steps": 146,
"stateful_callbacks": {
"TrainerControl": {
"args": {
"should_epoch_stop": false,
"should_evaluate": false,
"should_log": false,
"should_save": true,
"should_training_stop": false
},
"attributes": {}
}
},
"total_flos": 6392788746240.0,
"train_batch_size": 1,
"trial_name": null,
"trial_params": null
}