TAMA-vB / trainer_state.json
Naihao Deng
Added model files
852da28
{
"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": 180.72653528520956,
"learning_rate": 4.942922374429224e-07,
"loss": 3.2191,
"step": 10
},
{
"epoch": 0.13651877133105803,
"grad_norm": 67.78875632967043,
"learning_rate": 4.885844748858447e-07,
"loss": 1.3006,
"step": 20
},
{
"epoch": 0.20477815699658702,
"grad_norm": 58.498874329827956,
"learning_rate": 4.828767123287671e-07,
"loss": 1.0222,
"step": 30
},
{
"epoch": 0.27303754266211605,
"grad_norm": 65.81012271745634,
"learning_rate": 4.771689497716894e-07,
"loss": 1.0702,
"step": 40
},
{
"epoch": 0.3412969283276451,
"grad_norm": 67.6868122468499,
"learning_rate": 4.7146118721461187e-07,
"loss": 0.8233,
"step": 50
},
{
"epoch": 0.40955631399317405,
"grad_norm": 130.6895452398653,
"learning_rate": 4.657534246575342e-07,
"loss": 1.0114,
"step": 60
},
{
"epoch": 0.4778156996587031,
"grad_norm": 67.10774835755684,
"learning_rate": 4.600456621004566e-07,
"loss": 0.7751,
"step": 70
},
{
"epoch": 0.5460750853242321,
"grad_norm": 222.42047329326795,
"learning_rate": 4.54337899543379e-07,
"loss": 0.9296,
"step": 80
},
{
"epoch": 0.6143344709897611,
"grad_norm": 53.90173100677448,
"learning_rate": 4.4863013698630134e-07,
"loss": 0.8508,
"step": 90
},
{
"epoch": 0.6825938566552902,
"grad_norm": 73.87543764538118,
"learning_rate": 4.429223744292237e-07,
"loss": 0.7461,
"step": 100
},
{
"epoch": 0.7508532423208191,
"grad_norm": 51.730677807091766,
"learning_rate": 4.372146118721461e-07,
"loss": 0.6538,
"step": 110
},
{
"epoch": 0.8191126279863481,
"grad_norm": 58.532049779965085,
"learning_rate": 4.315068493150685e-07,
"loss": 0.7099,
"step": 120
},
{
"epoch": 0.8873720136518771,
"grad_norm": 69.81205859668461,
"learning_rate": 4.2579908675799087e-07,
"loss": 0.7059,
"step": 130
},
{
"epoch": 0.9556313993174061,
"grad_norm": 48.161843074686466,
"learning_rate": 4.200913242009132e-07,
"loss": 0.8547,
"step": 140
},
{
"epoch": 0.9965870307167235,
"eval_loss": 0.6474742889404297,
"eval_runtime": 10.4556,
"eval_samples_per_second": 24.867,
"eval_steps_per_second": 3.156,
"step": 146
},
{
"epoch": 1.023890784982935,
"grad_norm": 27.25525751270437,
"learning_rate": 4.143835616438356e-07,
"loss": 0.6076,
"step": 150
},
{
"epoch": 1.0921501706484642,
"grad_norm": 25.863617719976116,
"learning_rate": 4.0867579908675797e-07,
"loss": 0.5054,
"step": 160
},
{
"epoch": 1.1604095563139931,
"grad_norm": 52.70780610444248,
"learning_rate": 4.029680365296804e-07,
"loss": 0.4987,
"step": 170
},
{
"epoch": 1.2286689419795223,
"grad_norm": 25.718837256960583,
"learning_rate": 3.972602739726027e-07,
"loss": 0.5697,
"step": 180
},
{
"epoch": 1.2969283276450512,
"grad_norm": 36.98618805648008,
"learning_rate": 3.915525114155251e-07,
"loss": 0.615,
"step": 190
},
{
"epoch": 1.36518771331058,
"grad_norm": 32.58402570740449,
"learning_rate": 3.858447488584475e-07,
"loss": 0.5849,
"step": 200
},
{
"epoch": 1.4334470989761092,
"grad_norm": 23.37886970180941,
"learning_rate": 3.8013698630136986e-07,
"loss": 0.5924,
"step": 210
},
{
"epoch": 1.5017064846416384,
"grad_norm": 27.73682160230221,
"learning_rate": 3.744292237442922e-07,
"loss": 0.5095,
"step": 220
},
{
"epoch": 1.5699658703071673,
"grad_norm": 58.838640484256175,
"learning_rate": 3.687214611872146e-07,
"loss": 0.4661,
"step": 230
},
{
"epoch": 1.6382252559726962,
"grad_norm": 29.677456691329837,
"learning_rate": 3.6301369863013697e-07,
"loss": 0.5448,
"step": 240
},
{
"epoch": 1.7064846416382253,
"grad_norm": 83.73694041166883,
"learning_rate": 3.573059360730594e-07,
"loss": 0.4715,
"step": 250
},
{
"epoch": 1.7747440273037542,
"grad_norm": 72.88041601319203,
"learning_rate": 3.515981735159817e-07,
"loss": 0.4408,
"step": 260
},
{
"epoch": 1.8430034129692832,
"grad_norm": 48.86805911410425,
"learning_rate": 3.4589041095890407e-07,
"loss": 0.5299,
"step": 270
},
{
"epoch": 1.9112627986348123,
"grad_norm": 56.39227585073634,
"learning_rate": 3.401826484018265e-07,
"loss": 0.462,
"step": 280
},
{
"epoch": 1.9795221843003414,
"grad_norm": 47.10405902113264,
"learning_rate": 3.3447488584474886e-07,
"loss": 0.5135,
"step": 290
},
{
"epoch": 1.993174061433447,
"eval_loss": 0.6217488646507263,
"eval_runtime": 10.0993,
"eval_samples_per_second": 25.744,
"eval_steps_per_second": 3.268,
"step": 292
},
{
"epoch": 2.04778156996587,
"grad_norm": 47.53720316702887,
"learning_rate": 3.287671232876712e-07,
"loss": 0.4304,
"step": 300
},
{
"epoch": 2.1160409556313993,
"grad_norm": 29.27611194113683,
"learning_rate": 3.230593607305936e-07,
"loss": 0.4024,
"step": 310
},
{
"epoch": 2.1843003412969284,
"grad_norm": 22.633855883872762,
"learning_rate": 3.1735159817351596e-07,
"loss": 0.3316,
"step": 320
},
{
"epoch": 2.252559726962457,
"grad_norm": 33.60546914801217,
"learning_rate": 3.116438356164384e-07,
"loss": 0.3474,
"step": 330
},
{
"epoch": 2.3208191126279862,
"grad_norm": 38.727812421442835,
"learning_rate": 3.059360730593607e-07,
"loss": 0.3546,
"step": 340
},
{
"epoch": 2.3890784982935154,
"grad_norm": 41.67978939810039,
"learning_rate": 3.0022831050228307e-07,
"loss": 0.2542,
"step": 350
},
{
"epoch": 2.4573378839590445,
"grad_norm": 77.57463923271911,
"learning_rate": 2.945205479452055e-07,
"loss": 0.2766,
"step": 360
},
{
"epoch": 2.5255972696245736,
"grad_norm": 59.43311669158369,
"learning_rate": 2.8881278538812786e-07,
"loss": 0.44,
"step": 370
},
{
"epoch": 2.5938566552901023,
"grad_norm": 22.434362950673165,
"learning_rate": 2.831050228310502e-07,
"loss": 0.2935,
"step": 380
},
{
"epoch": 2.6621160409556315,
"grad_norm": 30.626819579935223,
"learning_rate": 2.773972602739726e-07,
"loss": 0.3208,
"step": 390
},
{
"epoch": 2.73037542662116,
"grad_norm": 64.9106394772347,
"learning_rate": 2.7168949771689496e-07,
"loss": 0.41,
"step": 400
},
{
"epoch": 2.7986348122866893,
"grad_norm": 112.81858040393827,
"learning_rate": 2.659817351598174e-07,
"loss": 0.4234,
"step": 410
},
{
"epoch": 2.8668941979522184,
"grad_norm": 25.381208288443705,
"learning_rate": 2.602739726027397e-07,
"loss": 0.3328,
"step": 420
},
{
"epoch": 2.9351535836177476,
"grad_norm": 45.82561826818154,
"learning_rate": 2.5456621004566206e-07,
"loss": 0.4019,
"step": 430
},
{
"epoch": 2.9897610921501707,
"eval_loss": 0.64452064037323,
"eval_runtime": 10.0372,
"eval_samples_per_second": 25.904,
"eval_steps_per_second": 3.288,
"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
}