SALAMA_NEW88 / checkpoint-300 /trainer_state.json
EYEDOL's picture
Upload folder using huggingface_hub
e1e5560 verified
{
"best_global_step": 300,
"best_metric": 0.5966701953613704,
"best_model_checkpoint": "./SALAMA_NEW88/checkpoint-300",
"epoch": 0.9787928221859706,
"eval_steps": 300,
"global_step": 300,
"is_hyper_param_search": false,
"is_local_process_zero": true,
"is_world_process_zero": true,
"log_history": [
{
"epoch": 0.03262642740619902,
"grad_norm": 2.3551647663116455,
"learning_rate": 1.8e-07,
"loss": 0.0153,
"step": 10
},
{
"epoch": 0.06525285481239804,
"grad_norm": 1.2086862325668335,
"learning_rate": 3.8e-07,
"loss": 0.012,
"step": 20
},
{
"epoch": 0.09787928221859707,
"grad_norm": 2.0345568656921387,
"learning_rate": 5.800000000000001e-07,
"loss": 0.0145,
"step": 30
},
{
"epoch": 0.13050570962479607,
"grad_norm": 1.01999032497406,
"learning_rate": 7.8e-07,
"loss": 0.0091,
"step": 40
},
{
"epoch": 0.1631321370309951,
"grad_norm": 0.9040848612785339,
"learning_rate": 9.800000000000001e-07,
"loss": 0.0112,
"step": 50
},
{
"epoch": 0.19575856443719414,
"grad_norm": 1.4125545024871826,
"learning_rate": 1.1800000000000001e-06,
"loss": 0.0064,
"step": 60
},
{
"epoch": 0.22838499184339314,
"grad_norm": 0.22181180119514465,
"learning_rate": 1.3800000000000001e-06,
"loss": 0.0089,
"step": 70
},
{
"epoch": 0.26101141924959215,
"grad_norm": 0.6269964575767517,
"learning_rate": 1.5800000000000001e-06,
"loss": 0.0069,
"step": 80
},
{
"epoch": 0.2936378466557912,
"grad_norm": 0.1682853251695633,
"learning_rate": 1.7800000000000001e-06,
"loss": 0.0077,
"step": 90
},
{
"epoch": 0.3262642740619902,
"grad_norm": 0.14965638518333435,
"learning_rate": 1.98e-06,
"loss": 0.0026,
"step": 100
},
{
"epoch": 0.35889070146818924,
"grad_norm": 0.34541550278663635,
"learning_rate": 2.1800000000000003e-06,
"loss": 0.0061,
"step": 110
},
{
"epoch": 0.3915171288743883,
"grad_norm": 0.45224177837371826,
"learning_rate": 2.38e-06,
"loss": 0.0034,
"step": 120
},
{
"epoch": 0.42414355628058725,
"grad_norm": 0.19250553846359253,
"learning_rate": 2.5800000000000003e-06,
"loss": 0.0038,
"step": 130
},
{
"epoch": 0.4567699836867863,
"grad_norm": 1.2514749765396118,
"learning_rate": 2.7800000000000005e-06,
"loss": 0.0058,
"step": 140
},
{
"epoch": 0.4893964110929853,
"grad_norm": 0.26269763708114624,
"learning_rate": 2.9800000000000003e-06,
"loss": 0.0046,
"step": 150
},
{
"epoch": 0.5220228384991843,
"grad_norm": 0.5023095607757568,
"learning_rate": 3.1800000000000005e-06,
"loss": 0.008,
"step": 160
},
{
"epoch": 0.5546492659053833,
"grad_norm": 0.5606697797775269,
"learning_rate": 3.3800000000000007e-06,
"loss": 0.0054,
"step": 170
},
{
"epoch": 0.5872756933115824,
"grad_norm": 0.11588776856660843,
"learning_rate": 3.58e-06,
"loss": 0.0039,
"step": 180
},
{
"epoch": 0.6199021207177814,
"grad_norm": 0.4340250492095947,
"learning_rate": 3.7800000000000002e-06,
"loss": 0.0053,
"step": 190
},
{
"epoch": 0.6525285481239804,
"grad_norm": 0.1779269576072693,
"learning_rate": 3.980000000000001e-06,
"loss": 0.0046,
"step": 200
},
{
"epoch": 0.6851549755301795,
"grad_norm": 0.42044180631637573,
"learning_rate": 4.18e-06,
"loss": 0.0076,
"step": 210
},
{
"epoch": 0.7177814029363785,
"grad_norm": 0.2029053121805191,
"learning_rate": 4.38e-06,
"loss": 0.005,
"step": 220
},
{
"epoch": 0.7504078303425775,
"grad_norm": 0.12045332789421082,
"learning_rate": 4.58e-06,
"loss": 0.0087,
"step": 230
},
{
"epoch": 0.7830342577487766,
"grad_norm": 0.11535464227199554,
"learning_rate": 4.78e-06,
"loss": 0.0036,
"step": 240
},
{
"epoch": 0.8156606851549756,
"grad_norm": 0.21480712294578552,
"learning_rate": 4.980000000000001e-06,
"loss": 0.006,
"step": 250
},
{
"epoch": 0.8482871125611745,
"grad_norm": 0.23418085277080536,
"learning_rate": 5.18e-06,
"loss": 0.0111,
"step": 260
},
{
"epoch": 0.8809135399673735,
"grad_norm": 0.8250060081481934,
"learning_rate": 5.380000000000001e-06,
"loss": 0.005,
"step": 270
},
{
"epoch": 0.9135399673735726,
"grad_norm": 0.6729530692100525,
"learning_rate": 5.580000000000001e-06,
"loss": 0.0042,
"step": 280
},
{
"epoch": 0.9461663947797716,
"grad_norm": 2.4627511501312256,
"learning_rate": 5.78e-06,
"loss": 0.0147,
"step": 290
},
{
"epoch": 0.9787928221859706,
"grad_norm": 2.2341601848602295,
"learning_rate": 5.98e-06,
"loss": 0.0153,
"step": 300
},
{
"epoch": 0.9787928221859706,
"eval_loss": 0.0076320902444422245,
"eval_runtime": 875.0468,
"eval_samples_per_second": 2.801,
"eval_steps_per_second": 0.351,
"eval_wer": 0.5966701953613704,
"step": 300
}
],
"logging_steps": 10,
"max_steps": 921,
"num_input_tokens_seen": 0,
"num_train_epochs": 3,
"save_steps": 300,
"stateful_callbacks": {
"TrainerControl": {
"args": {
"should_epoch_stop": false,
"should_evaluate": false,
"should_log": false,
"should_save": true,
"should_training_stop": false
},
"attributes": {}
}
},
"total_flos": 2.770419843072e+18,
"train_batch_size": 16,
"trial_name": null,
"trial_params": null
}