chuyish's picture
Model save
db29329 verified
{
"best_metric": null,
"best_model_checkpoint": null,
"epoch": 0.9974424552429667,
"eval_steps": 500,
"global_step": 195,
"is_hyper_param_search": false,
"is_local_process_zero": true,
"is_world_process_zero": true,
"log_history": [
{
"epoch": 0.02557544757033248,
"grad_norm": 142.70468406748688,
"learning_rate": 5e-06,
"loss": 3.5042,
"step": 5
},
{
"epoch": 0.05115089514066496,
"grad_norm": 39.35096392339812,
"learning_rate": 1e-05,
"loss": 1.8402,
"step": 10
},
{
"epoch": 0.07672634271099744,
"grad_norm": 2.7883188777304038,
"learning_rate": 1.5000000000000002e-05,
"loss": 0.8528,
"step": 15
},
{
"epoch": 0.10230179028132992,
"grad_norm": 1.9363912255406037,
"learning_rate": 2e-05,
"loss": 0.5077,
"step": 20
},
{
"epoch": 0.1278772378516624,
"grad_norm": 2.479337233395153,
"learning_rate": 1.9959742939952393e-05,
"loss": 0.4772,
"step": 25
},
{
"epoch": 0.1534526854219949,
"grad_norm": 2.437727400797437,
"learning_rate": 1.98392958859863e-05,
"loss": 0.4575,
"step": 30
},
{
"epoch": 0.17902813299232737,
"grad_norm": 2.064035897127794,
"learning_rate": 1.9639628606958535e-05,
"loss": 0.4183,
"step": 35
},
{
"epoch": 0.20460358056265984,
"grad_norm": 2.0573078173417567,
"learning_rate": 1.9362348706397374e-05,
"loss": 0.3801,
"step": 40
},
{
"epoch": 0.23017902813299232,
"grad_norm": 1.3860822147023304,
"learning_rate": 1.900968867902419e-05,
"loss": 0.3071,
"step": 45
},
{
"epoch": 0.2557544757033248,
"grad_norm": 1.1223824377196536,
"learning_rate": 1.8584487936018663e-05,
"loss": 0.2866,
"step": 50
},
{
"epoch": 0.2813299232736573,
"grad_norm": 1.3072552349963746,
"learning_rate": 1.8090169943749477e-05,
"loss": 0.3053,
"step": 55
},
{
"epoch": 0.3069053708439898,
"grad_norm": 1.1007498150237207,
"learning_rate": 1.7530714660036112e-05,
"loss": 0.2824,
"step": 60
},
{
"epoch": 0.33248081841432225,
"grad_norm": 1.0572125214610735,
"learning_rate": 1.691062648986865e-05,
"loss": 0.2796,
"step": 65
},
{
"epoch": 0.35805626598465473,
"grad_norm": 1.0088193429142867,
"learning_rate": 1.6234898018587336e-05,
"loss": 0.2592,
"step": 70
},
{
"epoch": 0.3836317135549872,
"grad_norm": 0.9536544858527741,
"learning_rate": 1.5508969814521026e-05,
"loss": 0.2729,
"step": 75
},
{
"epoch": 0.4092071611253197,
"grad_norm": 1.0177774752547195,
"learning_rate": 1.4738686624729987e-05,
"loss": 0.2578,
"step": 80
},
{
"epoch": 0.43478260869565216,
"grad_norm": 0.9024655077486329,
"learning_rate": 1.3930250316539237e-05,
"loss": 0.2537,
"step": 85
},
{
"epoch": 0.46035805626598464,
"grad_norm": 0.9603153657835733,
"learning_rate": 1.3090169943749475e-05,
"loss": 0.2658,
"step": 90
},
{
"epoch": 0.4859335038363171,
"grad_norm": 0.9976253438540076,
"learning_rate": 1.2225209339563144e-05,
"loss": 0.2961,
"step": 95
},
{
"epoch": 0.5115089514066496,
"grad_norm": 0.8607729610780752,
"learning_rate": 1.1342332658176556e-05,
"loss": 0.2087,
"step": 100
},
{
"epoch": 0.5370843989769821,
"grad_norm": 0.8346737121409833,
"learning_rate": 1.044864830350515e-05,
"loss": 0.2335,
"step": 105
},
{
"epoch": 0.5626598465473146,
"grad_norm": 0.9276457383456489,
"learning_rate": 9.551351696494854e-06,
"loss": 0.2321,
"step": 110
},
{
"epoch": 0.5882352941176471,
"grad_norm": 0.8201591451858892,
"learning_rate": 8.657667341823449e-06,
"loss": 0.1986,
"step": 115
},
{
"epoch": 0.6138107416879796,
"grad_norm": 0.9324949471856788,
"learning_rate": 7.774790660436857e-06,
"loss": 0.2346,
"step": 120
},
{
"epoch": 0.639386189258312,
"grad_norm": 0.8224133465787747,
"learning_rate": 6.909830056250527e-06,
"loss": 0.2273,
"step": 125
},
{
"epoch": 0.6649616368286445,
"grad_norm": 0.7394741748745267,
"learning_rate": 6.069749683460765e-06,
"loss": 0.2032,
"step": 130
},
{
"epoch": 0.690537084398977,
"grad_norm": 0.8165965679312597,
"learning_rate": 5.2613133752700145e-06,
"loss": 0.219,
"step": 135
},
{
"epoch": 0.7161125319693095,
"grad_norm": 1.117117835940415,
"learning_rate": 4.491030185478976e-06,
"loss": 0.2337,
"step": 140
},
{
"epoch": 0.7416879795396419,
"grad_norm": 0.8321965045011145,
"learning_rate": 3.7651019814126656e-06,
"loss": 0.2199,
"step": 145
},
{
"epoch": 0.7672634271099744,
"grad_norm": 0.9442319309396315,
"learning_rate": 3.089373510131354e-06,
"loss": 0.1942,
"step": 150
},
{
"epoch": 0.7928388746803069,
"grad_norm": 0.8339668845503085,
"learning_rate": 2.469285339963892e-06,
"loss": 0.2021,
"step": 155
},
{
"epoch": 0.8184143222506394,
"grad_norm": 0.8898393855240705,
"learning_rate": 1.9098300562505266e-06,
"loss": 0.2428,
"step": 160
},
{
"epoch": 0.8439897698209718,
"grad_norm": 0.8213143519432727,
"learning_rate": 1.4155120639813392e-06,
"loss": 0.1893,
"step": 165
},
{
"epoch": 0.8695652173913043,
"grad_norm": 0.9728880507025741,
"learning_rate": 9.903113209758098e-07,
"loss": 0.1854,
"step": 170
},
{
"epoch": 0.8951406649616368,
"grad_norm": 0.7472973685735516,
"learning_rate": 6.37651293602628e-07,
"loss": 0.2287,
"step": 175
},
{
"epoch": 0.9207161125319693,
"grad_norm": 0.818513207281356,
"learning_rate": 3.603713930414676e-07,
"loss": 0.1873,
"step": 180
},
{
"epoch": 0.9462915601023018,
"grad_norm": 0.809589496670413,
"learning_rate": 1.6070411401370335e-07,
"loss": 0.1982,
"step": 185
},
{
"epoch": 0.9718670076726342,
"grad_norm": 0.759402066262018,
"learning_rate": 4.025706004760932e-08,
"loss": 0.2064,
"step": 190
},
{
"epoch": 0.9974424552429667,
"grad_norm": 0.8249592638910015,
"learning_rate": 0.0,
"loss": 0.1977,
"step": 195
},
{
"epoch": 0.9974424552429667,
"step": 195,
"total_flos": 77502516822016.0,
"train_loss": 0.4037646987499335,
"train_runtime": 4195.4221,
"train_samples_per_second": 1.491,
"train_steps_per_second": 0.046
}
],
"logging_steps": 5,
"max_steps": 195,
"num_input_tokens_seen": 0,
"num_train_epochs": 1,
"save_steps": 100,
"stateful_callbacks": {
"TrainerControl": {
"args": {
"should_epoch_stop": false,
"should_evaluate": false,
"should_log": false,
"should_save": true,
"should_training_stop": true
},
"attributes": {}
}
},
"total_flos": 77502516822016.0,
"train_batch_size": 2,
"trial_name": null,
"trial_params": null
}