Sky-T1-32B-Preview-5k-1-epoch / trainer_state.json
ryanmarten's picture
Upload model
0e94c49 verified
{
"best_metric": null,
"best_model_checkpoint": null,
"epoch": 0.9984,
"eval_steps": 500,
"global_step": 52,
"is_hyper_param_search": false,
"is_local_process_zero": true,
"is_world_process_zero": true,
"log_history": [
{
"epoch": 0.0192,
"grad_norm": 2.5694146156311035,
"learning_rate": 1.6666666666666667e-06,
"loss": 0.5167,
"step": 1
},
{
"epoch": 0.0384,
"grad_norm": 2.4537222385406494,
"learning_rate": 3.3333333333333333e-06,
"loss": 0.5203,
"step": 2
},
{
"epoch": 0.0576,
"grad_norm": 2.321495532989502,
"learning_rate": 5e-06,
"loss": 0.4748,
"step": 3
},
{
"epoch": 0.0768,
"grad_norm": 1.8380646705627441,
"learning_rate": 6.666666666666667e-06,
"loss": 0.5253,
"step": 4
},
{
"epoch": 0.096,
"grad_norm": 1.4723880290985107,
"learning_rate": 8.333333333333334e-06,
"loss": 0.4606,
"step": 5
},
{
"epoch": 0.1152,
"grad_norm": 1.1131113767623901,
"learning_rate": 1e-05,
"loss": 0.4019,
"step": 6
},
{
"epoch": 0.1344,
"grad_norm": 0.9251970648765564,
"learning_rate": 9.988343845952697e-06,
"loss": 0.4271,
"step": 7
},
{
"epoch": 0.1536,
"grad_norm": 1.1454741954803467,
"learning_rate": 9.953429730181653e-06,
"loss": 0.3827,
"step": 8
},
{
"epoch": 0.1728,
"grad_norm": 0.9471750259399414,
"learning_rate": 9.895420438411616e-06,
"loss": 0.3547,
"step": 9
},
{
"epoch": 0.192,
"grad_norm": 0.7063102126121521,
"learning_rate": 9.814586436738998e-06,
"loss": 0.3375,
"step": 10
},
{
"epoch": 0.2112,
"grad_norm": 0.6931116580963135,
"learning_rate": 9.711304610594104e-06,
"loss": 0.3156,
"step": 11
},
{
"epoch": 0.2304,
"grad_norm": 0.7782188653945923,
"learning_rate": 9.586056507527266e-06,
"loss": 0.3111,
"step": 12
},
{
"epoch": 0.2496,
"grad_norm": 0.7079610228538513,
"learning_rate": 9.439426092011877e-06,
"loss": 0.2994,
"step": 13
},
{
"epoch": 0.2688,
"grad_norm": 0.6569077968597412,
"learning_rate": 9.272097022732444e-06,
"loss": 0.2973,
"step": 14
},
{
"epoch": 0.288,
"grad_norm": 0.680906355381012,
"learning_rate": 9.08484946505221e-06,
"loss": 0.2882,
"step": 15
},
{
"epoch": 0.3072,
"grad_norm": 0.5492838025093079,
"learning_rate": 8.8785564535221e-06,
"loss": 0.2531,
"step": 16
},
{
"epoch": 0.3264,
"grad_norm": 0.6430338025093079,
"learning_rate": 8.65417982139062e-06,
"loss": 0.2819,
"step": 17
},
{
"epoch": 0.3456,
"grad_norm": 0.5319820046424866,
"learning_rate": 8.412765716093273e-06,
"loss": 0.2765,
"step": 18
},
{
"epoch": 0.3648,
"grad_norm": 0.4264802932739258,
"learning_rate": 8.155439721630265e-06,
"loss": 0.2588,
"step": 19
},
{
"epoch": 0.384,
"grad_norm": 0.42277446389198303,
"learning_rate": 7.883401610574338e-06,
"loss": 0.2572,
"step": 20
},
{
"epoch": 0.4032,
"grad_norm": 0.46798449754714966,
"learning_rate": 7.597919750177168e-06,
"loss": 0.2319,
"step": 21
},
{
"epoch": 0.4224,
"grad_norm": 0.4637551009654999,
"learning_rate": 7.300325188655762e-06,
"loss": 0.2307,
"step": 22
},
{
"epoch": 0.4416,
"grad_norm": 0.477437824010849,
"learning_rate": 6.9920054492312086e-06,
"loss": 0.2495,
"step": 23
},
{
"epoch": 0.4608,
"grad_norm": 0.40543755888938904,
"learning_rate": 6.674398060854931e-06,
"loss": 0.2272,
"step": 24
},
{
"epoch": 0.48,
"grad_norm": 0.4067147672176361,
"learning_rate": 6.348983855785122e-06,
"loss": 0.2349,
"step": 25
},
{
"epoch": 0.4992,
"grad_norm": 0.37864550948143005,
"learning_rate": 6.0172800652631706e-06,
"loss": 0.2345,
"step": 26
},
{
"epoch": 0.5184,
"grad_norm": 0.3472854197025299,
"learning_rate": 5.680833245481234e-06,
"loss": 0.2192,
"step": 27
},
{
"epoch": 0.5376,
"grad_norm": 0.36279794573783875,
"learning_rate": 5.341212066823356e-06,
"loss": 0.2212,
"step": 28
},
{
"epoch": 0.5568,
"grad_norm": 0.35561856627464294,
"learning_rate": 5e-06,
"loss": 0.2218,
"step": 29
},
{
"epoch": 0.576,
"grad_norm": 0.34567970037460327,
"learning_rate": 4.6587879331766465e-06,
"loss": 0.2157,
"step": 30
},
{
"epoch": 0.5952,
"grad_norm": 0.3285064101219177,
"learning_rate": 4.319166754518768e-06,
"loss": 0.2214,
"step": 31
},
{
"epoch": 0.6144,
"grad_norm": 0.28714433312416077,
"learning_rate": 3.982719934736832e-06,
"loss": 0.1882,
"step": 32
},
{
"epoch": 0.6336,
"grad_norm": 0.31616437435150146,
"learning_rate": 3.6510161442148783e-06,
"loss": 0.1859,
"step": 33
},
{
"epoch": 0.6528,
"grad_norm": 0.3225463330745697,
"learning_rate": 3.3256019391450696e-06,
"loss": 0.2063,
"step": 34
},
{
"epoch": 0.672,
"grad_norm": 0.33429035544395447,
"learning_rate": 3.007994550768793e-06,
"loss": 0.2043,
"step": 35
},
{
"epoch": 0.6912,
"grad_norm": 0.33725297451019287,
"learning_rate": 2.6996748113442397e-06,
"loss": 0.2168,
"step": 36
},
{
"epoch": 0.7104,
"grad_norm": 0.3270690143108368,
"learning_rate": 2.4020802498228333e-06,
"loss": 0.2076,
"step": 37
},
{
"epoch": 0.7296,
"grad_norm": 0.34343093633651733,
"learning_rate": 2.1165983894256647e-06,
"loss": 0.2022,
"step": 38
},
{
"epoch": 0.7488,
"grad_norm": 0.3212158977985382,
"learning_rate": 1.8445602783697375e-06,
"loss": 0.2043,
"step": 39
},
{
"epoch": 0.768,
"grad_norm": 0.4333406388759613,
"learning_rate": 1.5872342839067305e-06,
"loss": 0.1961,
"step": 40
},
{
"epoch": 0.7872,
"grad_norm": 0.3265644609928131,
"learning_rate": 1.3458201786093795e-06,
"loss": 0.206,
"step": 41
},
{
"epoch": 0.8064,
"grad_norm": 0.3291531503200531,
"learning_rate": 1.1214435464779006e-06,
"loss": 0.2164,
"step": 42
},
{
"epoch": 0.8256,
"grad_norm": 0.3214084506034851,
"learning_rate": 9.151505349477901e-07,
"loss": 0.1954,
"step": 43
},
{
"epoch": 0.8448,
"grad_norm": 0.33091598749160767,
"learning_rate": 7.279029772675572e-07,
"loss": 0.2029,
"step": 44
},
{
"epoch": 0.864,
"grad_norm": 0.305397629737854,
"learning_rate": 5.60573907988124e-07,
"loss": 0.1895,
"step": 45
},
{
"epoch": 0.8832,
"grad_norm": 0.3404708504676819,
"learning_rate": 4.139434924727359e-07,
"loss": 0.206,
"step": 46
},
{
"epoch": 0.9024,
"grad_norm": 0.3258639872074127,
"learning_rate": 2.88695389405898e-07,
"loss": 0.2152,
"step": 47
},
{
"epoch": 0.9216,
"grad_norm": 0.3110244870185852,
"learning_rate": 1.8541356326100436e-07,
"loss": 0.2068,
"step": 48
},
{
"epoch": 0.9408,
"grad_norm": 0.31024959683418274,
"learning_rate": 1.0457956158838545e-07,
"loss": 0.1988,
"step": 49
},
{
"epoch": 0.96,
"grad_norm": 0.3411162495613098,
"learning_rate": 4.657026981834623e-08,
"loss": 0.2048,
"step": 50
},
{
"epoch": 0.9792,
"grad_norm": 0.32382944226264954,
"learning_rate": 1.1656154047303691e-08,
"loss": 0.2035,
"step": 51
},
{
"epoch": 0.9984,
"grad_norm": 0.3142213821411133,
"learning_rate": 0.0,
"loss": 0.2071,
"step": 52
},
{
"epoch": 0.9984,
"step": 52,
"total_flos": 149313677950976.0,
"train_loss": 0.26948727323458743,
"train_runtime": 7475.3088,
"train_samples_per_second": 0.669,
"train_steps_per_second": 0.007
}
],
"logging_steps": 1,
"max_steps": 52,
"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": 149313677950976.0,
"train_batch_size": 1,
"trial_name": null,
"trial_params": null
}