cherry-alpaca-5-percent-7B / trainer_state.json
Ming Li
Initial commit
a31655f
{
"best_metric": null,
"best_model_checkpoint": null,
"epoch": 2.961439588688946,
"global_step": 72,
"is_hyper_param_search": false,
"is_local_process_zero": true,
"is_world_process_zero": true,
"log_history": [
{
"epoch": 0.04,
"learning_rate": 6.666666666666667e-06,
"loss": 1.6215,
"step": 1
},
{
"epoch": 0.08,
"learning_rate": 1.3333333333333333e-05,
"loss": 1.6348,
"step": 2
},
{
"epoch": 0.12,
"learning_rate": 2e-05,
"loss": 1.4299,
"step": 3
},
{
"epoch": 0.16,
"learning_rate": 1.9989636736467278e-05,
"loss": 1.4334,
"step": 4
},
{
"epoch": 0.21,
"learning_rate": 1.9958568425315316e-05,
"loss": 1.468,
"step": 5
},
{
"epoch": 0.25,
"learning_rate": 1.9906859460363307e-05,
"loss": 1.4159,
"step": 6
},
{
"epoch": 0.29,
"learning_rate": 1.9834617016337424e-05,
"loss": 1.4128,
"step": 7
},
{
"epoch": 0.33,
"learning_rate": 1.9741990826734793e-05,
"loss": 1.3951,
"step": 8
},
{
"epoch": 0.37,
"learning_rate": 1.9629172873477995e-05,
"loss": 1.3818,
"step": 9
},
{
"epoch": 0.41,
"learning_rate": 1.9496396989003195e-05,
"loss": 1.342,
"step": 10
},
{
"epoch": 0.45,
"learning_rate": 1.9343938371606714e-05,
"loss": 1.2873,
"step": 11
},
{
"epoch": 0.49,
"learning_rate": 1.917211301505453e-05,
"loss": 1.3423,
"step": 12
},
{
"epoch": 0.53,
"learning_rate": 1.8981277053636963e-05,
"loss": 1.4581,
"step": 13
},
{
"epoch": 0.58,
"learning_rate": 1.8771826024025944e-05,
"loss": 1.3193,
"step": 14
},
{
"epoch": 0.62,
"learning_rate": 1.8544194045464888e-05,
"loss": 1.3919,
"step": 15
},
{
"epoch": 0.66,
"learning_rate": 1.8298852919990254e-05,
"loss": 1.374,
"step": 16
},
{
"epoch": 0.7,
"learning_rate": 1.8036311154549783e-05,
"loss": 1.3812,
"step": 17
},
{
"epoch": 0.74,
"learning_rate": 1.77571129070442e-05,
"loss": 1.3577,
"step": 18
},
{
"epoch": 0.78,
"learning_rate": 1.7461836858476858e-05,
"loss": 1.3745,
"step": 19
},
{
"epoch": 0.82,
"learning_rate": 1.7151095013548996e-05,
"loss": 1.3124,
"step": 20
},
{
"epoch": 0.86,
"learning_rate": 1.6825531432186545e-05,
"loss": 1.3827,
"step": 21
},
{
"epoch": 0.9,
"learning_rate": 1.648582089462756e-05,
"loss": 1.394,
"step": 22
},
{
"epoch": 0.95,
"learning_rate": 1.6132667502837164e-05,
"loss": 1.3605,
"step": 23
},
{
"epoch": 0.99,
"learning_rate": 1.5766803221148676e-05,
"loss": 1.35,
"step": 24
},
{
"epoch": 1.03,
"learning_rate": 1.538898635915576e-05,
"loss": 1.1304,
"step": 25
},
{
"epoch": 1.07,
"learning_rate": 1.5000000000000002e-05,
"loss": 1.0308,
"step": 26
},
{
"epoch": 1.11,
"learning_rate": 1.4600650377311523e-05,
"loss": 1.0333,
"step": 27
},
{
"epoch": 1.15,
"learning_rate": 1.4191765204166643e-05,
"loss": 0.9938,
"step": 28
},
{
"epoch": 1.19,
"learning_rate": 1.3774191957526144e-05,
"loss": 0.9677,
"step": 29
},
{
"epoch": 1.23,
"learning_rate": 1.3348796121709862e-05,
"loss": 1.0101,
"step": 30
},
{
"epoch": 1.28,
"learning_rate": 1.291645939454825e-05,
"loss": 0.9679,
"step": 31
},
{
"epoch": 1.32,
"learning_rate": 1.2478077859929e-05,
"loss": 0.9704,
"step": 32
},
{
"epoch": 1.36,
"learning_rate": 1.2034560130526341e-05,
"loss": 0.9165,
"step": 33
},
{
"epoch": 1.4,
"learning_rate": 1.1586825464562515e-05,
"loss": 0.9524,
"step": 34
},
{
"epoch": 1.44,
"learning_rate": 1.113580186050475e-05,
"loss": 0.9744,
"step": 35
},
{
"epoch": 1.48,
"learning_rate": 1.0682424133646712e-05,
"loss": 0.9251,
"step": 36
},
{
"epoch": 1.52,
"learning_rate": 1.0227631978561057e-05,
"loss": 0.92,
"step": 37
},
{
"epoch": 1.56,
"learning_rate": 9.772368021438943e-06,
"loss": 0.947,
"step": 38
},
{
"epoch": 1.6,
"learning_rate": 9.317575866353293e-06,
"loss": 0.9406,
"step": 39
},
{
"epoch": 1.65,
"learning_rate": 8.86419813949525e-06,
"loss": 0.9803,
"step": 40
},
{
"epoch": 1.69,
"learning_rate": 8.413174535437486e-06,
"loss": 0.932,
"step": 41
},
{
"epoch": 1.73,
"learning_rate": 7.965439869473664e-06,
"loss": 0.9271,
"step": 42
},
{
"epoch": 1.77,
"learning_rate": 7.521922140071003e-06,
"loss": 0.9081,
"step": 43
},
{
"epoch": 1.81,
"learning_rate": 7.0835406054517505e-06,
"loss": 0.922,
"step": 44
},
{
"epoch": 1.85,
"learning_rate": 6.651203878290139e-06,
"loss": 0.941,
"step": 45
},
{
"epoch": 1.89,
"learning_rate": 6.225808042473857e-06,
"loss": 0.9436,
"step": 46
},
{
"epoch": 1.93,
"learning_rate": 5.8082347958333625e-06,
"loss": 0.9453,
"step": 47
},
{
"epoch": 1.97,
"learning_rate": 5.399349622688479e-06,
"loss": 0.9162,
"step": 48
},
{
"epoch": 2.02,
"learning_rate": 5.000000000000003e-06,
"loss": 0.8596,
"step": 49
},
{
"epoch": 2.06,
"learning_rate": 4.611013640844245e-06,
"loss": 0.8061,
"step": 50
},
{
"epoch": 2.1,
"learning_rate": 4.2331967788513295e-06,
"loss": 0.7507,
"step": 51
},
{
"epoch": 2.14,
"learning_rate": 3.867332497162836e-06,
"loss": 0.7905,
"step": 52
},
{
"epoch": 2.18,
"learning_rate": 3.5141791053724405e-06,
"loss": 0.7714,
"step": 53
},
{
"epoch": 2.22,
"learning_rate": 3.174468567813461e-06,
"loss": 0.7438,
"step": 54
},
{
"epoch": 2.26,
"learning_rate": 2.8489049864510053e-06,
"loss": 0.6632,
"step": 55
},
{
"epoch": 2.3,
"learning_rate": 2.5381631415231455e-06,
"loss": 0.7124,
"step": 56
},
{
"epoch": 2.34,
"learning_rate": 2.2428870929558012e-06,
"loss": 0.7325,
"step": 57
},
{
"epoch": 2.39,
"learning_rate": 1.963688845450218e-06,
"loss": 0.6903,
"step": 58
},
{
"epoch": 2.43,
"learning_rate": 1.7011470800097496e-06,
"loss": 0.7075,
"step": 59
},
{
"epoch": 2.47,
"learning_rate": 1.4558059545351144e-06,
"loss": 0.6924,
"step": 60
},
{
"epoch": 2.51,
"learning_rate": 1.2281739759740575e-06,
"loss": 0.7008,
"step": 61
},
{
"epoch": 2.55,
"learning_rate": 1.01872294636304e-06,
"loss": 0.6721,
"step": 62
},
{
"epoch": 2.59,
"learning_rate": 8.278869849454718e-07,
"loss": 0.6658,
"step": 63
},
{
"epoch": 2.63,
"learning_rate": 6.560616283932897e-07,
"loss": 0.6464,
"step": 64
},
{
"epoch": 2.67,
"learning_rate": 5.036030109968082e-07,
"loss": 0.7287,
"step": 65
},
{
"epoch": 2.71,
"learning_rate": 3.708271265220087e-07,
"loss": 0.7194,
"step": 66
},
{
"epoch": 2.76,
"learning_rate": 2.5800917326521013e-07,
"loss": 0.7061,
"step": 67
},
{
"epoch": 2.8,
"learning_rate": 1.6538298366257975e-07,
"loss": 0.7292,
"step": 68
},
{
"epoch": 2.84,
"learning_rate": 9.314053963669245e-08,
"loss": 0.7474,
"step": 69
},
{
"epoch": 2.88,
"learning_rate": 4.143157468468717e-08,
"loss": 0.7519,
"step": 70
},
{
"epoch": 2.92,
"learning_rate": 1.0363263532724433e-08,
"loss": 0.7462,
"step": 71
},
{
"epoch": 2.96,
"learning_rate": 0.0,
"loss": 0.6997,
"step": 72
},
{
"epoch": 2.96,
"step": 72,
"total_flos": 4.919620776283341e+16,
"train_loss": 1.029880191716883,
"train_runtime": 10310.4521,
"train_samples_per_second": 0.905,
"train_steps_per_second": 0.007
}
],
"max_steps": 72,
"num_train_epochs": 3,
"total_flos": 4.919620776283341e+16,
"trial_name": null,
"trial_params": null
}