209 lines
5.4 KiB
JSON
209 lines
5.4 KiB
JSON
|
{
|
||
|
|
"best_global_step": null,
|
||
|
|
"best_metric": null,
|
||
|
|
"best_model_checkpoint": null,
|
||
|
|
"epoch": 7.0,
|
||
|
|
"eval_steps": 500,
|
||
|
|
"global_step": 77,
|
||
|
|
"is_hyper_param_search": false,
|
||
|
|
"is_local_process_zero": true,
|
||
|
|
"is_world_process_zero": true,
|
||
|
|
"log_history": [
|
||
|
|
{
|
||
|
|
"epoch": 0.46875,
|
||
|
|
"grad_norm": 1.673851302244658,
|
||
|
|
"learning_rate": 2e-05,
|
||
|
|
"loss": 0.3508,
|
||
|
|
"loss_nan_ranks": 0,
|
||
|
|
"loss_rank_avg": 0.09443844109773636,
|
||
|
|
"step": 5,
|
||
|
|
"valid_targets_mean": 6473.8,
|
||
|
|
"valid_targets_min": 1524
|
||
|
|
},
|
||
|
|
{
|
||
|
|
"epoch": 0.9375,
|
||
|
|
"grad_norm": 0.3907133266191781,
|
||
|
|
"learning_rate": 3.9979273472934556e-05,
|
||
|
|
"loss": 0.2609,
|
||
|
|
"loss_nan_ranks": 0,
|
||
|
|
"loss_rank_avg": 0.07388223707675934,
|
||
|
|
"step": 10,
|
||
|
|
"valid_targets_mean": 5224.7,
|
||
|
|
"valid_targets_min": 1096
|
||
|
|
},
|
||
|
|
{
|
||
|
|
"epoch": 1.375,
|
||
|
|
"grad_norm": 0.184278311171899,
|
||
|
|
"learning_rate": 3.925834574695599e-05,
|
||
|
|
"loss": 0.2051,
|
||
|
|
"loss_nan_ranks": 0,
|
||
|
|
"loss_rank_avg": 0.06569650769233704,
|
||
|
|
"step": 15,
|
||
|
|
"valid_targets_mean": 8057.2,
|
||
|
|
"valid_targets_min": 1524
|
||
|
|
},
|
||
|
|
{
|
||
|
|
"epoch": 1.84375,
|
||
|
|
"grad_norm": 0.16863686013098317,
|
||
|
|
"learning_rate": 3.754365204805189e-05,
|
||
|
|
"loss": 0.186,
|
||
|
|
"loss_nan_ranks": 0,
|
||
|
|
"loss_rank_avg": 0.05596768856048584,
|
||
|
|
"step": 20,
|
||
|
|
"valid_targets_mean": 4704.4,
|
||
|
|
"valid_targets_min": 1013
|
||
|
|
},
|
||
|
|
{
|
||
|
|
"epoch": 2.28125,
|
||
|
|
"grad_norm": 0.1823655395194005,
|
||
|
|
"learning_rate": 3.4923673716953717e-05,
|
||
|
|
"loss": 0.1513,
|
||
|
|
"loss_nan_ranks": 0,
|
||
|
|
"loss_rank_avg": 0.05063176155090332,
|
||
|
|
"step": 25,
|
||
|
|
"valid_targets_mean": 6318.8,
|
||
|
|
"valid_targets_min": 1096
|
||
|
|
},
|
||
|
|
{
|
||
|
|
"epoch": 2.75,
|
||
|
|
"grad_norm": 0.17231125581468304,
|
||
|
|
"learning_rate": 3.153360644229735e-05,
|
||
|
|
"loss": 0.1469,
|
||
|
|
"loss_nan_ranks": 0,
|
||
|
|
"loss_rank_avg": 0.04585719481110573,
|
||
|
|
"step": 30,
|
||
|
|
"valid_targets_mean": 4503.4,
|
||
|
|
"valid_targets_min": 978
|
||
|
|
},
|
||
|
|
{
|
||
|
|
"epoch": 3.1875,
|
||
|
|
"grad_norm": 0.17740764070105294,
|
||
|
|
"learning_rate": 2.7548383915052287e-05,
|
||
|
|
"loss": 0.135,
|
||
|
|
"loss_nan_ranks": 0,
|
||
|
|
"loss_rank_avg": 0.038901083171367645,
|
||
|
|
"step": 35,
|
||
|
|
"valid_targets_mean": 6662.6,
|
||
|
|
"valid_targets_min": 990
|
||
|
|
},
|
||
|
|
{
|
||
|
|
"epoch": 3.65625,
|
||
|
|
"grad_norm": 0.17905235430627578,
|
||
|
|
"learning_rate": 2.317365092912503e-05,
|
||
|
|
"loss": 0.1126,
|
||
|
|
"loss_nan_ranks": 0,
|
||
|
|
"loss_rank_avg": 0.029759323224425316,
|
||
|
|
"step": 40,
|
||
|
|
"valid_targets_mean": 5237.5,
|
||
|
|
"valid_targets_min": 1023
|
||
|
|
},
|
||
|
|
{
|
||
|
|
"epoch": 4.09375,
|
||
|
|
"grad_norm": 0.19664414903717908,
|
||
|
|
"learning_rate": 1.8635151732706586e-05,
|
||
|
|
"loss": 0.1021,
|
||
|
|
"loss_nan_ranks": 0,
|
||
|
|
"loss_rank_avg": 0.039972178637981415,
|
||
|
|
"step": 45,
|
||
|
|
"valid_targets_mean": 5255.4,
|
||
|
|
"valid_targets_min": 516
|
||
|
|
},
|
||
|
|
{
|
||
|
|
"epoch": 4.5625,
|
||
|
|
"grad_norm": 0.21650155303669674,
|
||
|
|
"learning_rate": 1.4167081210903501e-05,
|
||
|
|
"loss": 0.0882,
|
||
|
|
"loss_nan_ranks": 0,
|
||
|
|
"loss_rank_avg": 0.02802399918437004,
|
||
|
|
"step": 50,
|
||
|
|
"valid_targets_mean": 5966.8,
|
||
|
|
"valid_targets_min": 1185
|
||
|
|
},
|
||
|
|
{
|
||
|
|
"epoch": 5.0,
|
||
|
|
"grad_norm": 0.1866861112102521,
|
||
|
|
"learning_rate": 1.0000000000000006e-05,
|
||
|
|
"loss": 0.0833,
|
||
|
|
"loss_nan_ranks": 0,
|
||
|
|
"loss_rank_avg": 0.029194967821240425,
|
||
|
|
"step": 55,
|
||
|
|
"valid_targets_mean": 5956.2,
|
||
|
|
"valid_targets_min": 399
|
||
|
|
},
|
||
|
|
{
|
||
|
|
"epoch": 5.46875,
|
||
|
|
"grad_norm": 0.2006117168960997,
|
||
|
|
"learning_rate": 6.348937135626922e-06,
|
||
|
|
"loss": 0.0724,
|
||
|
|
"loss_nan_ranks": 0,
|
||
|
|
"loss_rank_avg": 0.01756685972213745,
|
||
|
|
"step": 60,
|
||
|
|
"valid_targets_mean": 4807.5,
|
||
|
|
"valid_targets_min": 436
|
||
|
|
},
|
||
|
|
{
|
||
|
|
"epoch": 5.9375,
|
||
|
|
"grad_norm": 0.19922579399109103,
|
||
|
|
"learning_rate": 3.402294160019499e-06,
|
||
|
|
"loss": 0.0722,
|
||
|
|
"loss_nan_ranks": 0,
|
||
|
|
"loss_rank_avg": 0.024522025138139725,
|
||
|
|
"step": 65,
|
||
|
|
"valid_targets_mean": 5606.9,
|
||
|
|
"valid_targets_min": 1425
|
||
|
|
},
|
||
|
|
{
|
||
|
|
"epoch": 6.375,
|
||
|
|
"grad_norm": 0.1953832655127811,
|
||
|
|
"learning_rate": 1.3121232567865793e-06,
|
||
|
|
"loss": 0.0636,
|
||
|
|
"loss_nan_ranks": 0,
|
||
|
|
"loss_rank_avg": 0.017598971724510193,
|
||
|
|
"step": 70,
|
||
|
|
"valid_targets_mean": 4612.0,
|
||
|
|
"valid_targets_min": 510
|
||
|
|
},
|
||
|
|
{
|
||
|
|
"epoch": 6.84375,
|
||
|
|
"grad_norm": 0.15480982820327163,
|
||
|
|
"learning_rate": 1.862810792733849e-07,
|
||
|
|
"loss": 0.0672,
|
||
|
|
"loss_nan_ranks": 0,
|
||
|
|
"loss_rank_avg": 0.020643679425120354,
|
||
|
|
"step": 75,
|
||
|
|
"valid_targets_mean": 7398.4,
|
||
|
|
"valid_targets_min": 1134
|
||
|
|
},
|
||
|
|
{
|
||
|
|
"epoch": 7.0,
|
||
|
|
"step": 77,
|
||
|
|
"total_flos": 2.1021193659010253e+17,
|
||
|
|
"train_loss": 0.0,
|
||
|
|
"train_runtime": 0.875,
|
||
|
|
"train_samples_per_second": 7999.638,
|
||
|
|
"train_steps_per_second": 87.996
|
||
|
|
}
|
||
|
|
],
|
||
|
|
"logging_steps": 5,
|
||
|
|
"max_steps": 77,
|
||
|
|
"num_input_tokens_seen": 0,
|
||
|
|
"num_train_epochs": 7,
|
||
|
|
"save_steps": 50,
|
||
|
|
"stateful_callbacks": {
|
||
|
|
"TrainerControl": {
|
||
|
|
"args": {
|
||
|
|
"should_epoch_stop": false,
|
||
|
|
"should_evaluate": false,
|
||
|
|
"should_log": false,
|
||
|
|
"should_save": true,
|
||
|
|
"should_training_stop": true
|
||
|
|
},
|
||
|
|
"attributes": {}
|
||
|
|
}
|
||
|
|
},
|
||
|
|
"total_flos": 2.1021193659010253e+17,
|
||
|
|
"train_batch_size": 1,
|
||
|
|
"trial_name": null,
|
||
|
|
"trial_params": null
|
||
|
|
}
|