| { | |
| "best_metric": null, | |
| "best_model_checkpoint": null, | |
| "epoch": 0.0005615516796010736, | |
| "eval_steps": 9, | |
| "global_step": 25, | |
| "is_hyper_param_search": false, | |
| "is_local_process_zero": true, | |
| "is_world_process_zero": true, | |
| "log_history": [ | |
| { | |
| "epoch": 2.246206718404295e-05, | |
| "grad_norm": 3.0712389945983887, | |
| "learning_rate": 1e-05, | |
| "loss": 4.5815, | |
| "step": 1 | |
| }, | |
| { | |
| "epoch": 2.246206718404295e-05, | |
| "eval_loss": 2.2074947357177734, | |
| "eval_runtime": 3664.8507, | |
| "eval_samples_per_second": 10.23, | |
| "eval_steps_per_second": 1.279, | |
| "step": 1 | |
| }, | |
| { | |
| "epoch": 4.49241343680859e-05, | |
| "grad_norm": 2.025412082672119, | |
| "learning_rate": 2e-05, | |
| "loss": 3.7769, | |
| "step": 2 | |
| }, | |
| { | |
| "epoch": 6.738620155212884e-05, | |
| "grad_norm": 2.5275726318359375, | |
| "learning_rate": 3e-05, | |
| "loss": 4.2228, | |
| "step": 3 | |
| }, | |
| { | |
| "epoch": 8.98482687361718e-05, | |
| "grad_norm": 2.5390024185180664, | |
| "learning_rate": 4e-05, | |
| "loss": 4.9002, | |
| "step": 4 | |
| }, | |
| { | |
| "epoch": 0.00011231033592021474, | |
| "grad_norm": 2.9198367595672607, | |
| "learning_rate": 5e-05, | |
| "loss": 4.7604, | |
| "step": 5 | |
| }, | |
| { | |
| "epoch": 0.00013477240310425767, | |
| "grad_norm": 3.3159918785095215, | |
| "learning_rate": 6e-05, | |
| "loss": 4.1514, | |
| "step": 6 | |
| }, | |
| { | |
| "epoch": 0.00015723447028830065, | |
| "grad_norm": 3.7733492851257324, | |
| "learning_rate": 7e-05, | |
| "loss": 4.502, | |
| "step": 7 | |
| }, | |
| { | |
| "epoch": 0.0001796965374723436, | |
| "grad_norm": 3.3140640258789062, | |
| "learning_rate": 8e-05, | |
| "loss": 3.4606, | |
| "step": 8 | |
| }, | |
| { | |
| "epoch": 0.00020215860465638654, | |
| "grad_norm": 5.731607437133789, | |
| "learning_rate": 9e-05, | |
| "loss": 4.2619, | |
| "step": 9 | |
| }, | |
| { | |
| "epoch": 0.00020215860465638654, | |
| "eval_loss": 2.0114121437072754, | |
| "eval_runtime": 3693.2986, | |
| "eval_samples_per_second": 10.151, | |
| "eval_steps_per_second": 1.269, | |
| "step": 9 | |
| }, | |
| { | |
| "epoch": 0.00022462067184042948, | |
| "grad_norm": 4.831848621368408, | |
| "learning_rate": 0.0001, | |
| "loss": 4.6822, | |
| "step": 10 | |
| }, | |
| { | |
| "epoch": 0.0002470827390244724, | |
| "grad_norm": 2.6047534942626953, | |
| "learning_rate": 9.99695413509548e-05, | |
| "loss": 4.2618, | |
| "step": 11 | |
| }, | |
| { | |
| "epoch": 0.00026954480620851535, | |
| "grad_norm": 2.650613784790039, | |
| "learning_rate": 9.987820251299122e-05, | |
| "loss": 3.7387, | |
| "step": 12 | |
| }, | |
| { | |
| "epoch": 0.0002920068733925583, | |
| "grad_norm": 2.0770010948181152, | |
| "learning_rate": 9.972609476841367e-05, | |
| "loss": 3.8619, | |
| "step": 13 | |
| }, | |
| { | |
| "epoch": 0.0003144689405766013, | |
| "grad_norm": 4.928474426269531, | |
| "learning_rate": 9.951340343707852e-05, | |
| "loss": 3.99, | |
| "step": 14 | |
| }, | |
| { | |
| "epoch": 0.00033693100776064424, | |
| "grad_norm": 3.021501064300537, | |
| "learning_rate": 9.924038765061042e-05, | |
| "loss": 4.1065, | |
| "step": 15 | |
| }, | |
| { | |
| "epoch": 0.0003593930749446872, | |
| "grad_norm": 2.1191632747650146, | |
| "learning_rate": 9.890738003669029e-05, | |
| "loss": 3.3574, | |
| "step": 16 | |
| }, | |
| { | |
| "epoch": 0.0003818551421287301, | |
| "grad_norm": 2.9330360889434814, | |
| "learning_rate": 9.851478631379982e-05, | |
| "loss": 4.126, | |
| "step": 17 | |
| }, | |
| { | |
| "epoch": 0.00040431720931277307, | |
| "grad_norm": 3.0402603149414062, | |
| "learning_rate": 9.806308479691595e-05, | |
| "loss": 3.7699, | |
| "step": 18 | |
| }, | |
| { | |
| "epoch": 0.00040431720931277307, | |
| "eval_loss": 1.824965000152588, | |
| "eval_runtime": 3686.782, | |
| "eval_samples_per_second": 10.169, | |
| "eval_steps_per_second": 1.271, | |
| "step": 18 | |
| }, | |
| { | |
| "epoch": 0.000426779276496816, | |
| "grad_norm": 2.7607719898223877, | |
| "learning_rate": 9.755282581475769e-05, | |
| "loss": 3.1566, | |
| "step": 19 | |
| }, | |
| { | |
| "epoch": 0.00044924134368085896, | |
| "grad_norm": 2.5294137001037598, | |
| "learning_rate": 9.698463103929542e-05, | |
| "loss": 4.1573, | |
| "step": 20 | |
| }, | |
| { | |
| "epoch": 0.0004717034108649019, | |
| "grad_norm": 3.3328206539154053, | |
| "learning_rate": 9.635919272833938e-05, | |
| "loss": 2.941, | |
| "step": 21 | |
| }, | |
| { | |
| "epoch": 0.0004941654780489448, | |
| "grad_norm": 3.416025161743164, | |
| "learning_rate": 9.567727288213005e-05, | |
| "loss": 3.0187, | |
| "step": 22 | |
| }, | |
| { | |
| "epoch": 0.0005166275452329877, | |
| "grad_norm": 2.709918975830078, | |
| "learning_rate": 9.493970231495835e-05, | |
| "loss": 3.157, | |
| "step": 23 | |
| }, | |
| { | |
| "epoch": 0.0005390896124170307, | |
| "grad_norm": 1.9112098217010498, | |
| "learning_rate": 9.414737964294636e-05, | |
| "loss": 3.7691, | |
| "step": 24 | |
| }, | |
| { | |
| "epoch": 0.0005615516796010736, | |
| "grad_norm": 4.849649906158447, | |
| "learning_rate": 9.330127018922194e-05, | |
| "loss": 4.0183, | |
| "step": 25 | |
| } | |
| ], | |
| "logging_steps": 1, | |
| "max_steps": 100, | |
| "num_input_tokens_seen": 0, | |
| "num_train_epochs": 1, | |
| "save_steps": 25, | |
| "stateful_callbacks": { | |
| "TrainerControl": { | |
| "args": { | |
| "should_epoch_stop": false, | |
| "should_evaluate": false, | |
| "should_log": false, | |
| "should_save": true, | |
| "should_training_stop": false | |
| }, | |
| "attributes": {} | |
| } | |
| }, | |
| "total_flos": 1.63364550475776e+16, | |
| "train_batch_size": 8, | |
| "trial_name": null, | |
| "trial_params": null | |
| } | |