| { | |
| "best_metric": null, | |
| "best_model_checkpoint": null, | |
| "epoch": 0.04849660523763336, | |
| "eval_steps": 9, | |
| "global_step": 25, | |
| "is_hyper_param_search": false, | |
| "is_local_process_zero": true, | |
| "is_world_process_zero": true, | |
| "log_history": [ | |
| { | |
| "epoch": 0.0019398642095053346, | |
| "grad_norm": 5.092199802398682, | |
| "learning_rate": 1e-05, | |
| "loss": 4.7383, | |
| "step": 1 | |
| }, | |
| { | |
| "epoch": 0.0019398642095053346, | |
| "eval_loss": 4.821082592010498, | |
| "eval_runtime": 9.658, | |
| "eval_samples_per_second": 44.937, | |
| "eval_steps_per_second": 5.695, | |
| "step": 1 | |
| }, | |
| { | |
| "epoch": 0.0038797284190106693, | |
| "grad_norm": 4.449008464813232, | |
| "learning_rate": 2e-05, | |
| "loss": 4.6036, | |
| "step": 2 | |
| }, | |
| { | |
| "epoch": 0.005819592628516004, | |
| "grad_norm": 4.542291164398193, | |
| "learning_rate": 3e-05, | |
| "loss": 4.7654, | |
| "step": 3 | |
| }, | |
| { | |
| "epoch": 0.007759456838021339, | |
| "grad_norm": 4.831502437591553, | |
| "learning_rate": 4e-05, | |
| "loss": 4.6307, | |
| "step": 4 | |
| }, | |
| { | |
| "epoch": 0.009699321047526674, | |
| "grad_norm": 4.798445224761963, | |
| "learning_rate": 5e-05, | |
| "loss": 4.689, | |
| "step": 5 | |
| }, | |
| { | |
| "epoch": 0.011639185257032008, | |
| "grad_norm": 4.600784778594971, | |
| "learning_rate": 6e-05, | |
| "loss": 4.4962, | |
| "step": 6 | |
| }, | |
| { | |
| "epoch": 0.013579049466537343, | |
| "grad_norm": 4.6975178718566895, | |
| "learning_rate": 7e-05, | |
| "loss": 4.6526, | |
| "step": 7 | |
| }, | |
| { | |
| "epoch": 0.015518913676042677, | |
| "grad_norm": 4.297379970550537, | |
| "learning_rate": 8e-05, | |
| "loss": 4.4603, | |
| "step": 8 | |
| }, | |
| { | |
| "epoch": 0.01745877788554801, | |
| "grad_norm": 4.602691650390625, | |
| "learning_rate": 9e-05, | |
| "loss": 4.0801, | |
| "step": 9 | |
| }, | |
| { | |
| "epoch": 0.01745877788554801, | |
| "eval_loss": 3.935520887374878, | |
| "eval_runtime": 7.7589, | |
| "eval_samples_per_second": 55.935, | |
| "eval_steps_per_second": 7.089, | |
| "step": 9 | |
| }, | |
| { | |
| "epoch": 0.019398642095053348, | |
| "grad_norm": 3.76698899269104, | |
| "learning_rate": 0.0001, | |
| "loss": 3.9645, | |
| "step": 10 | |
| }, | |
| { | |
| "epoch": 0.02133850630455868, | |
| "grad_norm": 3.087214708328247, | |
| "learning_rate": 9.99695413509548e-05, | |
| "loss": 3.9473, | |
| "step": 11 | |
| }, | |
| { | |
| "epoch": 0.023278370514064017, | |
| "grad_norm": 3.322976589202881, | |
| "learning_rate": 9.987820251299122e-05, | |
| "loss": 3.652, | |
| "step": 12 | |
| }, | |
| { | |
| "epoch": 0.02521823472356935, | |
| "grad_norm": 3.5539023876190186, | |
| "learning_rate": 9.972609476841367e-05, | |
| "loss": 3.3627, | |
| "step": 13 | |
| }, | |
| { | |
| "epoch": 0.027158098933074686, | |
| "grad_norm": 3.190157413482666, | |
| "learning_rate": 9.951340343707852e-05, | |
| "loss": 3.1173, | |
| "step": 14 | |
| }, | |
| { | |
| "epoch": 0.029097963142580018, | |
| "grad_norm": 3.779177188873291, | |
| "learning_rate": 9.924038765061042e-05, | |
| "loss": 2.8321, | |
| "step": 15 | |
| }, | |
| { | |
| "epoch": 0.031037827352085354, | |
| "grad_norm": 3.7224929332733154, | |
| "learning_rate": 9.890738003669029e-05, | |
| "loss": 2.504, | |
| "step": 16 | |
| }, | |
| { | |
| "epoch": 0.03297769156159069, | |
| "grad_norm": 3.4633424282073975, | |
| "learning_rate": 9.851478631379982e-05, | |
| "loss": 2.2944, | |
| "step": 17 | |
| }, | |
| { | |
| "epoch": 0.03491755577109602, | |
| "grad_norm": 2.928765296936035, | |
| "learning_rate": 9.806308479691595e-05, | |
| "loss": 2.2625, | |
| "step": 18 | |
| }, | |
| { | |
| "epoch": 0.03491755577109602, | |
| "eval_loss": 2.196943759918213, | |
| "eval_runtime": 7.7522, | |
| "eval_samples_per_second": 55.984, | |
| "eval_steps_per_second": 7.095, | |
| "step": 18 | |
| }, | |
| { | |
| "epoch": 0.03685741998060136, | |
| "grad_norm": 3.9546890258789062, | |
| "learning_rate": 9.755282581475769e-05, | |
| "loss": 2.3878, | |
| "step": 19 | |
| }, | |
| { | |
| "epoch": 0.038797284190106696, | |
| "grad_norm": 4.105186462402344, | |
| "learning_rate": 9.698463103929542e-05, | |
| "loss": 2.1788, | |
| "step": 20 | |
| }, | |
| { | |
| "epoch": 0.040737148399612025, | |
| "grad_norm": 3.8948333263397217, | |
| "learning_rate": 9.635919272833938e-05, | |
| "loss": 1.8275, | |
| "step": 21 | |
| }, | |
| { | |
| "epoch": 0.04267701260911736, | |
| "grad_norm": 3.057760238647461, | |
| "learning_rate": 9.567727288213005e-05, | |
| "loss": 1.621, | |
| "step": 22 | |
| }, | |
| { | |
| "epoch": 0.0446168768186227, | |
| "grad_norm": 2.3399696350097656, | |
| "learning_rate": 9.493970231495835e-05, | |
| "loss": 1.655, | |
| "step": 23 | |
| }, | |
| { | |
| "epoch": 0.04655674102812803, | |
| "grad_norm": 2.517136335372925, | |
| "learning_rate": 9.414737964294636e-05, | |
| "loss": 1.7654, | |
| "step": 24 | |
| }, | |
| { | |
| "epoch": 0.04849660523763336, | |
| "grad_norm": 2.6656148433685303, | |
| "learning_rate": 9.330127018922194e-05, | |
| "loss": 1.7639, | |
| "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": 645594775289856.0, | |
| "train_batch_size": 8, | |
| "trial_name": null, | |
| "trial_params": null | |
| } | |