| { | |
| "best_metric": null, | |
| "best_model_checkpoint": null, | |
| "epoch": 0.9937888198757764, | |
| "eval_steps": 500, | |
| "global_step": 80, | |
| "is_hyper_param_search": false, | |
| "is_local_process_zero": true, | |
| "is_world_process_zero": true, | |
| "log_history": [ | |
| { | |
| "epoch": 0.012422360248447204, | |
| "grad_norm": 21.184524536132812, | |
| "learning_rate": 2.5e-05, | |
| "loss": 3.7651, | |
| "step": 1 | |
| }, | |
| { | |
| "epoch": 0.062111801242236024, | |
| "grad_norm": 22.220394134521484, | |
| "learning_rate": 0.000125, | |
| "loss": 3.7771, | |
| "step": 5 | |
| }, | |
| { | |
| "epoch": 0.12422360248447205, | |
| "grad_norm": 12.797124862670898, | |
| "learning_rate": 0.00019961946980917456, | |
| "loss": 3.031, | |
| "step": 10 | |
| }, | |
| { | |
| "epoch": 0.18633540372670807, | |
| "grad_norm": 5.477193832397461, | |
| "learning_rate": 0.0001953716950748227, | |
| "loss": 2.3288, | |
| "step": 15 | |
| }, | |
| { | |
| "epoch": 0.2484472049689441, | |
| "grad_norm": 2.4697020053863525, | |
| "learning_rate": 0.00018660254037844388, | |
| "loss": 1.938, | |
| "step": 20 | |
| }, | |
| { | |
| "epoch": 0.3105590062111801, | |
| "grad_norm": 1.60536789894104, | |
| "learning_rate": 0.0001737277336810124, | |
| "loss": 1.6815, | |
| "step": 25 | |
| }, | |
| { | |
| "epoch": 0.37267080745341613, | |
| "grad_norm": 0.9478007555007935, | |
| "learning_rate": 0.0001573576436351046, | |
| "loss": 1.5625, | |
| "step": 30 | |
| }, | |
| { | |
| "epoch": 0.43478260869565216, | |
| "grad_norm": 0.5295054912567139, | |
| "learning_rate": 0.000138268343236509, | |
| "loss": 1.4703, | |
| "step": 35 | |
| }, | |
| { | |
| "epoch": 0.4968944099378882, | |
| "grad_norm": 0.42334404587745667, | |
| "learning_rate": 0.00011736481776669306, | |
| "loss": 1.4589, | |
| "step": 40 | |
| }, | |
| { | |
| "epoch": 0.5590062111801242, | |
| "grad_norm": 0.3187739849090576, | |
| "learning_rate": 9.563806126346642e-05, | |
| "loss": 1.4376, | |
| "step": 45 | |
| }, | |
| { | |
| "epoch": 0.6211180124223602, | |
| "grad_norm": 0.3597185015678406, | |
| "learning_rate": 7.411809548974792e-05, | |
| "loss": 1.4271, | |
| "step": 50 | |
| }, | |
| { | |
| "epoch": 0.6832298136645962, | |
| "grad_norm": 0.2957977056503296, | |
| "learning_rate": 5.382513867649663e-05, | |
| "loss": 1.4297, | |
| "step": 55 | |
| }, | |
| { | |
| "epoch": 0.7453416149068323, | |
| "grad_norm": 0.29805850982666016, | |
| "learning_rate": 3.5721239031346066e-05, | |
| "loss": 1.4191, | |
| "step": 60 | |
| }, | |
| { | |
| "epoch": 0.8074534161490683, | |
| "grad_norm": 0.27294573187828064, | |
| "learning_rate": 2.0664665970876496e-05, | |
| "loss": 1.403, | |
| "step": 65 | |
| }, | |
| { | |
| "epoch": 0.8695652173913043, | |
| "grad_norm": 0.33928054571151733, | |
| "learning_rate": 9.369221296335006e-06, | |
| "loss": 1.4174, | |
| "step": 70 | |
| }, | |
| { | |
| "epoch": 0.9316770186335404, | |
| "grad_norm": 0.31092315912246704, | |
| "learning_rate": 2.3703992880066638e-06, | |
| "loss": 1.3976, | |
| "step": 75 | |
| }, | |
| { | |
| "epoch": 0.9937888198757764, | |
| "grad_norm": 0.2693531811237335, | |
| "learning_rate": 0.0, | |
| "loss": 1.4059, | |
| "step": 80 | |
| }, | |
| { | |
| "epoch": 0.9937888198757764, | |
| "eval_loss": 1.4015706777572632, | |
| "eval_runtime": 2.2495, | |
| "eval_samples_per_second": 91.13, | |
| "eval_steps_per_second": 0.889, | |
| "step": 80 | |
| }, | |
| { | |
| "epoch": 0.9937888198757764, | |
| "step": 80, | |
| "total_flos": 8.089924109389005e+17, | |
| "train_loss": 1.786448487639427, | |
| "train_runtime": 830.0773, | |
| "train_samples_per_second": 24.8, | |
| "train_steps_per_second": 0.096 | |
| } | |
| ], | |
| "logging_steps": 5, | |
| "max_steps": 80, | |
| "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": 8.089924109389005e+17, | |
| "train_batch_size": 16, | |
| "trial_name": null, | |
| "trial_params": null | |
| } | |