| { | |
| "best_global_step": null, | |
| "best_metric": null, | |
| "best_model_checkpoint": null, | |
| "epoch": 1.0, | |
| "eval_steps": 500, | |
| "global_step": 198, | |
| "is_hyper_param_search": false, | |
| "is_local_process_zero": true, | |
| "is_world_process_zero": true, | |
| "log_history": [ | |
| { | |
| "epoch": 0.00506409241968666, | |
| "grad_norm": 5.678173069867947, | |
| "learning_rate": 0.0001, | |
| "loss": 0.3528, | |
| "mean_token_accuracy": 0.9147689286619425, | |
| "num_tokens": 586487.0, | |
| "step": 1 | |
| }, | |
| { | |
| "epoch": 0.05064092419686659, | |
| "grad_norm": 0.3134267771799764, | |
| "learning_rate": 0.0001, | |
| "loss": 0.2555, | |
| "mean_token_accuracy": 0.9199666447109647, | |
| "num_tokens": 5968280.0, | |
| "step": 10 | |
| }, | |
| { | |
| "epoch": 0.10128184839373318, | |
| "grad_norm": 0.14841672886560275, | |
| "learning_rate": 0.0001, | |
| "loss": 0.1589, | |
| "mean_token_accuracy": 0.9397787630558014, | |
| "num_tokens": 11930513.0, | |
| "step": 20 | |
| }, | |
| { | |
| "epoch": 0.15192277259059978, | |
| "grad_norm": 0.0924952729966239, | |
| "learning_rate": 0.0001, | |
| "loss": 0.139, | |
| "mean_token_accuracy": 0.946698647364974, | |
| "num_tokens": 17853273.0, | |
| "step": 30 | |
| }, | |
| { | |
| "epoch": 0.20256369678746636, | |
| "grad_norm": 0.08634084330086704, | |
| "learning_rate": 0.0001, | |
| "loss": 0.1277, | |
| "mean_token_accuracy": 0.95089195612818, | |
| "num_tokens": 23792525.0, | |
| "step": 40 | |
| }, | |
| { | |
| "epoch": 0.25320462098433294, | |
| "grad_norm": 0.07548012506474341, | |
| "learning_rate": 0.0001, | |
| "loss": 0.1212, | |
| "mean_token_accuracy": 0.953272457793355, | |
| "num_tokens": 29710036.0, | |
| "step": 50 | |
| }, | |
| { | |
| "epoch": 0.30384554518119955, | |
| "grad_norm": 0.07448026254865987, | |
| "learning_rate": 0.0001, | |
| "loss": 0.1151, | |
| "mean_token_accuracy": 0.9557252813130617, | |
| "num_tokens": 35685902.0, | |
| "step": 60 | |
| }, | |
| { | |
| "epoch": 0.35448646937806616, | |
| "grad_norm": 0.0879124617580733, | |
| "learning_rate": 0.0001, | |
| "loss": 0.1112, | |
| "mean_token_accuracy": 0.957416345924139, | |
| "num_tokens": 41642058.0, | |
| "step": 70 | |
| }, | |
| { | |
| "epoch": 0.4051273935749327, | |
| "grad_norm": 0.07273411152988447, | |
| "learning_rate": 0.0001, | |
| "loss": 0.1098, | |
| "mean_token_accuracy": 0.9579035054892302, | |
| "num_tokens": 47564730.0, | |
| "step": 80 | |
| }, | |
| { | |
| "epoch": 0.4557683177717993, | |
| "grad_norm": 0.07959899342781425, | |
| "learning_rate": 0.0001, | |
| "loss": 0.1076, | |
| "mean_token_accuracy": 0.9587001299485565, | |
| "num_tokens": 53504116.0, | |
| "step": 90 | |
| }, | |
| { | |
| "epoch": 0.5064092419686659, | |
| "grad_norm": 0.09016384504615442, | |
| "learning_rate": 0.0001, | |
| "loss": 0.1064, | |
| "mean_token_accuracy": 0.9593421731144189, | |
| "num_tokens": 59463168.0, | |
| "step": 100 | |
| }, | |
| { | |
| "epoch": 0.5570501661655325, | |
| "grad_norm": 0.08920769596857461, | |
| "learning_rate": 0.0001, | |
| "loss": 0.1061, | |
| "mean_token_accuracy": 0.9594598092138767, | |
| "num_tokens": 65384936.0, | |
| "step": 110 | |
| }, | |
| { | |
| "epoch": 0.6076910903623991, | |
| "grad_norm": 0.07392576254888708, | |
| "learning_rate": 0.0001, | |
| "loss": 0.1037, | |
| "mean_token_accuracy": 0.960389680787921, | |
| "num_tokens": 71307379.0, | |
| "step": 120 | |
| }, | |
| { | |
| "epoch": 0.6583320145592657, | |
| "grad_norm": 0.07677666218910574, | |
| "learning_rate": 0.0001, | |
| "loss": 0.1034, | |
| "mean_token_accuracy": 0.9603631895035505, | |
| "num_tokens": 77241947.0, | |
| "step": 130 | |
| }, | |
| { | |
| "epoch": 0.7089729387561323, | |
| "grad_norm": 0.08726421315898845, | |
| "learning_rate": 0.0001, | |
| "loss": 0.1039, | |
| "mean_token_accuracy": 0.9603854931890965, | |
| "num_tokens": 83153311.0, | |
| "step": 140 | |
| }, | |
| { | |
| "epoch": 0.7596138629529989, | |
| "grad_norm": 0.07666737355553582, | |
| "learning_rate": 0.0001, | |
| "loss": 0.1025, | |
| "mean_token_accuracy": 0.9607552452012896, | |
| "num_tokens": 89070746.0, | |
| "step": 150 | |
| }, | |
| { | |
| "epoch": 0.8102547871498654, | |
| "grad_norm": 0.07414032432163672, | |
| "learning_rate": 0.0001, | |
| "loss": 0.1015, | |
| "mean_token_accuracy": 0.9611390510573983, | |
| "num_tokens": 95023182.0, | |
| "step": 160 | |
| }, | |
| { | |
| "epoch": 0.860895711346732, | |
| "grad_norm": 0.07359359677738689, | |
| "learning_rate": 0.0001, | |
| "loss": 0.101, | |
| "mean_token_accuracy": 0.9613310528919101, | |
| "num_tokens": 100991573.0, | |
| "step": 170 | |
| }, | |
| { | |
| "epoch": 0.9115366355435987, | |
| "grad_norm": 0.07081191483858919, | |
| "learning_rate": 0.0001, | |
| "loss": 0.101, | |
| "mean_token_accuracy": 0.9613639689981938, | |
| "num_tokens": 106911646.0, | |
| "step": 180 | |
| }, | |
| { | |
| "epoch": 0.9621775597404653, | |
| "grad_norm": 0.06862894355772857, | |
| "learning_rate": 0.0001, | |
| "loss": 0.1005, | |
| "mean_token_accuracy": 0.9614966074004769, | |
| "num_tokens": 112858548.0, | |
| "step": 190 | |
| } | |
| ], | |
| "logging_steps": 10, | |
| "max_steps": 788, | |
| "num_input_tokens_seen": 0, | |
| "num_train_epochs": 4, | |
| "save_steps": 500, | |
| "stateful_callbacks": { | |
| "TrainerControl": { | |
| "args": { | |
| "should_epoch_stop": false, | |
| "should_evaluate": false, | |
| "should_log": false, | |
| "should_save": true, | |
| "should_training_stop": false | |
| }, | |
| "attributes": {} | |
| } | |
| }, | |
| "total_flos": 308608790560768.0, | |
| "train_batch_size": 8, | |
| "trial_name": null, | |
| "trial_params": null | |
| } | |