| { | |
| "best_metric": null, | |
| "best_model_checkpoint": null, | |
| "epoch": 3.9402985074626864, | |
| "eval_steps": 7, | |
| "global_step": 132, | |
| "is_hyper_param_search": false, | |
| "is_local_process_zero": true, | |
| "is_world_process_zero": true, | |
| "log_history": [ | |
| { | |
| "epoch": 0.029850746268656716, | |
| "grad_norm": 0.0, | |
| "learning_rate": 0.0, | |
| "loss": 0.8787, | |
| "step": 1 | |
| }, | |
| { | |
| "epoch": 0.208955223880597, | |
| "grad_norm": 3.4440434140026612, | |
| "learning_rate": 3.0000000000000004e-07, | |
| "loss": 0.8776, | |
| "step": 7 | |
| }, | |
| { | |
| "epoch": 0.208955223880597, | |
| "eval_loss": 0.7901861071586609, | |
| "eval_runtime": 150.9691, | |
| "eval_samples_per_second": 3.153, | |
| "eval_steps_per_second": 0.026, | |
| "step": 7 | |
| }, | |
| { | |
| "epoch": 0.417910447761194, | |
| "grad_norm": 2.901557012984361, | |
| "learning_rate": 1.8e-06, | |
| "loss": 0.8473, | |
| "step": 14 | |
| }, | |
| { | |
| "epoch": 0.417910447761194, | |
| "eval_loss": 0.7702628374099731, | |
| "eval_runtime": 143.7629, | |
| "eval_samples_per_second": 3.311, | |
| "eval_steps_per_second": 0.028, | |
| "step": 14 | |
| }, | |
| { | |
| "epoch": 0.6268656716417911, | |
| "grad_norm": 2.4942078074256164, | |
| "learning_rate": 2.5095609265912853e-06, | |
| "loss": 0.8293, | |
| "step": 21 | |
| }, | |
| { | |
| "epoch": 0.6268656716417911, | |
| "eval_loss": 0.760272204875946, | |
| "eval_runtime": 143.6002, | |
| "eval_samples_per_second": 3.315, | |
| "eval_steps_per_second": 0.028, | |
| "step": 21 | |
| }, | |
| { | |
| "epoch": 0.835820895522388, | |
| "grad_norm": 2.227024549115603, | |
| "learning_rate": 1.3197749551783641e-06, | |
| "loss": 0.8173, | |
| "step": 28 | |
| }, | |
| { | |
| "epoch": 0.835820895522388, | |
| "eval_loss": 0.7481057047843933, | |
| "eval_runtime": 139.8099, | |
| "eval_samples_per_second": 3.405, | |
| "eval_steps_per_second": 0.029, | |
| "step": 28 | |
| }, | |
| { | |
| "epoch": 1.044776119402985, | |
| "grad_norm": 1.890361948752719, | |
| "learning_rate": 6.783887430182062e-07, | |
| "loss": 0.7415, | |
| "step": 35 | |
| }, | |
| { | |
| "epoch": 1.044776119402985, | |
| "eval_loss": 0.7402028441429138, | |
| "eval_runtime": 139.4376, | |
| "eval_samples_per_second": 3.414, | |
| "eval_steps_per_second": 0.029, | |
| "step": 35 | |
| }, | |
| { | |
| "epoch": 1.2537313432835822, | |
| "grad_norm": 1.890868926771716, | |
| "learning_rate": 3.8102735091851235e-07, | |
| "loss": 0.6794, | |
| "step": 42 | |
| }, | |
| { | |
| "epoch": 1.2537313432835822, | |
| "eval_loss": 0.7418723106384277, | |
| "eval_runtime": 139.6461, | |
| "eval_samples_per_second": 3.409, | |
| "eval_steps_per_second": 0.029, | |
| "step": 42 | |
| }, | |
| { | |
| "epoch": 1.462686567164179, | |
| "grad_norm": 3.6823973563358674, | |
| "learning_rate": 1.9899658436440185e-07, | |
| "loss": 0.6688, | |
| "step": 49 | |
| }, | |
| { | |
| "epoch": 1.462686567164179, | |
| "eval_loss": 0.7392202615737915, | |
| "eval_runtime": 145.8028, | |
| "eval_samples_per_second": 3.265, | |
| "eval_steps_per_second": 0.027, | |
| "step": 49 | |
| }, | |
| { | |
| "epoch": 1.671641791044776, | |
| "grad_norm": 1.8056965838105696, | |
| "learning_rate": 1.1300091285551449e-07, | |
| "loss": 0.6498, | |
| "step": 56 | |
| }, | |
| { | |
| "epoch": 1.671641791044776, | |
| "eval_loss": 0.7367225289344788, | |
| "eval_runtime": 141.1059, | |
| "eval_samples_per_second": 3.373, | |
| "eval_steps_per_second": 0.028, | |
| "step": 56 | |
| }, | |
| { | |
| "epoch": 1.8805970149253732, | |
| "grad_norm": 1.5388706627418591, | |
| "learning_rate": 7.476064096023686e-08, | |
| "loss": 0.6701, | |
| "step": 63 | |
| }, | |
| { | |
| "epoch": 1.8805970149253732, | |
| "eval_loss": 0.7358315587043762, | |
| "eval_runtime": 143.703, | |
| "eval_samples_per_second": 3.312, | |
| "eval_steps_per_second": 0.028, | |
| "step": 63 | |
| }, | |
| { | |
| "epoch": 2.08955223880597, | |
| "grad_norm": 1.7131943205261349, | |
| "learning_rate": 5.89232146321995e-08, | |
| "loss": 0.664, | |
| "step": 70 | |
| }, | |
| { | |
| "epoch": 2.08955223880597, | |
| "eval_loss": 0.7354702353477478, | |
| "eval_runtime": 136.9065, | |
| "eval_samples_per_second": 3.477, | |
| "eval_steps_per_second": 0.029, | |
| "step": 70 | |
| }, | |
| { | |
| "epoch": 2.298507462686567, | |
| "grad_norm": 1.542023361934585, | |
| "learning_rate": 5.289674857255442e-08, | |
| "loss": 0.6447, | |
| "step": 77 | |
| }, | |
| { | |
| "epoch": 2.298507462686567, | |
| "eval_loss": 0.736127495765686, | |
| "eval_runtime": 136.173, | |
| "eval_samples_per_second": 3.496, | |
| "eval_steps_per_second": 0.029, | |
| "step": 77 | |
| }, | |
| { | |
| "epoch": 2.5074626865671643, | |
| "grad_norm": 2.5338510966510115, | |
| "learning_rate": 5.082712625717188e-08, | |
| "loss": 0.6412, | |
| "step": 84 | |
| }, | |
| { | |
| "epoch": 2.5074626865671643, | |
| "eval_loss": 0.7373142242431641, | |
| "eval_runtime": 138.002, | |
| "eval_samples_per_second": 3.449, | |
| "eval_steps_per_second": 0.029, | |
| "step": 84 | |
| }, | |
| { | |
| "epoch": 2.716417910447761, | |
| "grad_norm": 1.9360637903054263, | |
| "learning_rate": 5.020097212085352e-08, | |
| "loss": 0.6458, | |
| "step": 91 | |
| }, | |
| { | |
| "epoch": 2.716417910447761, | |
| "eval_loss": 0.7382717728614807, | |
| "eval_runtime": 139.7716, | |
| "eval_samples_per_second": 3.406, | |
| "eval_steps_per_second": 0.029, | |
| "step": 91 | |
| }, | |
| { | |
| "epoch": 2.925373134328358, | |
| "grad_norm": 2.733791738493569, | |
| "learning_rate": 5.0050722602692304e-08, | |
| "loss": 0.6356, | |
| "step": 98 | |
| }, | |
| { | |
| "epoch": 2.925373134328358, | |
| "eval_loss": 0.7387175559997559, | |
| "eval_runtime": 142.5673, | |
| "eval_samples_per_second": 3.339, | |
| "eval_steps_per_second": 0.028, | |
| "step": 98 | |
| }, | |
| { | |
| "epoch": 3.1343283582089554, | |
| "grad_norm": 1.3743623511988, | |
| "learning_rate": 5.001050931854095e-08, | |
| "loss": 0.6398, | |
| "step": 105 | |
| }, | |
| { | |
| "epoch": 3.1343283582089554, | |
| "eval_loss": 0.7387120723724365, | |
| "eval_runtime": 139.359, | |
| "eval_samples_per_second": 3.416, | |
| "eval_steps_per_second": 0.029, | |
| "step": 105 | |
| }, | |
| { | |
| "epoch": 3.343283582089552, | |
| "grad_norm": 1.5953948429140674, | |
| "learning_rate": 5.000119265172339e-08, | |
| "loss": 0.6228, | |
| "step": 112 | |
| }, | |
| { | |
| "epoch": 3.343283582089552, | |
| "eval_loss": 0.7390681505203247, | |
| "eval_runtime": 140.4569, | |
| "eval_samples_per_second": 3.389, | |
| "eval_steps_per_second": 0.028, | |
| "step": 112 | |
| }, | |
| { | |
| "epoch": 3.5522388059701493, | |
| "grad_norm": 1.8688576586340278, | |
| "learning_rate": 5.0000078923070654e-08, | |
| "loss": 0.6139, | |
| "step": 119 | |
| }, | |
| { | |
| "epoch": 3.5522388059701493, | |
| "eval_loss": 0.7394906282424927, | |
| "eval_runtime": 143.4431, | |
| "eval_samples_per_second": 3.318, | |
| "eval_steps_per_second": 0.028, | |
| "step": 119 | |
| }, | |
| { | |
| "epoch": 3.7611940298507465, | |
| "grad_norm": 3.2568458543215315, | |
| "learning_rate": 5.000000212746016e-08, | |
| "loss": 0.591, | |
| "step": 126 | |
| }, | |
| { | |
| "epoch": 3.7611940298507465, | |
| "eval_loss": 0.7398449778556824, | |
| "eval_runtime": 139.4002, | |
| "eval_samples_per_second": 3.415, | |
| "eval_steps_per_second": 0.029, | |
| "step": 126 | |
| }, | |
| { | |
| "epoch": 3.9402985074626864, | |
| "step": 132, | |
| "total_flos": 130172942548992.0, | |
| "train_loss": 0.6859688226020697, | |
| "train_runtime": 25080.2087, | |
| "train_samples_per_second": 0.683, | |
| "train_steps_per_second": 0.005 | |
| } | |
| ], | |
| "logging_steps": 7, | |
| "max_steps": 132, | |
| "num_input_tokens_seen": 0, | |
| "num_train_epochs": 4, | |
| "save_steps": 27, | |
| "total_flos": 130172942548992.0, | |
| "train_batch_size": 2, | |
| "trial_name": null, | |
| "trial_params": null | |
| } | |