| { | |
| "best_global_step": null, | |
| "best_metric": null, | |
| "best_model_checkpoint": null, | |
| "epoch": 12.470588235294118, | |
| "eval_steps": 500, | |
| "global_step": 112, | |
| "is_hyper_param_search": false, | |
| "is_local_process_zero": true, | |
| "is_world_process_zero": true, | |
| "log_history": [ | |
| { | |
| "epoch": 0.5882352941176471, | |
| "grad_norm": 0.8250285070833755, | |
| "learning_rate": 2.0833333333333336e-05, | |
| "loss": 1.1514, | |
| "mean_token_accuracy": 0.7545319080352784, | |
| "num_tokens": 2505918.0, | |
| "step": 5 | |
| }, | |
| { | |
| "epoch": 1.1176470588235294, | |
| "grad_norm": 0.769774425261519, | |
| "learning_rate": 4.166666666666667e-05, | |
| "loss": 1.0097, | |
| "mean_token_accuracy": 0.7732286784383986, | |
| "num_tokens": 4791423.0, | |
| "step": 10 | |
| }, | |
| { | |
| "epoch": 1.7058823529411766, | |
| "grad_norm": 0.38521393276298865, | |
| "learning_rate": 4.85e-05, | |
| "loss": 0.8391, | |
| "mean_token_accuracy": 0.806722441315651, | |
| "num_tokens": 7330272.0, | |
| "step": 15 | |
| }, | |
| { | |
| "epoch": 2.235294117647059, | |
| "grad_norm": 0.3989534309948224, | |
| "learning_rate": 4.600000000000001e-05, | |
| "loss": 0.7915, | |
| "mean_token_accuracy": 0.8170362280474769, | |
| "num_tokens": 9618952.0, | |
| "step": 20 | |
| }, | |
| { | |
| "epoch": 2.8235294117647056, | |
| "grad_norm": 0.35223023869070996, | |
| "learning_rate": 4.35e-05, | |
| "loss": 0.6523, | |
| "mean_token_accuracy": 0.8425902664661408, | |
| "num_tokens": 12114377.0, | |
| "step": 25 | |
| }, | |
| { | |
| "epoch": 3.3529411764705883, | |
| "grad_norm": 0.3419690748616924, | |
| "learning_rate": 4.1e-05, | |
| "loss": 0.6196, | |
| "mean_token_accuracy": 0.8515104750792185, | |
| "num_tokens": 14409009.0, | |
| "step": 30 | |
| }, | |
| { | |
| "epoch": 3.9411764705882355, | |
| "grad_norm": 1.0636105938490006, | |
| "learning_rate": 3.85e-05, | |
| "loss": 0.5473, | |
| "mean_token_accuracy": 0.8684158384799957, | |
| "num_tokens": 16918082.0, | |
| "step": 35 | |
| }, | |
| { | |
| "epoch": 4.470588235294118, | |
| "grad_norm": 0.8104215493367165, | |
| "learning_rate": 3.6e-05, | |
| "loss": 0.5313, | |
| "mean_token_accuracy": 0.8752750555674235, | |
| "num_tokens": 19217454.0, | |
| "step": 40 | |
| }, | |
| { | |
| "epoch": 5.0, | |
| "grad_norm": 0.4155095547714891, | |
| "learning_rate": 3.35e-05, | |
| "loss": 0.3995, | |
| "mean_token_accuracy": 0.8990385002560086, | |
| "num_tokens": 21475135.0, | |
| "step": 45 | |
| }, | |
| { | |
| "epoch": 5.588235294117647, | |
| "grad_norm": 0.38830329565569643, | |
| "learning_rate": 3.1e-05, | |
| "loss": 0.3932, | |
| "mean_token_accuracy": 0.9030953556299209, | |
| "num_tokens": 23999790.0, | |
| "step": 50 | |
| }, | |
| { | |
| "epoch": 6.117647058823529, | |
| "grad_norm": 0.6003093985868039, | |
| "learning_rate": 2.8499999999999998e-05, | |
| "loss": 0.3304, | |
| "mean_token_accuracy": 0.9205087688234117, | |
| "num_tokens": 26264984.0, | |
| "step": 55 | |
| }, | |
| { | |
| "epoch": 6.705882352941177, | |
| "grad_norm": 0.8142075697801212, | |
| "learning_rate": 2.6000000000000002e-05, | |
| "loss": 0.2441, | |
| "mean_token_accuracy": 0.9367312312126159, | |
| "num_tokens": 28792279.0, | |
| "step": 60 | |
| }, | |
| { | |
| "epoch": 7.235294117647059, | |
| "grad_norm": 0.44738871439201344, | |
| "learning_rate": 2.35e-05, | |
| "loss": 0.3344, | |
| "mean_token_accuracy": 0.9259601334730784, | |
| "num_tokens": 31085796.0, | |
| "step": 65 | |
| }, | |
| { | |
| "epoch": 7.823529411764706, | |
| "grad_norm": 0.5170032581958464, | |
| "learning_rate": 2.1e-05, | |
| "loss": 0.2198, | |
| "mean_token_accuracy": 0.9499045610427856, | |
| "num_tokens": 33579925.0, | |
| "step": 70 | |
| }, | |
| { | |
| "epoch": 8.352941176470589, | |
| "grad_norm": 0.41980493493671023, | |
| "learning_rate": 1.85e-05, | |
| "loss": 0.243, | |
| "mean_token_accuracy": 0.9482089314195845, | |
| "num_tokens": 35872214.0, | |
| "step": 75 | |
| }, | |
| { | |
| "epoch": 8.941176470588236, | |
| "grad_norm": 1.2078125761983167, | |
| "learning_rate": 1.6000000000000003e-05, | |
| "loss": 0.1887, | |
| "mean_token_accuracy": 0.9552906274795532, | |
| "num_tokens": 38392486.0, | |
| "step": 80 | |
| }, | |
| { | |
| "epoch": 9.470588235294118, | |
| "grad_norm": 0.4806987993157492, | |
| "learning_rate": 1.3500000000000001e-05, | |
| "loss": 0.1776, | |
| "mean_token_accuracy": 0.9594514999124739, | |
| "num_tokens": 40676022.0, | |
| "step": 85 | |
| }, | |
| { | |
| "epoch": 10.0, | |
| "grad_norm": 0.626687130036329, | |
| "learning_rate": 1.1000000000000001e-05, | |
| "loss": 0.1525, | |
| "mean_token_accuracy": 0.9653611448076036, | |
| "num_tokens": 42948208.0, | |
| "step": 90 | |
| }, | |
| { | |
| "epoch": 10.588235294117647, | |
| "grad_norm": 0.3484261346541397, | |
| "learning_rate": 8.500000000000002e-06, | |
| "loss": 0.1488, | |
| "mean_token_accuracy": 0.9670574784278869, | |
| "num_tokens": 45475525.0, | |
| "step": 95 | |
| }, | |
| { | |
| "epoch": 11.117647058823529, | |
| "grad_norm": 0.43421046707207084, | |
| "learning_rate": 6e-06, | |
| "loss": 0.1424, | |
| "mean_token_accuracy": 0.9697046114338769, | |
| "num_tokens": 47733376.0, | |
| "step": 100 | |
| }, | |
| { | |
| "epoch": 11.705882352941176, | |
| "grad_norm": 0.24969104543475654, | |
| "learning_rate": 3.5000000000000004e-06, | |
| "loss": 0.1501, | |
| "mean_token_accuracy": 0.9680212080478668, | |
| "num_tokens": 50275121.0, | |
| "step": 105 | |
| }, | |
| { | |
| "epoch": 12.235294117647058, | |
| "grad_norm": 0.1876144299655879, | |
| "learning_rate": 1.0000000000000002e-06, | |
| "loss": 0.0969, | |
| "mean_token_accuracy": 0.9780076709058549, | |
| "num_tokens": 52549898.0, | |
| "step": 110 | |
| }, | |
| { | |
| "epoch": 12.470588235294118, | |
| "mean_token_accuracy": 0.9807853102684021, | |
| "num_tokens": 53548989.0, | |
| "step": 112, | |
| "total_flos": 80266967384064.0, | |
| "train_loss": 0.4191697733476758, | |
| "train_runtime": 1031.2092, | |
| "train_samples_per_second": 7.318, | |
| "train_steps_per_second": 0.109 | |
| } | |
| ], | |
| "logging_steps": 5, | |
| "max_steps": 112, | |
| "num_input_tokens_seen": 0, | |
| "num_train_epochs": 14, | |
| "save_steps": 500, | |
| "stateful_callbacks": { | |
| "TrainerControl": { | |
| "args": { | |
| "should_epoch_stop": false, | |
| "should_evaluate": false, | |
| "should_log": false, | |
| "should_save": true, | |
| "should_training_stop": true | |
| }, | |
| "attributes": {} | |
| } | |
| }, | |
| "total_flos": 80266967384064.0, | |
| "train_batch_size": 2, | |
| "trial_name": null, | |
| "trial_params": null | |
| } | |