| { | |
| "best_global_step": null, | |
| "best_metric": null, | |
| "best_model_checkpoint": null, | |
| "epoch": 3.964936886395512, | |
| "eval_steps": 100, | |
| "global_step": 356, | |
| "is_hyper_param_search": false, | |
| "is_local_process_zero": true, | |
| "is_world_process_zero": true, | |
| "log_history": [ | |
| { | |
| "epoch": 0.11220196353436185, | |
| "grad_norm": 1.1029870510101318, | |
| "learning_rate": 5e-05, | |
| "loss": 2.5659, | |
| "mean_token_accuracy": 0.47926636449992654, | |
| "num_tokens": 37367.0, | |
| "step": 10 | |
| }, | |
| { | |
| "epoch": 0.2244039270687237, | |
| "grad_norm": 2.981795310974121, | |
| "learning_rate": 0.00010555555555555557, | |
| "loss": 2.0395, | |
| "mean_token_accuracy": 0.5484138417989015, | |
| "num_tokens": 73517.0, | |
| "step": 20 | |
| }, | |
| { | |
| "epoch": 0.33660589060308554, | |
| "grad_norm": 0.5173138976097107, | |
| "learning_rate": 0.0001611111111111111, | |
| "loss": 1.5375, | |
| "mean_token_accuracy": 0.6410307548940182, | |
| "num_tokens": 109542.0, | |
| "step": 30 | |
| }, | |
| { | |
| "epoch": 0.4488078541374474, | |
| "grad_norm": 0.4973823130130768, | |
| "learning_rate": 0.000198125, | |
| "loss": 1.481, | |
| "mean_token_accuracy": 0.6483925141394138, | |
| "num_tokens": 145327.0, | |
| "step": 40 | |
| }, | |
| { | |
| "epoch": 0.5610098176718092, | |
| "grad_norm": 0.605583131313324, | |
| "learning_rate": 0.00019187500000000002, | |
| "loss": 1.3906, | |
| "mean_token_accuracy": 0.661427416652441, | |
| "num_tokens": 180932.0, | |
| "step": 50 | |
| }, | |
| { | |
| "epoch": 0.6732117812061711, | |
| "grad_norm": 0.5373088717460632, | |
| "learning_rate": 0.000185625, | |
| "loss": 1.3255, | |
| "mean_token_accuracy": 0.669424007833004, | |
| "num_tokens": 217215.0, | |
| "step": 60 | |
| }, | |
| { | |
| "epoch": 0.7854137447405329, | |
| "grad_norm": 0.5484739542007446, | |
| "learning_rate": 0.000179375, | |
| "loss": 1.315, | |
| "mean_token_accuracy": 0.6726369611918926, | |
| "num_tokens": 254200.0, | |
| "step": 70 | |
| }, | |
| { | |
| "epoch": 0.8976157082748948, | |
| "grad_norm": 0.5383081436157227, | |
| "learning_rate": 0.000173125, | |
| "loss": 1.2989, | |
| "mean_token_accuracy": 0.6707686208188534, | |
| "num_tokens": 290771.0, | |
| "step": 80 | |
| }, | |
| { | |
| "epoch": 1.0, | |
| "grad_norm": 2.1042652130126953, | |
| "learning_rate": 0.000166875, | |
| "loss": 1.353, | |
| "mean_token_accuracy": 0.6659621212580432, | |
| "num_tokens": 323338.0, | |
| "step": 90 | |
| }, | |
| { | |
| "epoch": 1.1122019635343618, | |
| "grad_norm": 0.5776804089546204, | |
| "learning_rate": 0.00016062500000000001, | |
| "loss": 1.2447, | |
| "mean_token_accuracy": 0.6804150842130184, | |
| "num_tokens": 360543.0, | |
| "step": 100 | |
| }, | |
| { | |
| "epoch": 1.2244039270687237, | |
| "grad_norm": 0.6110250949859619, | |
| "learning_rate": 0.000154375, | |
| "loss": 1.2908, | |
| "mean_token_accuracy": 0.6726092718541622, | |
| "num_tokens": 397295.0, | |
| "step": 110 | |
| }, | |
| { | |
| "epoch": 1.3366058906030855, | |
| "grad_norm": 0.5095610022544861, | |
| "learning_rate": 0.000148125, | |
| "loss": 1.2549, | |
| "mean_token_accuracy": 0.6794664606451988, | |
| "num_tokens": 434747.0, | |
| "step": 120 | |
| }, | |
| { | |
| "epoch": 1.4488078541374474, | |
| "grad_norm": 0.6921575665473938, | |
| "learning_rate": 0.000141875, | |
| "loss": 1.2156, | |
| "mean_token_accuracy": 0.6838382914662361, | |
| "num_tokens": 471034.0, | |
| "step": 130 | |
| }, | |
| { | |
| "epoch": 1.5610098176718092, | |
| "grad_norm": 0.6114996075630188, | |
| "learning_rate": 0.000135625, | |
| "loss": 1.2229, | |
| "mean_token_accuracy": 0.6844779521226882, | |
| "num_tokens": 507771.0, | |
| "step": 140 | |
| }, | |
| { | |
| "epoch": 1.673211781206171, | |
| "grad_norm": 0.7110440731048584, | |
| "learning_rate": 0.00012937500000000001, | |
| "loss": 1.1964, | |
| "mean_token_accuracy": 0.6940846800804138, | |
| "num_tokens": 543317.0, | |
| "step": 150 | |
| }, | |
| { | |
| "epoch": 1.785413744740533, | |
| "grad_norm": 0.6362674236297607, | |
| "learning_rate": 0.000123125, | |
| "loss": 1.2173, | |
| "mean_token_accuracy": 0.6877869322896004, | |
| "num_tokens": 579484.0, | |
| "step": 160 | |
| }, | |
| { | |
| "epoch": 1.8976157082748948, | |
| "grad_norm": 0.5968947410583496, | |
| "learning_rate": 0.000116875, | |
| "loss": 1.1781, | |
| "mean_token_accuracy": 0.6954927623271943, | |
| "num_tokens": 614929.0, | |
| "step": 170 | |
| }, | |
| { | |
| "epoch": 2.0, | |
| "grad_norm": 2.4199445247650146, | |
| "learning_rate": 0.000110625, | |
| "loss": 1.1646, | |
| "mean_token_accuracy": 0.6967783880560365, | |
| "num_tokens": 646676.0, | |
| "step": 180 | |
| }, | |
| { | |
| "epoch": 2.112201963534362, | |
| "grad_norm": 0.7061767578125, | |
| "learning_rate": 0.000104375, | |
| "loss": 1.1006, | |
| "mean_token_accuracy": 0.7113809175789356, | |
| "num_tokens": 682566.0, | |
| "step": 190 | |
| }, | |
| { | |
| "epoch": 2.2244039270687237, | |
| "grad_norm": 0.7149446606636047, | |
| "learning_rate": 9.8125e-05, | |
| "loss": 1.1557, | |
| "mean_token_accuracy": 0.6986343055963516, | |
| "num_tokens": 719538.0, | |
| "step": 200 | |
| }, | |
| { | |
| "epoch": 2.3366058906030855, | |
| "grad_norm": 0.7239488959312439, | |
| "learning_rate": 9.1875e-05, | |
| "loss": 1.1258, | |
| "mean_token_accuracy": 0.7038173094391823, | |
| "num_tokens": 756020.0, | |
| "step": 210 | |
| }, | |
| { | |
| "epoch": 2.4488078541374474, | |
| "grad_norm": 0.818574070930481, | |
| "learning_rate": 8.5625e-05, | |
| "loss": 1.0903, | |
| "mean_token_accuracy": 0.7100009344518184, | |
| "num_tokens": 791787.0, | |
| "step": 220 | |
| }, | |
| { | |
| "epoch": 2.5610098176718092, | |
| "grad_norm": 0.6714032292366028, | |
| "learning_rate": 7.9375e-05, | |
| "loss": 1.1066, | |
| "mean_token_accuracy": 0.7094940036535263, | |
| "num_tokens": 827575.0, | |
| "step": 230 | |
| }, | |
| { | |
| "epoch": 2.673211781206171, | |
| "grad_norm": 0.7658008933067322, | |
| "learning_rate": 7.3125e-05, | |
| "loss": 1.1222, | |
| "mean_token_accuracy": 0.7040122233331203, | |
| "num_tokens": 864519.0, | |
| "step": 240 | |
| }, | |
| { | |
| "epoch": 2.785413744740533, | |
| "grad_norm": 0.7659981846809387, | |
| "learning_rate": 6.6875e-05, | |
| "loss": 1.1106, | |
| "mean_token_accuracy": 0.7076892741024494, | |
| "num_tokens": 900626.0, | |
| "step": 250 | |
| }, | |
| { | |
| "epoch": 2.897615708274895, | |
| "grad_norm": 0.7530002593994141, | |
| "learning_rate": 6.0624999999999996e-05, | |
| "loss": 1.1112, | |
| "mean_token_accuracy": 0.7089583992958068, | |
| "num_tokens": 936428.0, | |
| "step": 260 | |
| }, | |
| { | |
| "epoch": 3.0, | |
| "grad_norm": 2.5998833179473877, | |
| "learning_rate": 5.4375e-05, | |
| "loss": 1.1442, | |
| "mean_token_accuracy": 0.7058552944496886, | |
| "num_tokens": 970014.0, | |
| "step": 270 | |
| }, | |
| { | |
| "epoch": 3.112201963534362, | |
| "grad_norm": 0.8477901220321655, | |
| "learning_rate": 4.8125000000000004e-05, | |
| "loss": 1.0422, | |
| "mean_token_accuracy": 0.7182675130665302, | |
| "num_tokens": 1005733.0, | |
| "step": 280 | |
| }, | |
| { | |
| "epoch": 3.2244039270687237, | |
| "grad_norm": 0.832841157913208, | |
| "learning_rate": 4.1875e-05, | |
| "loss": 1.0555, | |
| "mean_token_accuracy": 0.7220007188618183, | |
| "num_tokens": 1042301.0, | |
| "step": 290 | |
| }, | |
| { | |
| "epoch": 3.3366058906030855, | |
| "grad_norm": 0.855139970779419, | |
| "learning_rate": 3.5625000000000005e-05, | |
| "loss": 1.0527, | |
| "mean_token_accuracy": 0.7194023981690407, | |
| "num_tokens": 1079434.0, | |
| "step": 300 | |
| }, | |
| { | |
| "epoch": 3.4488078541374474, | |
| "grad_norm": 0.9042227864265442, | |
| "learning_rate": 2.9375000000000003e-05, | |
| "loss": 1.0098, | |
| "mean_token_accuracy": 0.7282766819000244, | |
| "num_tokens": 1114936.0, | |
| "step": 310 | |
| }, | |
| { | |
| "epoch": 3.5610098176718092, | |
| "grad_norm": 0.8887130618095398, | |
| "learning_rate": 2.3125000000000003e-05, | |
| "loss": 1.0106, | |
| "mean_token_accuracy": 0.7245936967432499, | |
| "num_tokens": 1151776.0, | |
| "step": 320 | |
| }, | |
| { | |
| "epoch": 3.673211781206171, | |
| "grad_norm": 0.8810842633247375, | |
| "learning_rate": 1.6875000000000004e-05, | |
| "loss": 1.0078, | |
| "mean_token_accuracy": 0.7305177293717862, | |
| "num_tokens": 1187017.0, | |
| "step": 330 | |
| }, | |
| { | |
| "epoch": 3.785413744740533, | |
| "grad_norm": 0.9642019867897034, | |
| "learning_rate": 1.0625e-05, | |
| "loss": 1.0294, | |
| "mean_token_accuracy": 0.7268829271197319, | |
| "num_tokens": 1223241.0, | |
| "step": 340 | |
| }, | |
| { | |
| "epoch": 3.897615708274895, | |
| "grad_norm": 0.8219223022460938, | |
| "learning_rate": 4.375e-06, | |
| "loss": 1.0295, | |
| "mean_token_accuracy": 0.7224250309169292, | |
| "num_tokens": 1260361.0, | |
| "step": 350 | |
| } | |
| ], | |
| "logging_steps": 10, | |
| "max_steps": 356, | |
| "num_input_tokens_seen": 0, | |
| "num_train_epochs": 4, | |
| "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": 7.817406737670144e+16, | |
| "train_batch_size": 4, | |
| "trial_name": null, | |
| "trial_params": null | |
| } | |