| { | |
| "best_metric": null, | |
| "best_model_checkpoint": null, | |
| "epoch": 1.8850574712643677, | |
| "eval_steps": 6, | |
| "global_step": 42, | |
| "is_hyper_param_search": false, | |
| "is_local_process_zero": true, | |
| "is_world_process_zero": true, | |
| "log_history": [ | |
| { | |
| "epoch": 0.04597701149425287, | |
| "grad_norm": 2.537923574447632, | |
| "learning_rate": 0.0001, | |
| "loss": 1.4151, | |
| "step": 1 | |
| }, | |
| { | |
| "epoch": 0.04597701149425287, | |
| "eval_loss": 0.6549283862113953, | |
| "eval_runtime": 59.8262, | |
| "eval_samples_per_second": 3.31, | |
| "eval_steps_per_second": 0.418, | |
| "step": 1 | |
| }, | |
| { | |
| "epoch": 0.09195402298850575, | |
| "grad_norm": 2.5883562564849854, | |
| "learning_rate": 0.0002, | |
| "loss": 1.3772, | |
| "step": 2 | |
| }, | |
| { | |
| "epoch": 0.13793103448275862, | |
| "grad_norm": 2.2641923427581787, | |
| "learning_rate": 0.0001996917333733128, | |
| "loss": 1.3153, | |
| "step": 3 | |
| }, | |
| { | |
| "epoch": 0.1839080459770115, | |
| "grad_norm": 5.304536819458008, | |
| "learning_rate": 0.00019876883405951377, | |
| "loss": 1.1035, | |
| "step": 4 | |
| }, | |
| { | |
| "epoch": 0.22988505747126436, | |
| "grad_norm": 1.074016809463501, | |
| "learning_rate": 0.00019723699203976766, | |
| "loss": 0.9059, | |
| "step": 5 | |
| }, | |
| { | |
| "epoch": 0.27586206896551724, | |
| "grad_norm": 0.8219733834266663, | |
| "learning_rate": 0.00019510565162951537, | |
| "loss": 0.8678, | |
| "step": 6 | |
| }, | |
| { | |
| "epoch": 0.27586206896551724, | |
| "eval_loss": 0.6351205706596375, | |
| "eval_runtime": 59.9835, | |
| "eval_samples_per_second": 3.301, | |
| "eval_steps_per_second": 0.417, | |
| "step": 6 | |
| }, | |
| { | |
| "epoch": 0.3218390804597701, | |
| "grad_norm": 1.478880524635315, | |
| "learning_rate": 0.0001923879532511287, | |
| "loss": 0.8782, | |
| "step": 7 | |
| }, | |
| { | |
| "epoch": 0.367816091954023, | |
| "grad_norm": 0.7757106423377991, | |
| "learning_rate": 0.0001891006524188368, | |
| "loss": 0.7993, | |
| "step": 8 | |
| }, | |
| { | |
| "epoch": 0.41379310344827586, | |
| "grad_norm": 0.5512004494667053, | |
| "learning_rate": 0.00018526401643540922, | |
| "loss": 0.7126, | |
| "step": 9 | |
| }, | |
| { | |
| "epoch": 0.45977011494252873, | |
| "grad_norm": 1.5733872652053833, | |
| "learning_rate": 0.00018090169943749476, | |
| "loss": 0.685, | |
| "step": 10 | |
| }, | |
| { | |
| "epoch": 0.5057471264367817, | |
| "grad_norm": 0.34443506598472595, | |
| "learning_rate": 0.0001760405965600031, | |
| "loss": 0.653, | |
| "step": 11 | |
| }, | |
| { | |
| "epoch": 0.5517241379310345, | |
| "grad_norm": 0.5645097494125366, | |
| "learning_rate": 0.00017071067811865476, | |
| "loss": 0.6408, | |
| "step": 12 | |
| }, | |
| { | |
| "epoch": 0.5517241379310345, | |
| "eval_loss": 0.6028693318367004, | |
| "eval_runtime": 59.4134, | |
| "eval_samples_per_second": 3.333, | |
| "eval_steps_per_second": 0.421, | |
| "step": 12 | |
| }, | |
| { | |
| "epoch": 0.5977011494252874, | |
| "grad_norm": 0.5011037588119507, | |
| "learning_rate": 0.00016494480483301836, | |
| "loss": 0.6326, | |
| "step": 13 | |
| }, | |
| { | |
| "epoch": 0.6436781609195402, | |
| "grad_norm": 0.2876424789428711, | |
| "learning_rate": 0.00015877852522924732, | |
| "loss": 0.6937, | |
| "step": 14 | |
| }, | |
| { | |
| "epoch": 0.6896551724137931, | |
| "grad_norm": 0.47212648391723633, | |
| "learning_rate": 0.0001522498564715949, | |
| "loss": 0.6298, | |
| "step": 15 | |
| }, | |
| { | |
| "epoch": 0.735632183908046, | |
| "grad_norm": 0.6836615800857544, | |
| "learning_rate": 0.00014539904997395468, | |
| "loss": 0.5569, | |
| "step": 16 | |
| }, | |
| { | |
| "epoch": 0.7816091954022989, | |
| "grad_norm": 0.26638293266296387, | |
| "learning_rate": 0.000138268343236509, | |
| "loss": 0.6057, | |
| "step": 17 | |
| }, | |
| { | |
| "epoch": 0.8275862068965517, | |
| "grad_norm": 0.5527795553207397, | |
| "learning_rate": 0.00013090169943749476, | |
| "loss": 0.6546, | |
| "step": 18 | |
| }, | |
| { | |
| "epoch": 0.8275862068965517, | |
| "eval_loss": 0.5754240155220032, | |
| "eval_runtime": 59.3448, | |
| "eval_samples_per_second": 3.336, | |
| "eval_steps_per_second": 0.421, | |
| "step": 18 | |
| }, | |
| { | |
| "epoch": 0.8735632183908046, | |
| "grad_norm": 0.3253306746482849, | |
| "learning_rate": 0.00012334453638559057, | |
| "loss": 0.5521, | |
| "step": 19 | |
| }, | |
| { | |
| "epoch": 0.9195402298850575, | |
| "grad_norm": 0.17234736680984497, | |
| "learning_rate": 0.0001156434465040231, | |
| "loss": 0.5802, | |
| "step": 20 | |
| }, | |
| { | |
| "epoch": 0.9655172413793104, | |
| "grad_norm": 0.2543017864227295, | |
| "learning_rate": 0.0001078459095727845, | |
| "loss": 0.5579, | |
| "step": 21 | |
| }, | |
| { | |
| "epoch": 1.0114942528735633, | |
| "grad_norm": 0.8507859706878662, | |
| "learning_rate": 0.0001, | |
| "loss": 0.5795, | |
| "step": 22 | |
| }, | |
| { | |
| "epoch": 1.0114942528735633, | |
| "grad_norm": 0.29268795251846313, | |
| "learning_rate": 9.215409042721552e-05, | |
| "loss": 0.6311, | |
| "step": 23 | |
| }, | |
| { | |
| "epoch": 1.0574712643678161, | |
| "grad_norm": 0.18724839389324188, | |
| "learning_rate": 8.435655349597689e-05, | |
| "loss": 0.574, | |
| "step": 24 | |
| }, | |
| { | |
| "epoch": 1.0574712643678161, | |
| "eval_loss": 0.5620841979980469, | |
| "eval_runtime": 59.4173, | |
| "eval_samples_per_second": 3.332, | |
| "eval_steps_per_second": 0.421, | |
| "step": 24 | |
| }, | |
| { | |
| "epoch": 1.103448275862069, | |
| "grad_norm": 0.21775011718273163, | |
| "learning_rate": 7.66554636144095e-05, | |
| "loss": 0.5818, | |
| "step": 25 | |
| }, | |
| { | |
| "epoch": 1.1494252873563218, | |
| "grad_norm": 0.24518676102161407, | |
| "learning_rate": 6.909830056250527e-05, | |
| "loss": 0.5618, | |
| "step": 26 | |
| }, | |
| { | |
| "epoch": 1.1954022988505748, | |
| "grad_norm": 0.24137045443058014, | |
| "learning_rate": 6.173165676349103e-05, | |
| "loss": 0.6078, | |
| "step": 27 | |
| }, | |
| { | |
| "epoch": 1.2413793103448276, | |
| "grad_norm": 0.17265664041042328, | |
| "learning_rate": 5.4600950026045326e-05, | |
| "loss": 0.5907, | |
| "step": 28 | |
| }, | |
| { | |
| "epoch": 1.2873563218390804, | |
| "grad_norm": 0.1558091938495636, | |
| "learning_rate": 4.7750143528405126e-05, | |
| "loss": 0.553, | |
| "step": 29 | |
| }, | |
| { | |
| "epoch": 1.3333333333333333, | |
| "grad_norm": 0.15223024785518646, | |
| "learning_rate": 4.12214747707527e-05, | |
| "loss": 0.5518, | |
| "step": 30 | |
| }, | |
| { | |
| "epoch": 1.3333333333333333, | |
| "eval_loss": 0.5595065951347351, | |
| "eval_runtime": 59.4325, | |
| "eval_samples_per_second": 3.332, | |
| "eval_steps_per_second": 0.421, | |
| "step": 30 | |
| }, | |
| { | |
| "epoch": 1.3793103448275863, | |
| "grad_norm": 0.25727611780166626, | |
| "learning_rate": 3.5055195166981645e-05, | |
| "loss": 0.5463, | |
| "step": 31 | |
| }, | |
| { | |
| "epoch": 1.4252873563218391, | |
| "grad_norm": 0.13202741742134094, | |
| "learning_rate": 2.9289321881345254e-05, | |
| "loss": 0.5598, | |
| "step": 32 | |
| }, | |
| { | |
| "epoch": 1.471264367816092, | |
| "grad_norm": 0.20626722276210785, | |
| "learning_rate": 2.3959403439996907e-05, | |
| "loss": 0.5242, | |
| "step": 33 | |
| }, | |
| { | |
| "epoch": 1.5172413793103448, | |
| "grad_norm": 0.16128551959991455, | |
| "learning_rate": 1.9098300562505266e-05, | |
| "loss": 0.5715, | |
| "step": 34 | |
| }, | |
| { | |
| "epoch": 1.5632183908045976, | |
| "grad_norm": 0.346206396818161, | |
| "learning_rate": 1.4735983564590783e-05, | |
| "loss": 0.5265, | |
| "step": 35 | |
| }, | |
| { | |
| "epoch": 1.6091954022988506, | |
| "grad_norm": 0.1324978768825531, | |
| "learning_rate": 1.0899347581163221e-05, | |
| "loss": 0.5446, | |
| "step": 36 | |
| }, | |
| { | |
| "epoch": 1.6091954022988506, | |
| "eval_loss": 0.5576122403144836, | |
| "eval_runtime": 59.5124, | |
| "eval_samples_per_second": 3.327, | |
| "eval_steps_per_second": 0.42, | |
| "step": 36 | |
| }, | |
| { | |
| "epoch": 1.6551724137931034, | |
| "grad_norm": 0.1254015415906906, | |
| "learning_rate": 7.612046748871327e-06, | |
| "loss": 0.5739, | |
| "step": 37 | |
| }, | |
| { | |
| "epoch": 1.7011494252873565, | |
| "grad_norm": 0.18170268833637238, | |
| "learning_rate": 4.8943483704846475e-06, | |
| "loss": 0.5717, | |
| "step": 38 | |
| }, | |
| { | |
| "epoch": 1.7471264367816093, | |
| "grad_norm": 0.12662401795387268, | |
| "learning_rate": 2.7630079602323442e-06, | |
| "loss": 0.5424, | |
| "step": 39 | |
| }, | |
| { | |
| "epoch": 1.793103448275862, | |
| "grad_norm": 0.1479385495185852, | |
| "learning_rate": 1.231165940486234e-06, | |
| "loss": 0.563, | |
| "step": 40 | |
| }, | |
| { | |
| "epoch": 1.839080459770115, | |
| "grad_norm": 0.13152529299259186, | |
| "learning_rate": 3.0826662668720364e-07, | |
| "loss": 0.5698, | |
| "step": 41 | |
| }, | |
| { | |
| "epoch": 1.8850574712643677, | |
| "grad_norm": 0.17284046113491058, | |
| "learning_rate": 0.0, | |
| "loss": 0.6227, | |
| "step": 42 | |
| }, | |
| { | |
| "epoch": 1.8850574712643677, | |
| "eval_loss": 0.5568149089813232, | |
| "eval_runtime": 59.8384, | |
| "eval_samples_per_second": 3.309, | |
| "eval_steps_per_second": 0.418, | |
| "step": 42 | |
| } | |
| ], | |
| "logging_steps": 1, | |
| "max_steps": 42, | |
| "num_input_tokens_seen": 0, | |
| "num_train_epochs": 2, | |
| "save_steps": 21, | |
| "stateful_callbacks": { | |
| "TrainerControl": { | |
| "args": { | |
| "should_epoch_stop": false, | |
| "should_evaluate": false, | |
| "should_log": false, | |
| "should_save": true, | |
| "should_training_stop": true | |
| }, | |
| "attributes": {} | |
| } | |
| }, | |
| "total_flos": 2.3746008314177126e+17, | |
| "train_batch_size": 2, | |
| "trial_name": null, | |
| "trial_params": null | |
| } | |