| { | |
| "best_metric": null, | |
| "best_model_checkpoint": null, | |
| "epoch": 2.9802197802197803, | |
| "eval_steps": 100, | |
| "global_step": 339, | |
| "is_hyper_param_search": false, | |
| "is_local_process_zero": true, | |
| "is_world_process_zero": true, | |
| "log_history": [ | |
| { | |
| "epoch": 0.08791208791208792, | |
| "grad_norm": 27.515762329101562, | |
| "learning_rate": 1.4705882352941177e-05, | |
| "loss": 3.1401, | |
| "step": 10 | |
| }, | |
| { | |
| "epoch": 0.17582417582417584, | |
| "grad_norm": 6.303249835968018, | |
| "learning_rate": 4.411764705882353e-05, | |
| "loss": 2.7886, | |
| "step": 20 | |
| }, | |
| { | |
| "epoch": 0.26373626373626374, | |
| "grad_norm": 3.689723014831543, | |
| "learning_rate": 7.352941176470589e-05, | |
| "loss": 2.2838, | |
| "step": 30 | |
| }, | |
| { | |
| "epoch": 0.3516483516483517, | |
| "grad_norm": 3.294996500015259, | |
| "learning_rate": 9.999734761700061e-05, | |
| "loss": 2.0502, | |
| "step": 40 | |
| }, | |
| { | |
| "epoch": 0.43956043956043955, | |
| "grad_norm": 2.8345844745635986, | |
| "learning_rate": 9.967940201677627e-05, | |
| "loss": 1.9329, | |
| "step": 50 | |
| }, | |
| { | |
| "epoch": 0.5274725274725275, | |
| "grad_norm": 2.3837406635284424, | |
| "learning_rate": 9.883484239638841e-05, | |
| "loss": 1.9189, | |
| "step": 60 | |
| }, | |
| { | |
| "epoch": 0.6153846153846154, | |
| "grad_norm": 2.6968302726745605, | |
| "learning_rate": 9.747262129791496e-05, | |
| "loss": 1.9288, | |
| "step": 70 | |
| }, | |
| { | |
| "epoch": 0.7032967032967034, | |
| "grad_norm": 2.5380537509918213, | |
| "learning_rate": 9.560717860388061e-05, | |
| "loss": 1.7928, | |
| "step": 80 | |
| }, | |
| { | |
| "epoch": 0.7912087912087912, | |
| "grad_norm": 2.101832866668701, | |
| "learning_rate": 9.32582884709098e-05, | |
| "loss": 1.7827, | |
| "step": 90 | |
| }, | |
| { | |
| "epoch": 0.8791208791208791, | |
| "grad_norm": 2.518181085586548, | |
| "learning_rate": 9.045084971874738e-05, | |
| "loss": 1.7937, | |
| "step": 100 | |
| }, | |
| { | |
| "epoch": 0.8791208791208791, | |
| "eval_loss": 1.7743221521377563, | |
| "eval_runtime": 50.537, | |
| "eval_samples_per_second": 1.009, | |
| "eval_steps_per_second": 1.009, | |
| "step": 100 | |
| }, | |
| { | |
| "epoch": 0.967032967032967, | |
| "grad_norm": 2.1128666400909424, | |
| "learning_rate": 8.72146218965751e-05, | |
| "loss": 1.7837, | |
| "step": 110 | |
| }, | |
| { | |
| "epoch": 1.054945054945055, | |
| "grad_norm": 1.658250331878662, | |
| "learning_rate": 8.358390982438706e-05, | |
| "loss": 1.6638, | |
| "step": 120 | |
| }, | |
| { | |
| "epoch": 1.1428571428571428, | |
| "grad_norm": 2.094742774963379, | |
| "learning_rate": 7.959719995336364e-05, | |
| "loss": 1.5837, | |
| "step": 130 | |
| }, | |
| { | |
| "epoch": 1.2307692307692308, | |
| "grad_norm": 2.0394861698150635, | |
| "learning_rate": 7.529675239991482e-05, | |
| "loss": 1.6425, | |
| "step": 140 | |
| }, | |
| { | |
| "epoch": 1.3186813186813187, | |
| "grad_norm": 1.947106122970581, | |
| "learning_rate": 7.072815297793303e-05, | |
| "loss": 1.5648, | |
| "step": 150 | |
| }, | |
| { | |
| "epoch": 1.4065934065934065, | |
| "grad_norm": 2.5276968479156494, | |
| "learning_rate": 6.593982997782549e-05, | |
| "loss": 1.6034, | |
| "step": 160 | |
| }, | |
| { | |
| "epoch": 1.4945054945054945, | |
| "grad_norm": 2.3803257942199707, | |
| "learning_rate": 6.0982540814588384e-05, | |
| "loss": 1.5715, | |
| "step": 170 | |
| }, | |
| { | |
| "epoch": 1.5824175824175826, | |
| "grad_norm": 2.356159210205078, | |
| "learning_rate": 5.590883398658095e-05, | |
| "loss": 1.5342, | |
| "step": 180 | |
| }, | |
| { | |
| "epoch": 1.6703296703296702, | |
| "grad_norm": 2.287046194076538, | |
| "learning_rate": 5.077249204837026e-05, | |
| "loss": 1.5677, | |
| "step": 190 | |
| }, | |
| { | |
| "epoch": 1.7582417582417582, | |
| "grad_norm": 2.1357710361480713, | |
| "learning_rate": 4.562796150227338e-05, | |
| "loss": 1.6093, | |
| "step": 200 | |
| }, | |
| { | |
| "epoch": 1.7582417582417582, | |
| "eval_loss": 1.7231701612472534, | |
| "eval_runtime": 50.4704, | |
| "eval_samples_per_second": 1.01, | |
| "eval_steps_per_second": 1.01, | |
| "step": 200 | |
| }, | |
| { | |
| "epoch": 1.8461538461538463, | |
| "grad_norm": 2.670072555541992, | |
| "learning_rate": 4.0529775651888e-05, | |
| "loss": 1.5564, | |
| "step": 210 | |
| }, | |
| { | |
| "epoch": 1.934065934065934, | |
| "grad_norm": 2.400627851486206, | |
| "learning_rate": 3.5531976535508106e-05, | |
| "loss": 1.5771, | |
| "step": 220 | |
| }, | |
| { | |
| "epoch": 2.021978021978022, | |
| "grad_norm": 1.9662156105041504, | |
| "learning_rate": 3.068754206707392e-05, | |
| "loss": 1.4677, | |
| "step": 230 | |
| }, | |
| { | |
| "epoch": 2.10989010989011, | |
| "grad_norm": 2.1746368408203125, | |
| "learning_rate": 2.6047824457104763e-05, | |
| "loss": 1.3628, | |
| "step": 240 | |
| }, | |
| { | |
| "epoch": 2.197802197802198, | |
| "grad_norm": 2.274919271469116, | |
| "learning_rate": 2.1662005866492717e-05, | |
| "loss": 1.3521, | |
| "step": 250 | |
| }, | |
| { | |
| "epoch": 2.2857142857142856, | |
| "grad_norm": 2.1153922080993652, | |
| "learning_rate": 1.757657706336192e-05, | |
| "loss": 1.3069, | |
| "step": 260 | |
| }, | |
| { | |
| "epoch": 2.3736263736263736, | |
| "grad_norm": 2.8658852577209473, | |
| "learning_rate": 1.3834844609361064e-05, | |
| "loss": 1.297, | |
| "step": 270 | |
| }, | |
| { | |
| "epoch": 2.4615384615384617, | |
| "grad_norm": 2.429492235183716, | |
| "learning_rate": 1.0476471799336423e-05, | |
| "loss": 1.2933, | |
| "step": 280 | |
| }, | |
| { | |
| "epoch": 2.5494505494505493, | |
| "grad_norm": 2.1286540031433105, | |
| "learning_rate": 7.537058220539395e-06, | |
| "loss": 1.2501, | |
| "step": 290 | |
| }, | |
| { | |
| "epoch": 2.6373626373626373, | |
| "grad_norm": 2.603133201599121, | |
| "learning_rate": 5.047762388145583e-06, | |
| "loss": 1.3351, | |
| "step": 300 | |
| }, | |
| { | |
| "epoch": 2.6373626373626373, | |
| "eval_loss": 1.7396082878112793, | |
| "eval_runtime": 50.5148, | |
| "eval_samples_per_second": 1.01, | |
| "eval_steps_per_second": 1.01, | |
| "step": 300 | |
| }, | |
| { | |
| "epoch": 2.7252747252747254, | |
| "grad_norm": 2.373342275619507, | |
| "learning_rate": 3.034971457242525e-06, | |
| "loss": 1.3897, | |
| "step": 310 | |
| }, | |
| { | |
| "epoch": 2.813186813186813, | |
| "grad_norm": 2.549192190170288, | |
| "learning_rate": 1.520021512421449e-06, | |
| "loss": 1.2277, | |
| "step": 320 | |
| }, | |
| { | |
| "epoch": 2.901098901098901, | |
| "grad_norm": 2.8294496536254883, | |
| "learning_rate": 5.189713999728008e-07, | |
| "loss": 1.3033, | |
| "step": 330 | |
| } | |
| ], | |
| "logging_steps": 10, | |
| "max_steps": 339, | |
| "num_input_tokens_seen": 0, | |
| "num_train_epochs": 3, | |
| "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": 6.233392759268966e+16, | |
| "train_batch_size": 1, | |
| "trial_name": null, | |
| "trial_params": null | |
| } | |