| { | |
| "best_metric": null, | |
| "best_model_checkpoint": null, | |
| "epoch": 0.01204202667308908, | |
| "eval_steps": 9, | |
| "global_step": 100, | |
| "is_hyper_param_search": false, | |
| "is_local_process_zero": true, | |
| "is_world_process_zero": true, | |
| "log_history": [ | |
| { | |
| "epoch": 0.00012042026673089081, | |
| "eval_loss": 2.8115124702453613, | |
| "eval_runtime": 1968.0615, | |
| "eval_samples_per_second": 7.107, | |
| "eval_steps_per_second": 0.889, | |
| "step": 1 | |
| }, | |
| { | |
| "epoch": 0.00036126080019267245, | |
| "grad_norm": 41.35398864746094, | |
| "learning_rate": 3e-05, | |
| "loss": 11.0864, | |
| "step": 3 | |
| }, | |
| { | |
| "epoch": 0.0007225216003853449, | |
| "grad_norm": 32.59439468383789, | |
| "learning_rate": 6e-05, | |
| "loss": 10.0289, | |
| "step": 6 | |
| }, | |
| { | |
| "epoch": 0.0010837824005780173, | |
| "grad_norm": 23.11261749267578, | |
| "learning_rate": 9e-05, | |
| "loss": 9.0931, | |
| "step": 9 | |
| }, | |
| { | |
| "epoch": 0.0010837824005780173, | |
| "eval_loss": 2.0141799449920654, | |
| "eval_runtime": 1984.0515, | |
| "eval_samples_per_second": 7.05, | |
| "eval_steps_per_second": 0.882, | |
| "step": 9 | |
| }, | |
| { | |
| "epoch": 0.0014450432007706898, | |
| "grad_norm": 25.98846435546875, | |
| "learning_rate": 9.987820251299122e-05, | |
| "loss": 7.6576, | |
| "step": 12 | |
| }, | |
| { | |
| "epoch": 0.001806304000963362, | |
| "grad_norm": 32.36982727050781, | |
| "learning_rate": 9.924038765061042e-05, | |
| "loss": 7.2228, | |
| "step": 15 | |
| }, | |
| { | |
| "epoch": 0.0021675648011560346, | |
| "grad_norm": 16.200027465820312, | |
| "learning_rate": 9.806308479691595e-05, | |
| "loss": 6.5358, | |
| "step": 18 | |
| }, | |
| { | |
| "epoch": 0.0021675648011560346, | |
| "eval_loss": 1.6316649913787842, | |
| "eval_runtime": 1985.2556, | |
| "eval_samples_per_second": 7.045, | |
| "eval_steps_per_second": 0.881, | |
| "step": 18 | |
| }, | |
| { | |
| "epoch": 0.002528825601348707, | |
| "grad_norm": 18.970165252685547, | |
| "learning_rate": 9.635919272833938e-05, | |
| "loss": 6.6555, | |
| "step": 21 | |
| }, | |
| { | |
| "epoch": 0.0028900864015413796, | |
| "grad_norm": 15.357333183288574, | |
| "learning_rate": 9.414737964294636e-05, | |
| "loss": 6.3444, | |
| "step": 24 | |
| }, | |
| { | |
| "epoch": 0.0032513472017340517, | |
| "grad_norm": 15.496698379516602, | |
| "learning_rate": 9.145187862775209e-05, | |
| "loss": 5.7976, | |
| "step": 27 | |
| }, | |
| { | |
| "epoch": 0.0032513472017340517, | |
| "eval_loss": 1.5133966207504272, | |
| "eval_runtime": 1986.8922, | |
| "eval_samples_per_second": 7.04, | |
| "eval_steps_per_second": 0.88, | |
| "step": 27 | |
| }, | |
| { | |
| "epoch": 0.003612608001926724, | |
| "grad_norm": 13.662860870361328, | |
| "learning_rate": 8.83022221559489e-05, | |
| "loss": 6.0724, | |
| "step": 30 | |
| }, | |
| { | |
| "epoch": 0.003973868802119396, | |
| "grad_norm": 15.459967613220215, | |
| "learning_rate": 8.473291852294987e-05, | |
| "loss": 6.2818, | |
| "step": 33 | |
| }, | |
| { | |
| "epoch": 0.004335129602312069, | |
| "grad_norm": 11.845706939697266, | |
| "learning_rate": 8.07830737662829e-05, | |
| "loss": 5.7472, | |
| "step": 36 | |
| }, | |
| { | |
| "epoch": 0.004335129602312069, | |
| "eval_loss": 1.4950615167617798, | |
| "eval_runtime": 1983.0961, | |
| "eval_samples_per_second": 7.053, | |
| "eval_steps_per_second": 0.882, | |
| "step": 36 | |
| }, | |
| { | |
| "epoch": 0.004696390402504741, | |
| "grad_norm": 16.0750675201416, | |
| "learning_rate": 7.649596321166024e-05, | |
| "loss": 5.9232, | |
| "step": 39 | |
| }, | |
| { | |
| "epoch": 0.005057651202697414, | |
| "grad_norm": 15.54041576385498, | |
| "learning_rate": 7.191855733945387e-05, | |
| "loss": 5.7156, | |
| "step": 42 | |
| }, | |
| { | |
| "epoch": 0.005418912002890086, | |
| "grad_norm": 13.875934600830078, | |
| "learning_rate": 6.710100716628344e-05, | |
| "loss": 5.7691, | |
| "step": 45 | |
| }, | |
| { | |
| "epoch": 0.005418912002890086, | |
| "eval_loss": 1.454808235168457, | |
| "eval_runtime": 1987.0553, | |
| "eval_samples_per_second": 7.039, | |
| "eval_steps_per_second": 0.88, | |
| "step": 45 | |
| }, | |
| { | |
| "epoch": 0.005780172803082759, | |
| "grad_norm": 13.306793212890625, | |
| "learning_rate": 6.209609477998338e-05, | |
| "loss": 6.0942, | |
| "step": 48 | |
| }, | |
| { | |
| "epoch": 0.006141433603275431, | |
| "grad_norm": 13.40745735168457, | |
| "learning_rate": 5.695865504800327e-05, | |
| "loss": 5.8127, | |
| "step": 51 | |
| }, | |
| { | |
| "epoch": 0.006502694403468103, | |
| "grad_norm": 13.341890335083008, | |
| "learning_rate": 5.174497483512506e-05, | |
| "loss": 5.6365, | |
| "step": 54 | |
| }, | |
| { | |
| "epoch": 0.006502694403468103, | |
| "eval_loss": 1.4377180337905884, | |
| "eval_runtime": 1988.1753, | |
| "eval_samples_per_second": 7.035, | |
| "eval_steps_per_second": 0.88, | |
| "step": 54 | |
| }, | |
| { | |
| "epoch": 0.006863955203660776, | |
| "grad_norm": 14.141239166259766, | |
| "learning_rate": 4.6512176312793736e-05, | |
| "loss": 5.6544, | |
| "step": 57 | |
| }, | |
| { | |
| "epoch": 0.007225216003853448, | |
| "grad_norm": 13.24157428741455, | |
| "learning_rate": 4.131759111665349e-05, | |
| "loss": 5.7787, | |
| "step": 60 | |
| }, | |
| { | |
| "epoch": 0.007586476804046121, | |
| "grad_norm": 13.354693412780762, | |
| "learning_rate": 3.6218132209150045e-05, | |
| "loss": 5.9927, | |
| "step": 63 | |
| }, | |
| { | |
| "epoch": 0.007586476804046121, | |
| "eval_loss": 1.4332555532455444, | |
| "eval_runtime": 1985.8578, | |
| "eval_samples_per_second": 7.043, | |
| "eval_steps_per_second": 0.881, | |
| "step": 63 | |
| }, | |
| { | |
| "epoch": 0.007947737604238793, | |
| "grad_norm": 12.176066398620605, | |
| "learning_rate": 3.12696703292044e-05, | |
| "loss": 5.5682, | |
| "step": 66 | |
| }, | |
| { | |
| "epoch": 0.008308998404431465, | |
| "grad_norm": 15.984620094299316, | |
| "learning_rate": 2.6526421860705473e-05, | |
| "loss": 5.8425, | |
| "step": 69 | |
| }, | |
| { | |
| "epoch": 0.008670259204624138, | |
| "grad_norm": 13.588284492492676, | |
| "learning_rate": 2.2040354826462668e-05, | |
| "loss": 6.1731, | |
| "step": 72 | |
| }, | |
| { | |
| "epoch": 0.008670259204624138, | |
| "eval_loss": 1.423056721687317, | |
| "eval_runtime": 1986.6579, | |
| "eval_samples_per_second": 7.04, | |
| "eval_steps_per_second": 0.88, | |
| "step": 72 | |
| }, | |
| { | |
| "epoch": 0.009031520004816811, | |
| "grad_norm": 12.658801078796387, | |
| "learning_rate": 1.7860619515673033e-05, | |
| "loss": 5.9402, | |
| "step": 75 | |
| }, | |
| { | |
| "epoch": 0.009392780805009483, | |
| "grad_norm": 15.684614181518555, | |
| "learning_rate": 1.4033009983067452e-05, | |
| "loss": 5.6994, | |
| "step": 78 | |
| }, | |
| { | |
| "epoch": 0.009754041605202156, | |
| "grad_norm": 11.472796440124512, | |
| "learning_rate": 1.0599462319663905e-05, | |
| "loss": 5.8439, | |
| "step": 81 | |
| }, | |
| { | |
| "epoch": 0.009754041605202156, | |
| "eval_loss": 1.4167842864990234, | |
| "eval_runtime": 1987.061, | |
| "eval_samples_per_second": 7.039, | |
| "eval_steps_per_second": 0.88, | |
| "step": 81 | |
| }, | |
| { | |
| "epoch": 0.010115302405394828, | |
| "grad_norm": 11.89443588256836, | |
| "learning_rate": 7.597595192178702e-06, | |
| "loss": 5.465, | |
| "step": 84 | |
| }, | |
| { | |
| "epoch": 0.0104765632055875, | |
| "grad_norm": 13.721631050109863, | |
| "learning_rate": 5.060297685041659e-06, | |
| "loss": 5.8636, | |
| "step": 87 | |
| }, | |
| { | |
| "epoch": 0.010837824005780173, | |
| "grad_norm": 13.445150375366211, | |
| "learning_rate": 3.0153689607045845e-06, | |
| "loss": 5.7462, | |
| "step": 90 | |
| }, | |
| { | |
| "epoch": 0.010837824005780173, | |
| "eval_loss": 1.4135538339614868, | |
| "eval_runtime": 1984.5973, | |
| "eval_samples_per_second": 7.048, | |
| "eval_steps_per_second": 0.881, | |
| "step": 90 | |
| }, | |
| { | |
| "epoch": 0.011199084805972846, | |
| "grad_norm": 12.587129592895508, | |
| "learning_rate": 1.4852136862001764e-06, | |
| "loss": 5.5065, | |
| "step": 93 | |
| }, | |
| { | |
| "epoch": 0.011560345606165518, | |
| "grad_norm": 13.361394882202148, | |
| "learning_rate": 4.865965629214819e-07, | |
| "loss": 6.0172, | |
| "step": 96 | |
| }, | |
| { | |
| "epoch": 0.01192160640635819, | |
| "grad_norm": 13.288008689880371, | |
| "learning_rate": 3.04586490452119e-08, | |
| "loss": 5.6398, | |
| "step": 99 | |
| }, | |
| { | |
| "epoch": 0.01192160640635819, | |
| "eval_loss": 1.4129655361175537, | |
| "eval_runtime": 1979.296, | |
| "eval_samples_per_second": 7.067, | |
| "eval_steps_per_second": 0.884, | |
| "step": 99 | |
| } | |
| ], | |
| "logging_steps": 3, | |
| "max_steps": 100, | |
| "num_input_tokens_seen": 0, | |
| "num_train_epochs": 1, | |
| "save_steps": 9, | |
| "stateful_callbacks": { | |
| "TrainerControl": { | |
| "args": { | |
| "should_epoch_stop": false, | |
| "should_evaluate": false, | |
| "should_log": false, | |
| "should_save": true, | |
| "should_training_stop": true | |
| }, | |
| "attributes": {} | |
| } | |
| }, | |
| "total_flos": 2.812521215950848e+17, | |
| "train_batch_size": 8, | |
| "trial_name": null, | |
| "trial_params": null | |
| } | |