| { | |
| "best_metric": null, | |
| "best_model_checkpoint": null, | |
| "epoch": 0.05815644082582146, | |
| "eval_steps": 25, | |
| "global_step": 75, | |
| "is_hyper_param_search": false, | |
| "is_local_process_zero": true, | |
| "is_world_process_zero": true, | |
| "log_history": [ | |
| { | |
| "epoch": 0.0007754192110109528, | |
| "grad_norm": 13.743983268737793, | |
| "learning_rate": 3.3333333333333335e-05, | |
| "loss": 10.4552, | |
| "step": 1 | |
| }, | |
| { | |
| "epoch": 0.0007754192110109528, | |
| "eval_loss": 10.45703125, | |
| "eval_runtime": 17.3532, | |
| "eval_samples_per_second": 62.582, | |
| "eval_steps_per_second": 31.291, | |
| "step": 1 | |
| }, | |
| { | |
| "epoch": 0.0015508384220219056, | |
| "grad_norm": 15.994314193725586, | |
| "learning_rate": 6.666666666666667e-05, | |
| "loss": 10.4156, | |
| "step": 2 | |
| }, | |
| { | |
| "epoch": 0.0023262576330328583, | |
| "grad_norm": 17.251068115234375, | |
| "learning_rate": 0.0001, | |
| "loss": 10.3381, | |
| "step": 3 | |
| }, | |
| { | |
| "epoch": 0.0031016768440438112, | |
| "grad_norm": 13.243555068969727, | |
| "learning_rate": 9.99524110790929e-05, | |
| "loss": 10.5021, | |
| "step": 4 | |
| }, | |
| { | |
| "epoch": 0.003877096055054764, | |
| "grad_norm": 12.53420352935791, | |
| "learning_rate": 9.980973490458728e-05, | |
| "loss": 10.0685, | |
| "step": 5 | |
| }, | |
| { | |
| "epoch": 0.004652515266065717, | |
| "grad_norm": 13.710821151733398, | |
| "learning_rate": 9.957224306869053e-05, | |
| "loss": 9.9248, | |
| "step": 6 | |
| }, | |
| { | |
| "epoch": 0.0054279344770766695, | |
| "grad_norm": 14.05363941192627, | |
| "learning_rate": 9.924038765061042e-05, | |
| "loss": 9.9161, | |
| "step": 7 | |
| }, | |
| { | |
| "epoch": 0.0062033536880876225, | |
| "grad_norm": 13.404027938842773, | |
| "learning_rate": 9.881480035599667e-05, | |
| "loss": 9.5758, | |
| "step": 8 | |
| }, | |
| { | |
| "epoch": 0.006978772899098575, | |
| "grad_norm": 15.494867324829102, | |
| "learning_rate": 9.829629131445342e-05, | |
| "loss": 9.7933, | |
| "step": 9 | |
| }, | |
| { | |
| "epoch": 0.007754192110109528, | |
| "grad_norm": 21.381547927856445, | |
| "learning_rate": 9.768584753741134e-05, | |
| "loss": 9.1108, | |
| "step": 10 | |
| }, | |
| { | |
| "epoch": 0.00852961132112048, | |
| "grad_norm": 17.526710510253906, | |
| "learning_rate": 9.698463103929542e-05, | |
| "loss": 9.1686, | |
| "step": 11 | |
| }, | |
| { | |
| "epoch": 0.009305030532131433, | |
| "grad_norm": 16.56471824645996, | |
| "learning_rate": 9.619397662556435e-05, | |
| "loss": 9.0037, | |
| "step": 12 | |
| }, | |
| { | |
| "epoch": 0.010080449743142386, | |
| "grad_norm": 17.765729904174805, | |
| "learning_rate": 9.53153893518325e-05, | |
| "loss": 8.9391, | |
| "step": 13 | |
| }, | |
| { | |
| "epoch": 0.010855868954153339, | |
| "grad_norm": 18.965198516845703, | |
| "learning_rate": 9.435054165891109e-05, | |
| "loss": 8.794, | |
| "step": 14 | |
| }, | |
| { | |
| "epoch": 0.011631288165164292, | |
| "grad_norm": 20.40072250366211, | |
| "learning_rate": 9.330127018922194e-05, | |
| "loss": 8.0026, | |
| "step": 15 | |
| }, | |
| { | |
| "epoch": 0.012406707376175245, | |
| "grad_norm": 15.78911304473877, | |
| "learning_rate": 9.21695722906443e-05, | |
| "loss": 8.396, | |
| "step": 16 | |
| }, | |
| { | |
| "epoch": 0.013182126587186198, | |
| "grad_norm": 21.308853149414062, | |
| "learning_rate": 9.09576022144496e-05, | |
| "loss": 8.1367, | |
| "step": 17 | |
| }, | |
| { | |
| "epoch": 0.01395754579819715, | |
| "grad_norm": 17.67489242553711, | |
| "learning_rate": 8.966766701456177e-05, | |
| "loss": 7.9123, | |
| "step": 18 | |
| }, | |
| { | |
| "epoch": 0.014732965009208104, | |
| "grad_norm": 17.39400291442871, | |
| "learning_rate": 8.83022221559489e-05, | |
| "loss": 7.8025, | |
| "step": 19 | |
| }, | |
| { | |
| "epoch": 0.015508384220219057, | |
| "grad_norm": 18.812978744506836, | |
| "learning_rate": 8.68638668405062e-05, | |
| "loss": 7.7002, | |
| "step": 20 | |
| }, | |
| { | |
| "epoch": 0.01628380343123001, | |
| "grad_norm": 20.998043060302734, | |
| "learning_rate": 8.535533905932738e-05, | |
| "loss": 7.5169, | |
| "step": 21 | |
| }, | |
| { | |
| "epoch": 0.01705922264224096, | |
| "grad_norm": 17.887672424316406, | |
| "learning_rate": 8.377951038078302e-05, | |
| "loss": 7.2246, | |
| "step": 22 | |
| }, | |
| { | |
| "epoch": 0.017834641853251915, | |
| "grad_norm": 18.00612449645996, | |
| "learning_rate": 8.213938048432697e-05, | |
| "loss": 6.679, | |
| "step": 23 | |
| }, | |
| { | |
| "epoch": 0.018610061064262867, | |
| "grad_norm": 16.111370086669922, | |
| "learning_rate": 8.043807145043604e-05, | |
| "loss": 6.2643, | |
| "step": 24 | |
| }, | |
| { | |
| "epoch": 0.01938548027527382, | |
| "grad_norm": 22.258943557739258, | |
| "learning_rate": 7.86788218175523e-05, | |
| "loss": 6.6102, | |
| "step": 25 | |
| }, | |
| { | |
| "epoch": 0.01938548027527382, | |
| "eval_loss": 6.311874866485596, | |
| "eval_runtime": 17.3852, | |
| "eval_samples_per_second": 62.467, | |
| "eval_steps_per_second": 31.233, | |
| "step": 25 | |
| }, | |
| { | |
| "epoch": 0.020160899486284772, | |
| "grad_norm": 15.573179244995117, | |
| "learning_rate": 7.68649804173412e-05, | |
| "loss": 6.3441, | |
| "step": 26 | |
| }, | |
| { | |
| "epoch": 0.020936318697295727, | |
| "grad_norm": 15.074617385864258, | |
| "learning_rate": 7.500000000000001e-05, | |
| "loss": 6.2842, | |
| "step": 27 | |
| }, | |
| { | |
| "epoch": 0.021711737908306678, | |
| "grad_norm": 15.864977836608887, | |
| "learning_rate": 7.308743066175172e-05, | |
| "loss": 5.7357, | |
| "step": 28 | |
| }, | |
| { | |
| "epoch": 0.02248715711931763, | |
| "grad_norm": 19.382728576660156, | |
| "learning_rate": 7.113091308703498e-05, | |
| "loss": 6.0517, | |
| "step": 29 | |
| }, | |
| { | |
| "epoch": 0.023262576330328584, | |
| "grad_norm": 18.489917755126953, | |
| "learning_rate": 6.91341716182545e-05, | |
| "loss": 5.7561, | |
| "step": 30 | |
| }, | |
| { | |
| "epoch": 0.024037995541339535, | |
| "grad_norm": 11.977217674255371, | |
| "learning_rate": 6.710100716628344e-05, | |
| "loss": 5.3761, | |
| "step": 31 | |
| }, | |
| { | |
| "epoch": 0.02481341475235049, | |
| "grad_norm": 12.359403610229492, | |
| "learning_rate": 6.503528997521366e-05, | |
| "loss": 5.6858, | |
| "step": 32 | |
| }, | |
| { | |
| "epoch": 0.02558883396336144, | |
| "grad_norm": 17.48925018310547, | |
| "learning_rate": 6.294095225512603e-05, | |
| "loss": 5.4112, | |
| "step": 33 | |
| }, | |
| { | |
| "epoch": 0.026364253174372396, | |
| "grad_norm": 12.810553550720215, | |
| "learning_rate": 6.0821980696905146e-05, | |
| "loss": 5.1038, | |
| "step": 34 | |
| }, | |
| { | |
| "epoch": 0.027139672385383347, | |
| "grad_norm": 11.41131591796875, | |
| "learning_rate": 5.868240888334653e-05, | |
| "loss": 5.1019, | |
| "step": 35 | |
| }, | |
| { | |
| "epoch": 0.0279150915963943, | |
| "grad_norm": 11.404109001159668, | |
| "learning_rate": 5.6526309611002594e-05, | |
| "loss": 5.0096, | |
| "step": 36 | |
| }, | |
| { | |
| "epoch": 0.028690510807405253, | |
| "grad_norm": 10.934562683105469, | |
| "learning_rate": 5.435778713738292e-05, | |
| "loss": 4.8367, | |
| "step": 37 | |
| }, | |
| { | |
| "epoch": 0.029465930018416207, | |
| "grad_norm": 11.160900115966797, | |
| "learning_rate": 5.218096936826681e-05, | |
| "loss": 5.009, | |
| "step": 38 | |
| }, | |
| { | |
| "epoch": 0.03024134922942716, | |
| "grad_norm": 9.400223731994629, | |
| "learning_rate": 5e-05, | |
| "loss": 5.121, | |
| "step": 39 | |
| }, | |
| { | |
| "epoch": 0.031016768440438113, | |
| "grad_norm": 9.578547477722168, | |
| "learning_rate": 4.781903063173321e-05, | |
| "loss": 4.9154, | |
| "step": 40 | |
| }, | |
| { | |
| "epoch": 0.03179218765144907, | |
| "grad_norm": 8.08479118347168, | |
| "learning_rate": 4.564221286261709e-05, | |
| "loss": 4.4877, | |
| "step": 41 | |
| }, | |
| { | |
| "epoch": 0.03256760686246002, | |
| "grad_norm": 9.323592185974121, | |
| "learning_rate": 4.347369038899744e-05, | |
| "loss": 4.7919, | |
| "step": 42 | |
| }, | |
| { | |
| "epoch": 0.03334302607347097, | |
| "grad_norm": 12.592964172363281, | |
| "learning_rate": 4.131759111665349e-05, | |
| "loss": 4.9646, | |
| "step": 43 | |
| }, | |
| { | |
| "epoch": 0.03411844528448192, | |
| "grad_norm": 8.313202857971191, | |
| "learning_rate": 3.917801930309486e-05, | |
| "loss": 4.5304, | |
| "step": 44 | |
| }, | |
| { | |
| "epoch": 0.03489386449549287, | |
| "grad_norm": 9.623037338256836, | |
| "learning_rate": 3.705904774487396e-05, | |
| "loss": 4.5277, | |
| "step": 45 | |
| }, | |
| { | |
| "epoch": 0.03566928370650383, | |
| "grad_norm": 6.551573753356934, | |
| "learning_rate": 3.4964710024786354e-05, | |
| "loss": 4.7194, | |
| "step": 46 | |
| }, | |
| { | |
| "epoch": 0.03644470291751478, | |
| "grad_norm": 9.742829322814941, | |
| "learning_rate": 3.289899283371657e-05, | |
| "loss": 4.4445, | |
| "step": 47 | |
| }, | |
| { | |
| "epoch": 0.03722012212852573, | |
| "grad_norm": 6.696950435638428, | |
| "learning_rate": 3.086582838174551e-05, | |
| "loss": 4.3009, | |
| "step": 48 | |
| }, | |
| { | |
| "epoch": 0.037995541339536684, | |
| "grad_norm": 6.955257415771484, | |
| "learning_rate": 2.886908691296504e-05, | |
| "loss": 4.2683, | |
| "step": 49 | |
| }, | |
| { | |
| "epoch": 0.03877096055054764, | |
| "grad_norm": 11.268085479736328, | |
| "learning_rate": 2.6912569338248315e-05, | |
| "loss": 4.3482, | |
| "step": 50 | |
| }, | |
| { | |
| "epoch": 0.03877096055054764, | |
| "eval_loss": 4.515497207641602, | |
| "eval_runtime": 17.341, | |
| "eval_samples_per_second": 62.626, | |
| "eval_steps_per_second": 31.313, | |
| "step": 50 | |
| }, | |
| { | |
| "epoch": 0.039546379761558594, | |
| "grad_norm": 7.341830253601074, | |
| "learning_rate": 2.500000000000001e-05, | |
| "loss": 4.3773, | |
| "step": 51 | |
| }, | |
| { | |
| "epoch": 0.040321798972569545, | |
| "grad_norm": 8.497244834899902, | |
| "learning_rate": 2.3135019582658802e-05, | |
| "loss": 4.6826, | |
| "step": 52 | |
| }, | |
| { | |
| "epoch": 0.041097218183580496, | |
| "grad_norm": 7.568642616271973, | |
| "learning_rate": 2.132117818244771e-05, | |
| "loss": 4.4897, | |
| "step": 53 | |
| }, | |
| { | |
| "epoch": 0.041872637394591454, | |
| "grad_norm": 7.663918495178223, | |
| "learning_rate": 1.9561928549563968e-05, | |
| "loss": 5.1256, | |
| "step": 54 | |
| }, | |
| { | |
| "epoch": 0.042648056605602405, | |
| "grad_norm": 6.138592720031738, | |
| "learning_rate": 1.7860619515673033e-05, | |
| "loss": 4.3068, | |
| "step": 55 | |
| }, | |
| { | |
| "epoch": 0.043423475816613356, | |
| "grad_norm": 6.359229564666748, | |
| "learning_rate": 1.622048961921699e-05, | |
| "loss": 4.6431, | |
| "step": 56 | |
| }, | |
| { | |
| "epoch": 0.04419889502762431, | |
| "grad_norm": 6.404804706573486, | |
| "learning_rate": 1.4644660940672627e-05, | |
| "loss": 4.4604, | |
| "step": 57 | |
| }, | |
| { | |
| "epoch": 0.04497431423863526, | |
| "grad_norm": 5.194124221801758, | |
| "learning_rate": 1.3136133159493802e-05, | |
| "loss": 4.4148, | |
| "step": 58 | |
| }, | |
| { | |
| "epoch": 0.04574973344964622, | |
| "grad_norm": 6.9789557456970215, | |
| "learning_rate": 1.1697777844051105e-05, | |
| "loss": 4.5563, | |
| "step": 59 | |
| }, | |
| { | |
| "epoch": 0.04652515266065717, | |
| "grad_norm": 6.215468406677246, | |
| "learning_rate": 1.0332332985438248e-05, | |
| "loss": 4.6267, | |
| "step": 60 | |
| }, | |
| { | |
| "epoch": 0.04730057187166812, | |
| "grad_norm": 5.669643878936768, | |
| "learning_rate": 9.042397785550405e-06, | |
| "loss": 4.3564, | |
| "step": 61 | |
| }, | |
| { | |
| "epoch": 0.04807599108267907, | |
| "grad_norm": 6.695188045501709, | |
| "learning_rate": 7.830427709355725e-06, | |
| "loss": 4.6728, | |
| "step": 62 | |
| }, | |
| { | |
| "epoch": 0.04885141029369003, | |
| "grad_norm": 6.3594069480896, | |
| "learning_rate": 6.698729810778065e-06, | |
| "loss": 4.4725, | |
| "step": 63 | |
| }, | |
| { | |
| "epoch": 0.04962682950470098, | |
| "grad_norm": 6.124025344848633, | |
| "learning_rate": 5.649458341088915e-06, | |
| "loss": 4.7726, | |
| "step": 64 | |
| }, | |
| { | |
| "epoch": 0.05040224871571193, | |
| "grad_norm": 6.306771755218506, | |
| "learning_rate": 4.684610648167503e-06, | |
| "loss": 4.7565, | |
| "step": 65 | |
| }, | |
| { | |
| "epoch": 0.05117766792672288, | |
| "grad_norm": 6.299499034881592, | |
| "learning_rate": 3.8060233744356633e-06, | |
| "loss": 4.1523, | |
| "step": 66 | |
| }, | |
| { | |
| "epoch": 0.05195308713773384, | |
| "grad_norm": 5.291078567504883, | |
| "learning_rate": 3.0153689607045845e-06, | |
| "loss": 4.4084, | |
| "step": 67 | |
| }, | |
| { | |
| "epoch": 0.05272850634874479, | |
| "grad_norm": 7.21616268157959, | |
| "learning_rate": 2.314152462588659e-06, | |
| "loss": 4.2842, | |
| "step": 68 | |
| }, | |
| { | |
| "epoch": 0.05350392555975574, | |
| "grad_norm": 5.060047626495361, | |
| "learning_rate": 1.70370868554659e-06, | |
| "loss": 4.4137, | |
| "step": 69 | |
| }, | |
| { | |
| "epoch": 0.054279344770766694, | |
| "grad_norm": 5.063629627227783, | |
| "learning_rate": 1.1851996440033319e-06, | |
| "loss": 4.4392, | |
| "step": 70 | |
| }, | |
| { | |
| "epoch": 0.05505476398177765, | |
| "grad_norm": 4.903884410858154, | |
| "learning_rate": 7.596123493895991e-07, | |
| "loss": 4.3334, | |
| "step": 71 | |
| }, | |
| { | |
| "epoch": 0.0558301831927886, | |
| "grad_norm": 6.626821517944336, | |
| "learning_rate": 4.277569313094809e-07, | |
| "loss": 4.8866, | |
| "step": 72 | |
| }, | |
| { | |
| "epoch": 0.056605602403799554, | |
| "grad_norm": 5.395899772644043, | |
| "learning_rate": 1.9026509541272275e-07, | |
| "loss": 4.1787, | |
| "step": 73 | |
| }, | |
| { | |
| "epoch": 0.057381021614810505, | |
| "grad_norm": 5.96885347366333, | |
| "learning_rate": 4.7588920907110094e-08, | |
| "loss": 4.9024, | |
| "step": 74 | |
| }, | |
| { | |
| "epoch": 0.05815644082582146, | |
| "grad_norm": 6.069552898406982, | |
| "learning_rate": 0.0, | |
| "loss": 4.4114, | |
| "step": 75 | |
| }, | |
| { | |
| "epoch": 0.05815644082582146, | |
| "eval_loss": 4.3782525062561035, | |
| "eval_runtime": 17.3208, | |
| "eval_samples_per_second": 62.699, | |
| "eval_steps_per_second": 31.35, | |
| "step": 75 | |
| } | |
| ], | |
| "logging_steps": 1, | |
| "max_steps": 75, | |
| "num_input_tokens_seen": 0, | |
| "num_train_epochs": 1, | |
| "save_steps": 25, | |
| "stateful_callbacks": { | |
| "TrainerControl": { | |
| "args": { | |
| "should_epoch_stop": false, | |
| "should_evaluate": false, | |
| "should_log": false, | |
| "should_save": true, | |
| "should_training_stop": true | |
| }, | |
| "attributes": {} | |
| } | |
| }, | |
| "total_flos": 43502587084800.0, | |
| "train_batch_size": 2, | |
| "trial_name": null, | |
| "trial_params": null | |
| } | |