| { |
| "best_metric": null, |
| "best_model_checkpoint": null, |
| "epoch": 1.0, |
| "eval_steps": 1000, |
| "global_step": 19760, |
| "is_hyper_param_search": false, |
| "is_local_process_zero": true, |
| "is_world_process_zero": true, |
| "log_history": [ |
| { |
| "epoch": 0.025303643724696356, |
| "grad_norm": 1.9980295896530151, |
| "learning_rate": 4.8734817813765186e-05, |
| "loss": 4.276, |
| "step": 500 |
| }, |
| { |
| "epoch": 0.05060728744939271, |
| "grad_norm": 1.815592885017395, |
| "learning_rate": 4.746963562753037e-05, |
| "loss": 2.801, |
| "step": 1000 |
| }, |
| { |
| "epoch": 0.05060728744939271, |
| "eval_accuracy": 0.46582789392362756, |
| "eval_loss": 2.355369806289673, |
| "eval_runtime": 55.5712, |
| "eval_samples_per_second": 114.358, |
| "eval_steps_per_second": 3.581, |
| "step": 1000 |
| }, |
| { |
| "epoch": 0.07591093117408906, |
| "grad_norm": 4.090913772583008, |
| "learning_rate": 4.6204453441295545e-05, |
| "loss": 2.2031, |
| "step": 1500 |
| }, |
| { |
| "epoch": 0.10121457489878542, |
| "grad_norm": 1.4052000045776367, |
| "learning_rate": 4.4939271255060735e-05, |
| "loss": 1.8977, |
| "step": 2000 |
| }, |
| { |
| "epoch": 0.10121457489878542, |
| "eval_accuracy": 0.583393284126768, |
| "eval_loss": 1.7247928380966187, |
| "eval_runtime": 54.7342, |
| "eval_samples_per_second": 116.107, |
| "eval_steps_per_second": 3.636, |
| "step": 2000 |
| }, |
| { |
| "epoch": 0.12651821862348178, |
| "grad_norm": 1.7114479541778564, |
| "learning_rate": 4.367408906882591e-05, |
| "loss": 1.7504, |
| "step": 2500 |
| }, |
| { |
| "epoch": 0.15182186234817813, |
| "grad_norm": 1.3184257745742798, |
| "learning_rate": 4.2408906882591095e-05, |
| "loss": 1.658, |
| "step": 3000 |
| }, |
| { |
| "epoch": 0.15182186234817813, |
| "eval_accuracy": 0.6145087841948328, |
| "eval_loss": 1.5497987270355225, |
| "eval_runtime": 54.8048, |
| "eval_samples_per_second": 115.957, |
| "eval_steps_per_second": 3.631, |
| "step": 3000 |
| }, |
| { |
| "epoch": 0.1771255060728745, |
| "grad_norm": 1.22968327999115, |
| "learning_rate": 4.114372469635628e-05, |
| "loss": 1.5931, |
| "step": 3500 |
| }, |
| { |
| "epoch": 0.20242914979757085, |
| "grad_norm": 1.487292766571045, |
| "learning_rate": 3.9878542510121455e-05, |
| "loss": 1.5426, |
| "step": 4000 |
| }, |
| { |
| "epoch": 0.20242914979757085, |
| "eval_accuracy": 0.6321363940155341, |
| "eval_loss": 1.4541865587234497, |
| "eval_runtime": 55.1613, |
| "eval_samples_per_second": 115.208, |
| "eval_steps_per_second": 3.608, |
| "step": 4000 |
| }, |
| { |
| "epoch": 0.22773279352226722, |
| "grad_norm": 1.28114914894104, |
| "learning_rate": 3.8613360323886645e-05, |
| "loss": 1.5061, |
| "step": 4500 |
| }, |
| { |
| "epoch": 0.25303643724696356, |
| "grad_norm": 1.1875263452529907, |
| "learning_rate": 3.734817813765182e-05, |
| "loss": 1.4721, |
| "step": 5000 |
| }, |
| { |
| "epoch": 0.25303643724696356, |
| "eval_accuracy": 0.643541426805811, |
| "eval_loss": 1.393043875694275, |
| "eval_runtime": 55.2636, |
| "eval_samples_per_second": 114.994, |
| "eval_steps_per_second": 3.601, |
| "step": 5000 |
| }, |
| { |
| "epoch": 0.2783400809716599, |
| "grad_norm": 1.2500560283660889, |
| "learning_rate": 3.6082995951417005e-05, |
| "loss": 1.4487, |
| "step": 5500 |
| }, |
| { |
| "epoch": 0.30364372469635625, |
| "grad_norm": 1.1801079511642456, |
| "learning_rate": 3.481781376518219e-05, |
| "loss": 1.4237, |
| "step": 6000 |
| }, |
| { |
| "epoch": 0.30364372469635625, |
| "eval_accuracy": 0.651714884947544, |
| "eval_loss": 1.3496800661087036, |
| "eval_runtime": 55.4085, |
| "eval_samples_per_second": 114.694, |
| "eval_steps_per_second": 3.592, |
| "step": 6000 |
| }, |
| { |
| "epoch": 0.32894736842105265, |
| "grad_norm": 1.1773282289505005, |
| "learning_rate": 3.355263157894737e-05, |
| "loss": 1.4045, |
| "step": 6500 |
| }, |
| { |
| "epoch": 0.354251012145749, |
| "grad_norm": 1.2128751277923584, |
| "learning_rate": 3.2287449392712554e-05, |
| "loss": 1.387, |
| "step": 7000 |
| }, |
| { |
| "epoch": 0.354251012145749, |
| "eval_accuracy": 0.6580368287837641, |
| "eval_loss": 1.316208839416504, |
| "eval_runtime": 54.4605, |
| "eval_samples_per_second": 116.69, |
| "eval_steps_per_second": 3.654, |
| "step": 7000 |
| }, |
| { |
| "epoch": 0.37955465587044535, |
| "grad_norm": 1.1830706596374512, |
| "learning_rate": 3.102226720647773e-05, |
| "loss": 1.3691, |
| "step": 7500 |
| }, |
| { |
| "epoch": 0.4048582995951417, |
| "grad_norm": 1.2444064617156982, |
| "learning_rate": 2.9757085020242914e-05, |
| "loss": 1.3537, |
| "step": 8000 |
| }, |
| { |
| "epoch": 0.4048582995951417, |
| "eval_accuracy": 0.6632618922916124, |
| "eval_loss": 1.2898718118667603, |
| "eval_runtime": 54.6011, |
| "eval_samples_per_second": 116.39, |
| "eval_steps_per_second": 3.645, |
| "step": 8000 |
| }, |
| { |
| "epoch": 0.43016194331983804, |
| "grad_norm": 1.07850182056427, |
| "learning_rate": 2.84919028340081e-05, |
| "loss": 1.3451, |
| "step": 8500 |
| }, |
| { |
| "epoch": 0.45546558704453444, |
| "grad_norm": 1.1335291862487793, |
| "learning_rate": 2.722672064777328e-05, |
| "loss": 1.3306, |
| "step": 9000 |
| }, |
| { |
| "epoch": 0.45546558704453444, |
| "eval_accuracy": 0.6675778879631574, |
| "eval_loss": 1.2683097124099731, |
| "eval_runtime": 54.9704, |
| "eval_samples_per_second": 115.608, |
| "eval_steps_per_second": 3.62, |
| "step": 9000 |
| }, |
| { |
| "epoch": 0.4807692307692308, |
| "grad_norm": 1.1087806224822998, |
| "learning_rate": 2.5961538461538464e-05, |
| "loss": 1.3194, |
| "step": 9500 |
| }, |
| { |
| "epoch": 0.5060728744939271, |
| "grad_norm": 1.095122218132019, |
| "learning_rate": 2.4696356275303644e-05, |
| "loss": 1.3127, |
| "step": 10000 |
| }, |
| { |
| "epoch": 0.5060728744939271, |
| "eval_accuracy": 0.6715556365666769, |
| "eval_loss": 1.247435450553894, |
| "eval_runtime": 54.4433, |
| "eval_samples_per_second": 116.727, |
| "eval_steps_per_second": 3.655, |
| "step": 10000 |
| }, |
| { |
| "epoch": 0.5313765182186235, |
| "grad_norm": 1.0680406093597412, |
| "learning_rate": 2.3431174089068827e-05, |
| "loss": 1.3044, |
| "step": 10500 |
| }, |
| { |
| "epoch": 0.5566801619433198, |
| "grad_norm": 1.1252034902572632, |
| "learning_rate": 2.216599190283401e-05, |
| "loss": 1.2925, |
| "step": 11000 |
| }, |
| { |
| "epoch": 0.5566801619433198, |
| "eval_accuracy": 0.674490802802267, |
| "eval_loss": 1.2325994968414307, |
| "eval_runtime": 54.7111, |
| "eval_samples_per_second": 116.156, |
| "eval_steps_per_second": 3.637, |
| "step": 11000 |
| }, |
| { |
| "epoch": 0.5819838056680162, |
| "grad_norm": 1.0976389646530151, |
| "learning_rate": 2.090080971659919e-05, |
| "loss": 1.2861, |
| "step": 11500 |
| }, |
| { |
| "epoch": 0.6072874493927125, |
| "grad_norm": 1.0920414924621582, |
| "learning_rate": 1.9635627530364373e-05, |
| "loss": 1.2779, |
| "step": 12000 |
| }, |
| { |
| "epoch": 0.6072874493927125, |
| "eval_accuracy": 0.6777880579865301, |
| "eval_loss": 1.2170982360839844, |
| "eval_runtime": 54.8191, |
| "eval_samples_per_second": 115.927, |
| "eval_steps_per_second": 3.63, |
| "step": 12000 |
| }, |
| { |
| "epoch": 0.6325910931174089, |
| "grad_norm": 1.1419272422790527, |
| "learning_rate": 1.8370445344129557e-05, |
| "loss": 1.2751, |
| "step": 12500 |
| }, |
| { |
| "epoch": 0.6578947368421053, |
| "grad_norm": 1.1571863889694214, |
| "learning_rate": 1.7105263157894737e-05, |
| "loss": 1.262, |
| "step": 13000 |
| }, |
| { |
| "epoch": 0.6578947368421053, |
| "eval_accuracy": 0.6801773220645838, |
| "eval_loss": 1.2050944566726685, |
| "eval_runtime": 54.1554, |
| "eval_samples_per_second": 117.347, |
| "eval_steps_per_second": 3.675, |
| "step": 13000 |
| }, |
| { |
| "epoch": 0.6831983805668016, |
| "grad_norm": 1.0533004999160767, |
| "learning_rate": 1.584008097165992e-05, |
| "loss": 1.2596, |
| "step": 13500 |
| }, |
| { |
| "epoch": 0.708502024291498, |
| "grad_norm": 1.1271319389343262, |
| "learning_rate": 1.4574898785425101e-05, |
| "loss": 1.2502, |
| "step": 14000 |
| }, |
| { |
| "epoch": 0.708502024291498, |
| "eval_accuracy": 0.6823253986016352, |
| "eval_loss": 1.194899082183838, |
| "eval_runtime": 54.6327, |
| "eval_samples_per_second": 116.322, |
| "eval_steps_per_second": 3.643, |
| "step": 14000 |
| }, |
| { |
| "epoch": 0.7338056680161943, |
| "grad_norm": 1.1062027215957642, |
| "learning_rate": 1.3309716599190283e-05, |
| "loss": 1.2472, |
| "step": 14500 |
| }, |
| { |
| "epoch": 0.7591093117408907, |
| "grad_norm": 1.0716649293899536, |
| "learning_rate": 1.2044534412955466e-05, |
| "loss": 1.2413, |
| "step": 15000 |
| }, |
| { |
| "epoch": 0.7591093117408907, |
| "eval_accuracy": 0.6842521301951265, |
| "eval_loss": 1.18516206741333, |
| "eval_runtime": 54.3013, |
| "eval_samples_per_second": 117.032, |
| "eval_steps_per_second": 3.665, |
| "step": 15000 |
| }, |
| { |
| "epoch": 0.7844129554655871, |
| "grad_norm": 1.0834686756134033, |
| "learning_rate": 1.077935222672065e-05, |
| "loss": 1.2397, |
| "step": 15500 |
| }, |
| { |
| "epoch": 0.8097165991902834, |
| "grad_norm": 1.1064422130584717, |
| "learning_rate": 9.51417004048583e-06, |
| "loss": 1.2354, |
| "step": 16000 |
| }, |
| { |
| "epoch": 0.8097165991902834, |
| "eval_accuracy": 0.6857009474455732, |
| "eval_loss": 1.1772840023040771, |
| "eval_runtime": 54.1739, |
| "eval_samples_per_second": 117.307, |
| "eval_steps_per_second": 3.673, |
| "step": 16000 |
| }, |
| { |
| "epoch": 0.8350202429149798, |
| "grad_norm": 1.0644614696502686, |
| "learning_rate": 8.248987854251013e-06, |
| "loss": 1.2285, |
| "step": 16500 |
| }, |
| { |
| "epoch": 0.8603238866396761, |
| "grad_norm": 1.1000635623931885, |
| "learning_rate": 6.983805668016195e-06, |
| "loss": 1.2254, |
| "step": 17000 |
| }, |
| { |
| "epoch": 0.8603238866396761, |
| "eval_accuracy": 0.6874041806353169, |
| "eval_loss": 1.1698853969573975, |
| "eval_runtime": 53.9492, |
| "eval_samples_per_second": 117.796, |
| "eval_steps_per_second": 3.689, |
| "step": 17000 |
| }, |
| { |
| "epoch": 0.8856275303643725, |
| "grad_norm": 1.0928003787994385, |
| "learning_rate": 5.718623481781377e-06, |
| "loss": 1.2212, |
| "step": 17500 |
| }, |
| { |
| "epoch": 0.9109311740890689, |
| "grad_norm": 1.0563207864761353, |
| "learning_rate": 4.453441295546559e-06, |
| "loss": 1.2186, |
| "step": 18000 |
| }, |
| { |
| "epoch": 0.9109311740890689, |
| "eval_accuracy": 0.6887204062656462, |
| "eval_loss": 1.1639304161071777, |
| "eval_runtime": 53.8445, |
| "eval_samples_per_second": 118.025, |
| "eval_steps_per_second": 3.696, |
| "step": 18000 |
| }, |
| { |
| "epoch": 0.9362348178137652, |
| "grad_norm": 1.1007548570632935, |
| "learning_rate": 3.1882591093117414e-06, |
| "loss": 1.22, |
| "step": 18500 |
| }, |
| { |
| "epoch": 0.9615384615384616, |
| "grad_norm": 1.0710889101028442, |
| "learning_rate": 1.9230769230769234e-06, |
| "loss": 1.2155, |
| "step": 19000 |
| }, |
| { |
| "epoch": 0.9615384615384616, |
| "eval_accuracy": 0.6896582381773113, |
| "eval_loss": 1.1596901416778564, |
| "eval_runtime": 54.1254, |
| "eval_samples_per_second": 117.413, |
| "eval_steps_per_second": 3.677, |
| "step": 19000 |
| }, |
| { |
| "epoch": 0.9868421052631579, |
| "grad_norm": 1.0729280710220337, |
| "learning_rate": 6.578947368421053e-07, |
| "loss": 1.2135, |
| "step": 19500 |
| }, |
| { |
| "epoch": 1.0, |
| "step": 19760, |
| "total_flos": 3.30439310180352e+17, |
| "train_loss": 1.4863598595746617, |
| "train_runtime": 7692.9101, |
| "train_samples_per_second": 82.195, |
| "train_steps_per_second": 2.569 |
| } |
| ], |
| "logging_steps": 500, |
| "max_steps": 19760, |
| "num_input_tokens_seen": 0, |
| "num_train_epochs": 1, |
| "save_steps": 1000, |
| "stateful_callbacks": { |
| "TrainerControl": { |
| "args": { |
| "should_epoch_stop": false, |
| "should_evaluate": false, |
| "should_log": false, |
| "should_save": true, |
| "should_training_stop": true |
| }, |
| "attributes": {} |
| } |
| }, |
| "total_flos": 3.30439310180352e+17, |
| "train_batch_size": 32, |
| "trial_name": null, |
| "trial_params": null |
| } |
|
|
