| { | |
| "best_metric": null, | |
| "best_model_checkpoint": null, | |
| "epoch": 0.010311670232785956, | |
| "eval_steps": 9, | |
| "global_step": 100, | |
| "is_hyper_param_search": false, | |
| "is_local_process_zero": true, | |
| "is_world_process_zero": true, | |
| "log_history": [ | |
| { | |
| "epoch": 0.00010311670232785956, | |
| "eval_loss": 2.0299148559570312, | |
| "eval_runtime": 1161.575, | |
| "eval_samples_per_second": 14.062, | |
| "eval_steps_per_second": 1.758, | |
| "step": 1 | |
| }, | |
| { | |
| "epoch": 0.0003093501069835787, | |
| "grad_norm": 22.63811492919922, | |
| "learning_rate": 3e-05, | |
| "loss": 8.112, | |
| "step": 3 | |
| }, | |
| { | |
| "epoch": 0.0006187002139671574, | |
| "grad_norm": 13.994317054748535, | |
| "learning_rate": 6e-05, | |
| "loss": 8.7124, | |
| "step": 6 | |
| }, | |
| { | |
| "epoch": 0.000928050320950736, | |
| "grad_norm": 7.255701065063477, | |
| "learning_rate": 9e-05, | |
| "loss": 6.6208, | |
| "step": 9 | |
| }, | |
| { | |
| "epoch": 0.000928050320950736, | |
| "eval_loss": 1.5573339462280273, | |
| "eval_runtime": 1167.6686, | |
| "eval_samples_per_second": 13.989, | |
| "eval_steps_per_second": 1.749, | |
| "step": 9 | |
| }, | |
| { | |
| "epoch": 0.0012374004279343148, | |
| "grad_norm": 6.881002902984619, | |
| "learning_rate": 9.987820251299122e-05, | |
| "loss": 6.2024, | |
| "step": 12 | |
| }, | |
| { | |
| "epoch": 0.0015467505349178933, | |
| "grad_norm": 7.591043949127197, | |
| "learning_rate": 9.924038765061042e-05, | |
| "loss": 6.0058, | |
| "step": 15 | |
| }, | |
| { | |
| "epoch": 0.001856100641901472, | |
| "grad_norm": 5.465282440185547, | |
| "learning_rate": 9.806308479691595e-05, | |
| "loss": 5.2033, | |
| "step": 18 | |
| }, | |
| { | |
| "epoch": 0.001856100641901472, | |
| "eval_loss": 1.4896771907806396, | |
| "eval_runtime": 1167.5877, | |
| "eval_samples_per_second": 13.99, | |
| "eval_steps_per_second": 1.749, | |
| "step": 18 | |
| }, | |
| { | |
| "epoch": 0.0021654507488850505, | |
| "grad_norm": 8.633875846862793, | |
| "learning_rate": 9.635919272833938e-05, | |
| "loss": 5.7537, | |
| "step": 21 | |
| }, | |
| { | |
| "epoch": 0.0024748008558686295, | |
| "grad_norm": 10.359308242797852, | |
| "learning_rate": 9.414737964294636e-05, | |
| "loss": 6.1929, | |
| "step": 24 | |
| }, | |
| { | |
| "epoch": 0.002784150962852208, | |
| "grad_norm": 5.20798397064209, | |
| "learning_rate": 9.145187862775209e-05, | |
| "loss": 5.966, | |
| "step": 27 | |
| }, | |
| { | |
| "epoch": 0.002784150962852208, | |
| "eval_loss": 1.4642568826675415, | |
| "eval_runtime": 1167.6125, | |
| "eval_samples_per_second": 13.989, | |
| "eval_steps_per_second": 1.749, | |
| "step": 27 | |
| }, | |
| { | |
| "epoch": 0.0030935010698357867, | |
| "grad_norm": 7.362185478210449, | |
| "learning_rate": 8.83022221559489e-05, | |
| "loss": 6.0468, | |
| "step": 30 | |
| }, | |
| { | |
| "epoch": 0.0034028511768193653, | |
| "grad_norm": 7.112700939178467, | |
| "learning_rate": 8.473291852294987e-05, | |
| "loss": 6.4047, | |
| "step": 33 | |
| }, | |
| { | |
| "epoch": 0.003712201283802944, | |
| "grad_norm": 6.42846155166626, | |
| "learning_rate": 8.07830737662829e-05, | |
| "loss": 5.5313, | |
| "step": 36 | |
| }, | |
| { | |
| "epoch": 0.003712201283802944, | |
| "eval_loss": 1.4507426023483276, | |
| "eval_runtime": 1167.4801, | |
| "eval_samples_per_second": 13.991, | |
| "eval_steps_per_second": 1.749, | |
| "step": 36 | |
| }, | |
| { | |
| "epoch": 0.004021551390786523, | |
| "grad_norm": 4.613251686096191, | |
| "learning_rate": 7.649596321166024e-05, | |
| "loss": 5.437, | |
| "step": 39 | |
| }, | |
| { | |
| "epoch": 0.004330901497770101, | |
| "grad_norm": 7.317685604095459, | |
| "learning_rate": 7.191855733945387e-05, | |
| "loss": 5.8941, | |
| "step": 42 | |
| }, | |
| { | |
| "epoch": 0.00464025160475368, | |
| "grad_norm": 6.0703606605529785, | |
| "learning_rate": 6.710100716628344e-05, | |
| "loss": 5.9139, | |
| "step": 45 | |
| }, | |
| { | |
| "epoch": 0.00464025160475368, | |
| "eval_loss": 1.442267656326294, | |
| "eval_runtime": 1167.7611, | |
| "eval_samples_per_second": 13.987, | |
| "eval_steps_per_second": 1.749, | |
| "step": 45 | |
| }, | |
| { | |
| "epoch": 0.004949601711737259, | |
| "grad_norm": 4.372334957122803, | |
| "learning_rate": 6.209609477998338e-05, | |
| "loss": 5.2514, | |
| "step": 48 | |
| }, | |
| { | |
| "epoch": 0.005258951818720837, | |
| "grad_norm": 6.997920513153076, | |
| "learning_rate": 5.695865504800327e-05, | |
| "loss": 6.0713, | |
| "step": 51 | |
| }, | |
| { | |
| "epoch": 0.005568301925704416, | |
| "grad_norm": 6.882256031036377, | |
| "learning_rate": 5.174497483512506e-05, | |
| "loss": 6.3112, | |
| "step": 54 | |
| }, | |
| { | |
| "epoch": 0.005568301925704416, | |
| "eval_loss": 1.4337407350540161, | |
| "eval_runtime": 1167.2604, | |
| "eval_samples_per_second": 13.993, | |
| "eval_steps_per_second": 1.749, | |
| "step": 54 | |
| }, | |
| { | |
| "epoch": 0.005877652032687994, | |
| "grad_norm": 4.815347194671631, | |
| "learning_rate": 4.6512176312793736e-05, | |
| "loss": 5.721, | |
| "step": 57 | |
| }, | |
| { | |
| "epoch": 0.006187002139671573, | |
| "grad_norm": 8.502628326416016, | |
| "learning_rate": 4.131759111665349e-05, | |
| "loss": 5.9287, | |
| "step": 60 | |
| }, | |
| { | |
| "epoch": 0.0064963522466551515, | |
| "grad_norm": 8.623035430908203, | |
| "learning_rate": 3.6218132209150045e-05, | |
| "loss": 6.0742, | |
| "step": 63 | |
| }, | |
| { | |
| "epoch": 0.0064963522466551515, | |
| "eval_loss": 1.42971670627594, | |
| "eval_runtime": 1166.8218, | |
| "eval_samples_per_second": 13.999, | |
| "eval_steps_per_second": 1.75, | |
| "step": 63 | |
| }, | |
| { | |
| "epoch": 0.0068057023536387306, | |
| "grad_norm": 6.28512716293335, | |
| "learning_rate": 3.12696703292044e-05, | |
| "loss": 6.1469, | |
| "step": 66 | |
| }, | |
| { | |
| "epoch": 0.00711505246062231, | |
| "grad_norm": 6.072748184204102, | |
| "learning_rate": 2.6526421860705473e-05, | |
| "loss": 6.4233, | |
| "step": 69 | |
| }, | |
| { | |
| "epoch": 0.007424402567605888, | |
| "grad_norm": 5.338780403137207, | |
| "learning_rate": 2.2040354826462668e-05, | |
| "loss": 5.9596, | |
| "step": 72 | |
| }, | |
| { | |
| "epoch": 0.007424402567605888, | |
| "eval_loss": 1.4262299537658691, | |
| "eval_runtime": 1166.5213, | |
| "eval_samples_per_second": 14.002, | |
| "eval_steps_per_second": 1.751, | |
| "step": 72 | |
| }, | |
| { | |
| "epoch": 0.007733752674589467, | |
| "grad_norm": 5.755375385284424, | |
| "learning_rate": 1.7860619515673033e-05, | |
| "loss": 4.9325, | |
| "step": 75 | |
| }, | |
| { | |
| "epoch": 0.008043102781573046, | |
| "grad_norm": 7.211399078369141, | |
| "learning_rate": 1.4033009983067452e-05, | |
| "loss": 5.6665, | |
| "step": 78 | |
| }, | |
| { | |
| "epoch": 0.008352452888556624, | |
| "grad_norm": 9.436430931091309, | |
| "learning_rate": 1.0599462319663905e-05, | |
| "loss": 5.676, | |
| "step": 81 | |
| }, | |
| { | |
| "epoch": 0.008352452888556624, | |
| "eval_loss": 1.4237958192825317, | |
| "eval_runtime": 1166.3522, | |
| "eval_samples_per_second": 14.004, | |
| "eval_steps_per_second": 1.751, | |
| "step": 81 | |
| }, | |
| { | |
| "epoch": 0.008661802995540202, | |
| "grad_norm": 9.019287109375, | |
| "learning_rate": 7.597595192178702e-06, | |
| "loss": 6.0424, | |
| "step": 84 | |
| }, | |
| { | |
| "epoch": 0.008971153102523782, | |
| "grad_norm": 5.952609062194824, | |
| "learning_rate": 5.060297685041659e-06, | |
| "loss": 5.8633, | |
| "step": 87 | |
| }, | |
| { | |
| "epoch": 0.00928050320950736, | |
| "grad_norm": 5.024654388427734, | |
| "learning_rate": 3.0153689607045845e-06, | |
| "loss": 5.4096, | |
| "step": 90 | |
| }, | |
| { | |
| "epoch": 0.00928050320950736, | |
| "eval_loss": 1.422675371170044, | |
| "eval_runtime": 1166.5377, | |
| "eval_samples_per_second": 14.002, | |
| "eval_steps_per_second": 1.75, | |
| "step": 90 | |
| }, | |
| { | |
| "epoch": 0.009589853316490938, | |
| "grad_norm": 3.980428457260132, | |
| "learning_rate": 1.4852136862001764e-06, | |
| "loss": 4.9397, | |
| "step": 93 | |
| }, | |
| { | |
| "epoch": 0.009899203423474518, | |
| "grad_norm": 4.345184803009033, | |
| "learning_rate": 4.865965629214819e-07, | |
| "loss": 5.0577, | |
| "step": 96 | |
| }, | |
| { | |
| "epoch": 0.010208553530458096, | |
| "grad_norm": 8.817083358764648, | |
| "learning_rate": 3.04586490452119e-08, | |
| "loss": 6.205, | |
| "step": 99 | |
| }, | |
| { | |
| "epoch": 0.010208553530458096, | |
| "eval_loss": 1.4224265813827515, | |
| "eval_runtime": 1166.663, | |
| "eval_samples_per_second": 14.001, | |
| "eval_steps_per_second": 1.75, | |
| "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": 1.4625093572571955e+17, | |
| "train_batch_size": 8, | |
| "trial_name": null, | |
| "trial_params": null | |
| } | |