| { |
| "best_global_step": null, |
| "best_metric": null, |
| "best_model_checkpoint": null, |
| "epoch": 1.0, |
| "eval_steps": 600, |
| "global_step": 2850, |
| "is_hyper_param_search": false, |
| "is_local_process_zero": true, |
| "is_world_process_zero": true, |
| "log_history": [ |
| { |
| "epoch": 0.03508771929824561, |
| "grad_norm": 2.7452964782714844, |
| "learning_rate": 9.999454188754037e-06, |
| "loss": 3.2364, |
| "step": 100 |
| }, |
| { |
| "epoch": 0.07017543859649122, |
| "grad_norm": 1.8586595058441162, |
| "learning_rate": 9.958816479499802e-06, |
| "loss": 2.8017, |
| "step": 200 |
| }, |
| { |
| "epoch": 0.10526315789473684, |
| "grad_norm": 1.7550557851791382, |
| "learning_rate": 9.85418552940924e-06, |
| "loss": 2.5137, |
| "step": 300 |
| }, |
| { |
| "epoch": 0.14035087719298245, |
| "grad_norm": 1.5816640853881836, |
| "learning_rate": 9.68691159486222e-06, |
| "loss": 2.3289, |
| "step": 400 |
| }, |
| { |
| "epoch": 0.17543859649122806, |
| "grad_norm": 2.3989932537078857, |
| "learning_rate": 9.459153336342227e-06, |
| "loss": 2.1808, |
| "step": 500 |
| }, |
| { |
| "epoch": 0.21052631578947367, |
| "grad_norm": 1.958235263824463, |
| "learning_rate": 9.173849961047222e-06, |
| "loss": 2.0186, |
| "step": 600 |
| }, |
| { |
| "epoch": 0.21052631578947367, |
| "eval_loss": 1.981339454650879, |
| "eval_runtime": 229.6674, |
| "eval_samples_per_second": 38.599, |
| "eval_steps_per_second": 1.38, |
| "step": 600 |
| }, |
| { |
| "epoch": 0.24561403508771928, |
| "grad_norm": 2.1183085441589355, |
| "learning_rate": 8.834683292590185e-06, |
| "loss": 1.9565, |
| "step": 700 |
| }, |
| { |
| "epoch": 0.2807017543859649, |
| "grad_norm": 1.6860677003860474, |
| "learning_rate": 8.44603025727762e-06, |
| "loss": 1.9274, |
| "step": 800 |
| }, |
| { |
| "epoch": 0.3157894736842105, |
| "grad_norm": 1.8299016952514648, |
| "learning_rate": 8.012906400127846e-06, |
| "loss": 1.9046, |
| "step": 900 |
| }, |
| { |
| "epoch": 0.3508771929824561, |
| "grad_norm": 2.3082456588745117, |
| "learning_rate": 7.54090115955165e-06, |
| "loss": 1.8956, |
| "step": 1000 |
| }, |
| { |
| "epoch": 0.38596491228070173, |
| "grad_norm": 1.582141637802124, |
| "learning_rate": 7.036105735971767e-06, |
| "loss": 1.8842, |
| "step": 1100 |
| }, |
| { |
| "epoch": 0.42105263157894735, |
| "grad_norm": 2.031499147415161, |
| "learning_rate": 6.5050344852325334e-06, |
| "loss": 1.8814, |
| "step": 1200 |
| }, |
| { |
| "epoch": 0.42105263157894735, |
| "eval_loss": 1.880121111869812, |
| "eval_runtime": 207.849, |
| "eval_samples_per_second": 42.651, |
| "eval_steps_per_second": 1.525, |
| "step": 1200 |
| }, |
| { |
| "epoch": 0.45614035087719296, |
| "grad_norm": 2.076190710067749, |
| "learning_rate": 5.9545408512132104e-06, |
| "loss": 1.8773, |
| "step": 1300 |
| }, |
| { |
| "epoch": 0.49122807017543857, |
| "grad_norm": 1.5713521242141724, |
| "learning_rate": 5.391728922529762e-06, |
| "loss": 1.8761, |
| "step": 1400 |
| }, |
| { |
| "epoch": 0.5263157894736842, |
| "grad_norm": 1.424909234046936, |
| "learning_rate": 4.8238617546807515e-06, |
| "loss": 1.868, |
| "step": 1500 |
| }, |
| { |
| "epoch": 0.5614035087719298, |
| "grad_norm": 1.1872214078903198, |
| "learning_rate": 4.258267640734784e-06, |
| "loss": 1.8624, |
| "step": 1600 |
| }, |
| { |
| "epoch": 0.5964912280701754, |
| "grad_norm": 1.2520153522491455, |
| "learning_rate": 3.702245540130868e-06, |
| "loss": 1.8613, |
| "step": 1700 |
| }, |
| { |
| "epoch": 0.631578947368421, |
| "grad_norm": 1.815755009651184, |
| "learning_rate": 3.1629708860275753e-06, |
| "loss": 1.858, |
| "step": 1800 |
| }, |
| { |
| "epoch": 0.631578947368421, |
| "eval_loss": 1.8608616590499878, |
| "eval_runtime": 202.0389, |
| "eval_samples_per_second": 43.878, |
| "eval_steps_per_second": 1.569, |
| "step": 1800 |
| }, |
| { |
| "epoch": 0.6666666666666666, |
| "grad_norm": 1.4919559955596924, |
| "learning_rate": 2.647402986751724e-06, |
| "loss": 1.8568, |
| "step": 1900 |
| }, |
| { |
| "epoch": 0.7017543859649122, |
| "grad_norm": 1.4752705097198486, |
| "learning_rate": 2.162195216325574e-06, |
| "loss": 1.8567, |
| "step": 2000 |
| }, |
| { |
| "epoch": 0.7368421052631579, |
| "grad_norm": 1.4310791492462158, |
| "learning_rate": 1.7136091530585648e-06, |
| "loss": 1.8595, |
| "step": 2100 |
| }, |
| { |
| "epoch": 0.7719298245614035, |
| "grad_norm": 1.4420198202133179, |
| "learning_rate": 1.3074337742401638e-06, |
| "loss": 1.8555, |
| "step": 2200 |
| }, |
| { |
| "epoch": 0.8070175438596491, |
| "grad_norm": 1.2063950300216675, |
| "learning_rate": 9.489107497216321e-07, |
| "loss": 1.853, |
| "step": 2300 |
| }, |
| { |
| "epoch": 0.8421052631578947, |
| "grad_norm": 1.46232008934021, |
| "learning_rate": 6.426667984687179e-07, |
| "loss": 1.8549, |
| "step": 2400 |
| }, |
| { |
| "epoch": 0.8421052631578947, |
| "eval_loss": 1.854006052017212, |
| "eval_runtime": 209.5961, |
| "eval_samples_per_second": 42.296, |
| "eval_steps_per_second": 1.512, |
| "step": 2400 |
| }, |
| { |
| "epoch": 0.8771929824561403, |
| "grad_norm": 1.4567153453826904, |
| "learning_rate": 3.9265398102010576e-07, |
| "loss": 1.8519, |
| "step": 2500 |
| }, |
| { |
| "epoch": 0.9122807017543859, |
| "grad_norm": 1.1266202926635742, |
| "learning_rate": 2.020986983739559e-07, |
| "loss": 1.8476, |
| "step": 2600 |
| }, |
| { |
| "epoch": 0.9473684210526315, |
| "grad_norm": 1.330444097518921, |
| "learning_rate": 7.346005546900125e-08, |
| "loss": 1.8542, |
| "step": 2700 |
| }, |
| { |
| "epoch": 0.9824561403508771, |
| "grad_norm": 1.5576547384262085, |
| "learning_rate": 8.398126577059407e-09, |
| "loss": 1.8518, |
| "step": 2800 |
| }, |
| { |
| "epoch": 1.0, |
| "step": 2850, |
| "total_flos": 2.0843841016110776e+18, |
| "train_loss": 2.007047855310273, |
| "train_runtime": 5060.4296, |
| "train_samples_per_second": 31.534, |
| "train_steps_per_second": 0.563 |
| } |
| ], |
| "logging_steps": 100, |
| "max_steps": 2850, |
| "num_input_tokens_seen": 0, |
| "num_train_epochs": 1, |
| "save_steps": 2000, |
| "stateful_callbacks": { |
| "TrainerControl": { |
| "args": { |
| "should_epoch_stop": false, |
| "should_evaluate": false, |
| "should_log": false, |
| "should_save": true, |
| "should_training_stop": true |
| }, |
| "attributes": {} |
| } |
| }, |
| "total_flos": 2.0843841016110776e+18, |
| "train_batch_size": 4, |
| "trial_name": null, |
| "trial_params": null |
| } |
|
|
