| { | |
| "best_metric": null, | |
| "best_model_checkpoint": null, | |
| "epoch": 7.68, | |
| "eval_steps": 500, | |
| "global_step": 48, | |
| "is_hyper_param_search": false, | |
| "is_local_process_zero": true, | |
| "is_world_process_zero": true, | |
| "log_history": [ | |
| { | |
| "epoch": 0.16, | |
| "grad_norm": 21.84854475798701, | |
| "learning_rate": 2.0000000000000003e-06, | |
| "loss": 2.7666, | |
| "step": 1 | |
| }, | |
| { | |
| "epoch": 0.32, | |
| "grad_norm": 21.08410666706505, | |
| "learning_rate": 4.000000000000001e-06, | |
| "loss": 2.7148, | |
| "step": 2 | |
| }, | |
| { | |
| "epoch": 0.48, | |
| "grad_norm": 19.817509197835424, | |
| "learning_rate": 6e-06, | |
| "loss": 2.7169, | |
| "step": 3 | |
| }, | |
| { | |
| "epoch": 0.64, | |
| "grad_norm": 14.56433231107304, | |
| "learning_rate": 8.000000000000001e-06, | |
| "loss": 2.5238, | |
| "step": 4 | |
| }, | |
| { | |
| "epoch": 0.8, | |
| "grad_norm": 7.910921625273512, | |
| "learning_rate": 1e-05, | |
| "loss": 2.3043, | |
| "step": 5 | |
| }, | |
| { | |
| "epoch": 0.96, | |
| "grad_norm": 8.065606760172034, | |
| "learning_rate": 1.2e-05, | |
| "loss": 2.1385, | |
| "step": 6 | |
| }, | |
| { | |
| "epoch": 1.12, | |
| "grad_norm": 6.92294371298597, | |
| "learning_rate": 1.4000000000000001e-05, | |
| "loss": 1.8466, | |
| "step": 7 | |
| }, | |
| { | |
| "epoch": 1.28, | |
| "grad_norm": 7.374155243701679, | |
| "learning_rate": 1.6000000000000003e-05, | |
| "loss": 1.696, | |
| "step": 8 | |
| }, | |
| { | |
| "epoch": 1.44, | |
| "grad_norm": 5.529647167446027, | |
| "learning_rate": 1.8e-05, | |
| "loss": 1.3348, | |
| "step": 9 | |
| }, | |
| { | |
| "epoch": 1.6, | |
| "grad_norm": 5.470706825652522, | |
| "learning_rate": 2e-05, | |
| "loss": 1.2157, | |
| "step": 10 | |
| }, | |
| { | |
| "epoch": 1.76, | |
| "grad_norm": 5.744223049944512, | |
| "learning_rate": 2.2000000000000003e-05, | |
| "loss": 1.0832, | |
| "step": 11 | |
| }, | |
| { | |
| "epoch": 1.92, | |
| "grad_norm": 7.00854296530452, | |
| "learning_rate": 2.4e-05, | |
| "loss": 0.9879, | |
| "step": 12 | |
| }, | |
| { | |
| "epoch": 2.08, | |
| "grad_norm": 5.336567904924906, | |
| "learning_rate": 2.6000000000000002e-05, | |
| "loss": 0.5616, | |
| "step": 13 | |
| }, | |
| { | |
| "epoch": 2.24, | |
| "grad_norm": 4.476173984946472, | |
| "learning_rate": 2.8000000000000003e-05, | |
| "loss": 0.3954, | |
| "step": 14 | |
| }, | |
| { | |
| "epoch": 2.4, | |
| "grad_norm": 4.943428331011744, | |
| "learning_rate": 3e-05, | |
| "loss": 0.4176, | |
| "step": 15 | |
| }, | |
| { | |
| "epoch": 2.56, | |
| "grad_norm": 4.390162860701155, | |
| "learning_rate": 3.2000000000000005e-05, | |
| "loss": 0.3047, | |
| "step": 16 | |
| }, | |
| { | |
| "epoch": 2.7199999999999998, | |
| "grad_norm": 3.6439162726811514, | |
| "learning_rate": 3.4000000000000007e-05, | |
| "loss": 0.3502, | |
| "step": 17 | |
| }, | |
| { | |
| "epoch": 2.88, | |
| "grad_norm": 3.7835908637146614, | |
| "learning_rate": 3.6e-05, | |
| "loss": 0.4052, | |
| "step": 18 | |
| }, | |
| { | |
| "epoch": 3.04, | |
| "grad_norm": 3.045882254850862, | |
| "learning_rate": 3.8e-05, | |
| "loss": 0.299, | |
| "step": 19 | |
| }, | |
| { | |
| "epoch": 3.2, | |
| "grad_norm": 3.026990390481408, | |
| "learning_rate": 4e-05, | |
| "loss": 0.1532, | |
| "step": 20 | |
| }, | |
| { | |
| "epoch": 3.36, | |
| "grad_norm": 2.7854589808693317, | |
| "learning_rate": 4.2e-05, | |
| "loss": 0.1184, | |
| "step": 21 | |
| }, | |
| { | |
| "epoch": 3.52, | |
| "grad_norm": 3.934431288327914, | |
| "learning_rate": 4.4000000000000006e-05, | |
| "loss": 0.1605, | |
| "step": 22 | |
| }, | |
| { | |
| "epoch": 3.68, | |
| "grad_norm": 2.5384235711882246, | |
| "learning_rate": 4.600000000000001e-05, | |
| "loss": 0.1256, | |
| "step": 23 | |
| }, | |
| { | |
| "epoch": 3.84, | |
| "grad_norm": 3.0549054370297046, | |
| "learning_rate": 4.8e-05, | |
| "loss": 0.1509, | |
| "step": 24 | |
| }, | |
| { | |
| "epoch": 4.0, | |
| "grad_norm": 2.742034577764137, | |
| "learning_rate": 5e-05, | |
| "loss": 0.1955, | |
| "step": 25 | |
| }, | |
| { | |
| "epoch": 4.16, | |
| "grad_norm": 2.7052086450717185, | |
| "learning_rate": 5.2000000000000004e-05, | |
| "loss": 0.0883, | |
| "step": 26 | |
| }, | |
| { | |
| "epoch": 4.32, | |
| "grad_norm": 2.141311755890699, | |
| "learning_rate": 5.4000000000000005e-05, | |
| "loss": 0.0982, | |
| "step": 27 | |
| }, | |
| { | |
| "epoch": 4.48, | |
| "grad_norm": 2.405168716519016, | |
| "learning_rate": 5.6000000000000006e-05, | |
| "loss": 0.1167, | |
| "step": 28 | |
| }, | |
| { | |
| "epoch": 4.64, | |
| "grad_norm": 2.1040576646851976, | |
| "learning_rate": 5.8e-05, | |
| "loss": 0.1011, | |
| "step": 29 | |
| }, | |
| { | |
| "epoch": 4.8, | |
| "grad_norm": 2.1434366495348742, | |
| "learning_rate": 6e-05, | |
| "loss": 0.1093, | |
| "step": 30 | |
| }, | |
| { | |
| "epoch": 4.96, | |
| "grad_norm": 2.4885010419741356, | |
| "learning_rate": 6.2e-05, | |
| "loss": 0.126, | |
| "step": 31 | |
| }, | |
| { | |
| "epoch": 5.12, | |
| "grad_norm": 2.5185987231548705, | |
| "learning_rate": 6.400000000000001e-05, | |
| "loss": 0.0798, | |
| "step": 32 | |
| }, | |
| { | |
| "epoch": 5.28, | |
| "grad_norm": 2.4229963679451823, | |
| "learning_rate": 6.6e-05, | |
| "loss": 0.0777, | |
| "step": 33 | |
| }, | |
| { | |
| "epoch": 5.44, | |
| "grad_norm": 2.35757612847466, | |
| "learning_rate": 6.800000000000001e-05, | |
| "loss": 0.1035, | |
| "step": 34 | |
| }, | |
| { | |
| "epoch": 5.6, | |
| "grad_norm": 2.261180396997714, | |
| "learning_rate": 7e-05, | |
| "loss": 0.095, | |
| "step": 35 | |
| }, | |
| { | |
| "epoch": 5.76, | |
| "grad_norm": 2.397540189265323, | |
| "learning_rate": 7.2e-05, | |
| "loss": 0.1024, | |
| "step": 36 | |
| }, | |
| { | |
| "epoch": 5.92, | |
| "grad_norm": 2.440377463659988, | |
| "learning_rate": 7.4e-05, | |
| "loss": 0.1276, | |
| "step": 37 | |
| }, | |
| { | |
| "epoch": 6.08, | |
| "grad_norm": 1.9038257788340682, | |
| "learning_rate": 7.6e-05, | |
| "loss": 0.0839, | |
| "step": 38 | |
| }, | |
| { | |
| "epoch": 6.24, | |
| "grad_norm": 3.613615935736148, | |
| "learning_rate": 7.800000000000001e-05, | |
| "loss": 0.061, | |
| "step": 39 | |
| }, | |
| { | |
| "epoch": 6.4, | |
| "grad_norm": 2.225128405341641, | |
| "learning_rate": 8e-05, | |
| "loss": 0.0956, | |
| "step": 40 | |
| }, | |
| { | |
| "epoch": 6.5600000000000005, | |
| "grad_norm": 2.184522719061086, | |
| "learning_rate": 8.2e-05, | |
| "loss": 0.0882, | |
| "step": 41 | |
| }, | |
| { | |
| "epoch": 6.72, | |
| "grad_norm": 2.865012553917456, | |
| "learning_rate": 8.4e-05, | |
| "loss": 0.1075, | |
| "step": 42 | |
| }, | |
| { | |
| "epoch": 6.88, | |
| "grad_norm": 2.4858897371258255, | |
| "learning_rate": 8.6e-05, | |
| "loss": 0.1397, | |
| "step": 43 | |
| }, | |
| { | |
| "epoch": 7.04, | |
| "grad_norm": 2.5069104688577597, | |
| "learning_rate": 8.800000000000001e-05, | |
| "loss": 0.1138, | |
| "step": 44 | |
| }, | |
| { | |
| "epoch": 7.2, | |
| "grad_norm": 2.0377663889161544, | |
| "learning_rate": 9e-05, | |
| "loss": 0.0679, | |
| "step": 45 | |
| }, | |
| { | |
| "epoch": 7.36, | |
| "grad_norm": 1.7737003765568096, | |
| "learning_rate": 9.200000000000001e-05, | |
| "loss": 0.0625, | |
| "step": 46 | |
| }, | |
| { | |
| "epoch": 7.52, | |
| "grad_norm": 2.546223689211966, | |
| "learning_rate": 9.4e-05, | |
| "loss": 0.107, | |
| "step": 47 | |
| }, | |
| { | |
| "epoch": 7.68, | |
| "grad_norm": 2.629646137634436, | |
| "learning_rate": 9.6e-05, | |
| "loss": 0.1055, | |
| "step": 48 | |
| } | |
| ], | |
| "logging_steps": 1, | |
| "max_steps": 48, | |
| "num_input_tokens_seen": 0, | |
| "num_train_epochs": 8, | |
| "save_steps": 1, | |
| "stateful_callbacks": { | |
| "TrainerControl": { | |
| "args": { | |
| "should_epoch_stop": false, | |
| "should_evaluate": false, | |
| "should_log": false, | |
| "should_save": true, | |
| "should_training_stop": true | |
| }, | |
| "attributes": {} | |
| } | |
| }, | |
| "total_flos": 39991514234880.0, | |
| "train_batch_size": 4, | |
| "trial_name": null, | |
| "trial_params": null | |
| } | |