| { | |
| "best_metric": null, | |
| "best_model_checkpoint": null, | |
| "epoch": 2.0408163265306123, | |
| "eval_steps": 5, | |
| "global_step": 50, | |
| "is_hyper_param_search": false, | |
| "is_local_process_zero": true, | |
| "is_world_process_zero": true, | |
| "log_history": [ | |
| { | |
| "epoch": 0.04081632653061224, | |
| "eval_loss": 11.601783752441406, | |
| "eval_runtime": 3.7438, | |
| "eval_samples_per_second": 11.218, | |
| "eval_steps_per_second": 1.603, | |
| "step": 1 | |
| }, | |
| { | |
| "epoch": 0.12244897959183673, | |
| "grad_norm": 36.64659881591797, | |
| "learning_rate": 3e-05, | |
| "loss": 11.4918, | |
| "step": 3 | |
| }, | |
| { | |
| "epoch": 0.20408163265306123, | |
| "eval_loss": 0.6978029608726501, | |
| "eval_runtime": 3.2607, | |
| "eval_samples_per_second": 12.881, | |
| "eval_steps_per_second": 1.84, | |
| "step": 5 | |
| }, | |
| { | |
| "epoch": 0.24489795918367346, | |
| "grad_norm": 18.22076988220215, | |
| "learning_rate": 6e-05, | |
| "loss": 4.8189, | |
| "step": 6 | |
| }, | |
| { | |
| "epoch": 0.3673469387755102, | |
| "grad_norm": 3.5493392944335938, | |
| "learning_rate": 9e-05, | |
| "loss": 0.1762, | |
| "step": 9 | |
| }, | |
| { | |
| "epoch": 0.40816326530612246, | |
| "eval_loss": 2.6691372394561768, | |
| "eval_runtime": 3.2757, | |
| "eval_samples_per_second": 12.822, | |
| "eval_steps_per_second": 1.832, | |
| "step": 10 | |
| }, | |
| { | |
| "epoch": 0.4897959183673469, | |
| "grad_norm": 21.09944725036621, | |
| "learning_rate": 9.938441702975689e-05, | |
| "loss": 1.0168, | |
| "step": 12 | |
| }, | |
| { | |
| "epoch": 0.6122448979591837, | |
| "grad_norm": 44.65542984008789, | |
| "learning_rate": 9.619397662556435e-05, | |
| "loss": 0.5669, | |
| "step": 15 | |
| }, | |
| { | |
| "epoch": 0.6122448979591837, | |
| "eval_loss": 0.3454976975917816, | |
| "eval_runtime": 3.2845, | |
| "eval_samples_per_second": 12.787, | |
| "eval_steps_per_second": 1.827, | |
| "step": 15 | |
| }, | |
| { | |
| "epoch": 0.7346938775510204, | |
| "grad_norm": 0.9508678913116455, | |
| "learning_rate": 9.045084971874738e-05, | |
| "loss": 0.1641, | |
| "step": 18 | |
| }, | |
| { | |
| "epoch": 0.8163265306122449, | |
| "eval_loss": 0.04804961383342743, | |
| "eval_runtime": 3.2862, | |
| "eval_samples_per_second": 12.781, | |
| "eval_steps_per_second": 1.826, | |
| "step": 20 | |
| }, | |
| { | |
| "epoch": 0.8571428571428571, | |
| "grad_norm": 1.182706594467163, | |
| "learning_rate": 8.247240241650918e-05, | |
| "loss": 0.0288, | |
| "step": 21 | |
| }, | |
| { | |
| "epoch": 0.9795918367346939, | |
| "grad_norm": 3.4299917221069336, | |
| "learning_rate": 7.269952498697734e-05, | |
| "loss": 0.092, | |
| "step": 24 | |
| }, | |
| { | |
| "epoch": 1.0204081632653061, | |
| "eval_loss": 0.03517497330904007, | |
| "eval_runtime": 3.2834, | |
| "eval_samples_per_second": 12.792, | |
| "eval_steps_per_second": 1.827, | |
| "step": 25 | |
| }, | |
| { | |
| "epoch": 1.1020408163265305, | |
| "grad_norm": 1.1145570278167725, | |
| "learning_rate": 6.167226819279528e-05, | |
| "loss": 0.0348, | |
| "step": 27 | |
| }, | |
| { | |
| "epoch": 1.2244897959183674, | |
| "grad_norm": 2.1262824535369873, | |
| "learning_rate": 5e-05, | |
| "loss": 0.0754, | |
| "step": 30 | |
| }, | |
| { | |
| "epoch": 1.2244897959183674, | |
| "eval_loss": 0.01700435020029545, | |
| "eval_runtime": 3.2868, | |
| "eval_samples_per_second": 12.778, | |
| "eval_steps_per_second": 1.825, | |
| "step": 30 | |
| }, | |
| { | |
| "epoch": 1.346938775510204, | |
| "grad_norm": 0.5034695863723755, | |
| "learning_rate": 3.832773180720475e-05, | |
| "loss": 0.0288, | |
| "step": 33 | |
| }, | |
| { | |
| "epoch": 1.4285714285714286, | |
| "eval_loss": 0.01104520633816719, | |
| "eval_runtime": 3.2861, | |
| "eval_samples_per_second": 12.781, | |
| "eval_steps_per_second": 1.826, | |
| "step": 35 | |
| }, | |
| { | |
| "epoch": 1.469387755102041, | |
| "grad_norm": 1.8535966873168945, | |
| "learning_rate": 2.7300475013022663e-05, | |
| "loss": 0.0725, | |
| "step": 36 | |
| }, | |
| { | |
| "epoch": 1.5918367346938775, | |
| "grad_norm": 2.2284762859344482, | |
| "learning_rate": 1.7527597583490822e-05, | |
| "loss": 0.051, | |
| "step": 39 | |
| }, | |
| { | |
| "epoch": 1.6326530612244898, | |
| "eval_loss": 0.008920343592762947, | |
| "eval_runtime": 3.293, | |
| "eval_samples_per_second": 12.754, | |
| "eval_steps_per_second": 1.822, | |
| "step": 40 | |
| }, | |
| { | |
| "epoch": 1.7142857142857144, | |
| "grad_norm": 0.635669469833374, | |
| "learning_rate": 9.549150281252633e-06, | |
| "loss": 0.0118, | |
| "step": 42 | |
| }, | |
| { | |
| "epoch": 1.836734693877551, | |
| "grad_norm": 0.5687570571899414, | |
| "learning_rate": 3.8060233744356633e-06, | |
| "loss": 0.0289, | |
| "step": 45 | |
| }, | |
| { | |
| "epoch": 1.836734693877551, | |
| "eval_loss": 0.009489011950790882, | |
| "eval_runtime": 3.2913, | |
| "eval_samples_per_second": 12.761, | |
| "eval_steps_per_second": 1.823, | |
| "step": 45 | |
| }, | |
| { | |
| "epoch": 1.9591836734693877, | |
| "grad_norm": 1.9518921375274658, | |
| "learning_rate": 6.15582970243117e-07, | |
| "loss": 0.0413, | |
| "step": 48 | |
| }, | |
| { | |
| "epoch": 2.0408163265306123, | |
| "eval_loss": 0.008860735222697258, | |
| "eval_runtime": 3.2883, | |
| "eval_samples_per_second": 12.773, | |
| "eval_steps_per_second": 1.825, | |
| "step": 50 | |
| } | |
| ], | |
| "logging_steps": 3, | |
| "max_steps": 50, | |
| "num_input_tokens_seen": 0, | |
| "num_train_epochs": 3, | |
| "save_steps": 5, | |
| "stateful_callbacks": { | |
| "TrainerControl": { | |
| "args": { | |
| "should_epoch_stop": false, | |
| "should_evaluate": false, | |
| "should_log": false, | |
| "should_save": true, | |
| "should_training_stop": true | |
| }, | |
| "attributes": {} | |
| } | |
| }, | |
| "total_flos": 7.523711473994957e+16, | |
| "train_batch_size": 8, | |
| "trial_name": null, | |
| "trial_params": null | |
| } | |