| { | |
| "best_metric": null, | |
| "best_model_checkpoint": null, | |
| "epoch": 0.026698804116065636, | |
| "eval_steps": 2, | |
| "global_step": 18, | |
| "is_hyper_param_search": false, | |
| "is_local_process_zero": true, | |
| "is_world_process_zero": true, | |
| "log_history": [ | |
| { | |
| "epoch": 0.0014832668953369797, | |
| "grad_norm": 20.178142547607422, | |
| "learning_rate": 2e-05, | |
| "loss": 43.4533, | |
| "step": 1 | |
| }, | |
| { | |
| "epoch": 0.0014832668953369797, | |
| "eval_loss": 2.656465768814087, | |
| "eval_runtime": 154.8536, | |
| "eval_samples_per_second": 3.668, | |
| "eval_steps_per_second": 3.668, | |
| "step": 1 | |
| }, | |
| { | |
| "epoch": 0.0029665337906739594, | |
| "grad_norm": 17.634891510009766, | |
| "learning_rate": 4e-05, | |
| "loss": 39.2663, | |
| "step": 2 | |
| }, | |
| { | |
| "epoch": 0.0029665337906739594, | |
| "eval_loss": 2.655576705932617, | |
| "eval_runtime": 154.4307, | |
| "eval_samples_per_second": 3.678, | |
| "eval_steps_per_second": 3.678, | |
| "step": 2 | |
| }, | |
| { | |
| "epoch": 0.004449800686010939, | |
| "grad_norm": 21.489526748657227, | |
| "learning_rate": 6e-05, | |
| "loss": 40.7739, | |
| "step": 3 | |
| }, | |
| { | |
| "epoch": 0.005933067581347919, | |
| "grad_norm": 23.668588638305664, | |
| "learning_rate": 8e-05, | |
| "loss": 39.5743, | |
| "step": 4 | |
| }, | |
| { | |
| "epoch": 0.005933067581347919, | |
| "eval_loss": 2.6245453357696533, | |
| "eval_runtime": 150.3826, | |
| "eval_samples_per_second": 3.777, | |
| "eval_steps_per_second": 3.777, | |
| "step": 4 | |
| }, | |
| { | |
| "epoch": 0.007416334476684898, | |
| "grad_norm": 25.051509857177734, | |
| "learning_rate": 0.0001, | |
| "loss": 43.102, | |
| "step": 5 | |
| }, | |
| { | |
| "epoch": 0.008899601372021879, | |
| "grad_norm": 29.83518409729004, | |
| "learning_rate": 0.00012, | |
| "loss": 37.8648, | |
| "step": 6 | |
| }, | |
| { | |
| "epoch": 0.008899601372021879, | |
| "eval_loss": 2.4578418731689453, | |
| "eval_runtime": 150.3798, | |
| "eval_samples_per_second": 3.777, | |
| "eval_steps_per_second": 3.777, | |
| "step": 6 | |
| }, | |
| { | |
| "epoch": 0.010382868267358857, | |
| "grad_norm": 47.01485061645508, | |
| "learning_rate": 0.00014, | |
| "loss": 38.7341, | |
| "step": 7 | |
| }, | |
| { | |
| "epoch": 0.011866135162695838, | |
| "grad_norm": 45.00392532348633, | |
| "learning_rate": 0.00016, | |
| "loss": 36.6476, | |
| "step": 8 | |
| }, | |
| { | |
| "epoch": 0.011866135162695838, | |
| "eval_loss": 1.9450047016143799, | |
| "eval_runtime": 150.402, | |
| "eval_samples_per_second": 3.777, | |
| "eval_steps_per_second": 3.777, | |
| "step": 8 | |
| }, | |
| { | |
| "epoch": 0.013349402058032818, | |
| "grad_norm": 52.0974006652832, | |
| "learning_rate": 0.00018, | |
| "loss": 31.804, | |
| "step": 9 | |
| }, | |
| { | |
| "epoch": 0.014832668953369797, | |
| "grad_norm": 42.790775299072266, | |
| "learning_rate": 0.0002, | |
| "loss": 23.4278, | |
| "step": 10 | |
| }, | |
| { | |
| "epoch": 0.014832668953369797, | |
| "eval_loss": 1.1998307704925537, | |
| "eval_runtime": 151.4492, | |
| "eval_samples_per_second": 3.75, | |
| "eval_steps_per_second": 3.75, | |
| "step": 10 | |
| }, | |
| { | |
| "epoch": 0.016315935848706775, | |
| "grad_norm": 54.17241668701172, | |
| "learning_rate": 0.00019510565162951537, | |
| "loss": 18.5255, | |
| "step": 11 | |
| }, | |
| { | |
| "epoch": 0.017799202744043757, | |
| "grad_norm": 42.78046417236328, | |
| "learning_rate": 0.00018090169943749476, | |
| "loss": 11.9862, | |
| "step": 12 | |
| }, | |
| { | |
| "epoch": 0.017799202744043757, | |
| "eval_loss": 0.5600711703300476, | |
| "eval_runtime": 151.2427, | |
| "eval_samples_per_second": 3.756, | |
| "eval_steps_per_second": 3.756, | |
| "step": 12 | |
| }, | |
| { | |
| "epoch": 0.019282469639380736, | |
| "grad_norm": 31.308900833129883, | |
| "learning_rate": 0.00015877852522924732, | |
| "loss": 11.0217, | |
| "step": 13 | |
| }, | |
| { | |
| "epoch": 0.020765736534717714, | |
| "grad_norm": 26.355045318603516, | |
| "learning_rate": 0.00013090169943749476, | |
| "loss": 4.9234, | |
| "step": 14 | |
| }, | |
| { | |
| "epoch": 0.020765736534717714, | |
| "eval_loss": 0.39810341596603394, | |
| "eval_runtime": 151.4481, | |
| "eval_samples_per_second": 3.75, | |
| "eval_steps_per_second": 3.75, | |
| "step": 14 | |
| }, | |
| { | |
| "epoch": 0.022249003430054697, | |
| "grad_norm": 26.284854888916016, | |
| "learning_rate": 0.0001, | |
| "loss": 6.4574, | |
| "step": 15 | |
| }, | |
| { | |
| "epoch": 0.023732270325391675, | |
| "grad_norm": 35.7542839050293, | |
| "learning_rate": 6.909830056250527e-05, | |
| "loss": 9.4431, | |
| "step": 16 | |
| }, | |
| { | |
| "epoch": 0.023732270325391675, | |
| "eval_loss": 0.36851683259010315, | |
| "eval_runtime": 151.278, | |
| "eval_samples_per_second": 3.755, | |
| "eval_steps_per_second": 3.755, | |
| "step": 16 | |
| }, | |
| { | |
| "epoch": 0.025215537220728654, | |
| "grad_norm": 25.165361404418945, | |
| "learning_rate": 4.12214747707527e-05, | |
| "loss": 4.9765, | |
| "step": 17 | |
| }, | |
| { | |
| "epoch": 0.026698804116065636, | |
| "grad_norm": 14.660453796386719, | |
| "learning_rate": 1.9098300562505266e-05, | |
| "loss": 1.5801, | |
| "step": 18 | |
| }, | |
| { | |
| "epoch": 0.026698804116065636, | |
| "eval_loss": 0.34035831689834595, | |
| "eval_runtime": 151.1958, | |
| "eval_samples_per_second": 3.757, | |
| "eval_steps_per_second": 3.757, | |
| "step": 18 | |
| } | |
| ], | |
| "logging_steps": 1, | |
| "max_steps": 20, | |
| "num_input_tokens_seen": 0, | |
| "num_train_epochs": 1, | |
| "save_steps": 2, | |
| "stateful_callbacks": { | |
| "TrainerControl": { | |
| "args": { | |
| "should_epoch_stop": false, | |
| "should_evaluate": false, | |
| "should_log": false, | |
| "should_save": true, | |
| "should_training_stop": false | |
| }, | |
| "attributes": {} | |
| } | |
| }, | |
| "total_flos": 2927327503712256.0, | |
| "train_batch_size": 1, | |
| "trial_name": null, | |
| "trial_params": null | |
| } | |