| { | |
| "best_metric": null, | |
| "best_model_checkpoint": null, | |
| "epoch": 2.967741935483871, | |
| "eval_steps": 500, | |
| "global_step": 138, | |
| "is_hyper_param_search": false, | |
| "is_local_process_zero": true, | |
| "is_world_process_zero": true, | |
| "log_history": [ | |
| { | |
| "epoch": 0.21505376344086022, | |
| "grad_norm": 0.09814453125, | |
| "learning_rate": 0.0002, | |
| "loss": 1.8867, | |
| "step": 10 | |
| }, | |
| { | |
| "epoch": 0.43010752688172044, | |
| "grad_norm": 0.054443359375, | |
| "learning_rate": 0.0002, | |
| "loss": 1.6558, | |
| "step": 20 | |
| }, | |
| { | |
| "epoch": 0.6451612903225806, | |
| "grad_norm": 0.04833984375, | |
| "learning_rate": 0.0002, | |
| "loss": 1.6351, | |
| "step": 30 | |
| }, | |
| { | |
| "epoch": 0.8602150537634409, | |
| "grad_norm": 0.056884765625, | |
| "learning_rate": 0.0002, | |
| "loss": 1.623, | |
| "step": 40 | |
| }, | |
| { | |
| "epoch": 0.989247311827957, | |
| "eval_loss": 1.6275275945663452, | |
| "eval_runtime": 77.4322, | |
| "eval_samples_per_second": 4.068, | |
| "eval_steps_per_second": 0.517, | |
| "step": 46 | |
| }, | |
| { | |
| "epoch": 1.075268817204301, | |
| "grad_norm": 0.044677734375, | |
| "learning_rate": 0.0002, | |
| "loss": 1.6062, | |
| "step": 50 | |
| }, | |
| { | |
| "epoch": 1.2903225806451613, | |
| "grad_norm": 0.042724609375, | |
| "learning_rate": 0.0002, | |
| "loss": 1.6058, | |
| "step": 60 | |
| }, | |
| { | |
| "epoch": 1.5053763440860215, | |
| "grad_norm": 0.03759765625, | |
| "learning_rate": 0.0002, | |
| "loss": 1.5806, | |
| "step": 70 | |
| }, | |
| { | |
| "epoch": 1.7204301075268817, | |
| "grad_norm": 0.055419921875, | |
| "learning_rate": 0.0002, | |
| "loss": 1.6058, | |
| "step": 80 | |
| }, | |
| { | |
| "epoch": 1.935483870967742, | |
| "grad_norm": 0.037841796875, | |
| "learning_rate": 0.0002, | |
| "loss": 1.5825, | |
| "step": 90 | |
| }, | |
| { | |
| "epoch": 2.0, | |
| "eval_loss": 1.6166476011276245, | |
| "eval_runtime": 79.9595, | |
| "eval_samples_per_second": 3.939, | |
| "eval_steps_per_second": 0.5, | |
| "step": 93 | |
| }, | |
| { | |
| "epoch": 2.150537634408602, | |
| "grad_norm": 0.042236328125, | |
| "learning_rate": 0.0002, | |
| "loss": 1.5724, | |
| "step": 100 | |
| }, | |
| { | |
| "epoch": 2.3655913978494625, | |
| "grad_norm": 0.046875, | |
| "learning_rate": 0.0002, | |
| "loss": 1.5522, | |
| "step": 110 | |
| }, | |
| { | |
| "epoch": 2.5806451612903225, | |
| "grad_norm": 0.05029296875, | |
| "learning_rate": 0.0002, | |
| "loss": 1.569, | |
| "step": 120 | |
| }, | |
| { | |
| "epoch": 2.795698924731183, | |
| "grad_norm": 0.057861328125, | |
| "learning_rate": 0.0002, | |
| "loss": 1.5582, | |
| "step": 130 | |
| }, | |
| { | |
| "epoch": 2.967741935483871, | |
| "eval_loss": 1.6177211999893188, | |
| "eval_runtime": 77.5729, | |
| "eval_samples_per_second": 4.061, | |
| "eval_steps_per_second": 0.516, | |
| "step": 138 | |
| } | |
| ], | |
| "logging_steps": 10, | |
| "max_steps": 138, | |
| "num_input_tokens_seen": 0, | |
| "num_train_epochs": 3, | |
| "save_steps": 500, | |
| "total_flos": 3.8319031292390277e+18, | |
| "train_batch_size": 8, | |
| "trial_name": null, | |
| "trial_params": null | |
| } | |