| { | |
| "best_metric": 0.9311172366142273, | |
| "best_model_checkpoint": "./outputs/checkpoint-2100", | |
| "epoch": 2.8, | |
| "eval_steps": 100, | |
| "global_step": 2100, | |
| "is_hyper_param_search": false, | |
| "is_local_process_zero": true, | |
| "is_world_process_zero": true, | |
| "log_history": [ | |
| { | |
| "epoch": 0.13, | |
| "learning_rate": 0.0002, | |
| "loss": 1.0007, | |
| "step": 100 | |
| }, | |
| { | |
| "epoch": 0.13, | |
| "eval_loss": 1.0926419496536255, | |
| "eval_runtime": 890.2071, | |
| "eval_samples_per_second": 2.168, | |
| "eval_steps_per_second": 0.272, | |
| "step": 100 | |
| }, | |
| { | |
| "epoch": 0.27, | |
| "learning_rate": 0.0002, | |
| "loss": 0.897, | |
| "step": 200 | |
| }, | |
| { | |
| "epoch": 0.27, | |
| "eval_loss": 1.0682977437973022, | |
| "eval_runtime": 890.4733, | |
| "eval_samples_per_second": 2.167, | |
| "eval_steps_per_second": 0.272, | |
| "step": 200 | |
| }, | |
| { | |
| "epoch": 0.4, | |
| "learning_rate": 0.0002, | |
| "loss": 0.8814, | |
| "step": 300 | |
| }, | |
| { | |
| "epoch": 0.4, | |
| "eval_loss": 1.0527396202087402, | |
| "eval_runtime": 888.2495, | |
| "eval_samples_per_second": 2.173, | |
| "eval_steps_per_second": 0.272, | |
| "step": 300 | |
| }, | |
| { | |
| "epoch": 0.53, | |
| "learning_rate": 0.0002, | |
| "loss": 0.8697, | |
| "step": 400 | |
| }, | |
| { | |
| "epoch": 0.53, | |
| "eval_loss": 1.039888620376587, | |
| "eval_runtime": 883.1791, | |
| "eval_samples_per_second": 2.185, | |
| "eval_steps_per_second": 0.274, | |
| "step": 400 | |
| }, | |
| { | |
| "epoch": 0.67, | |
| "learning_rate": 0.0002, | |
| "loss": 0.8552, | |
| "step": 500 | |
| }, | |
| { | |
| "epoch": 0.67, | |
| "eval_loss": 1.0277502536773682, | |
| "eval_runtime": 724.238, | |
| "eval_samples_per_second": 2.665, | |
| "eval_steps_per_second": 0.334, | |
| "step": 500 | |
| }, | |
| { | |
| "epoch": 0.8, | |
| "learning_rate": 0.0002, | |
| "loss": 0.85, | |
| "step": 600 | |
| }, | |
| { | |
| "epoch": 0.8, | |
| "eval_loss": 1.0173273086547852, | |
| "eval_runtime": 724.2021, | |
| "eval_samples_per_second": 2.665, | |
| "eval_steps_per_second": 0.334, | |
| "step": 600 | |
| }, | |
| { | |
| "epoch": 0.93, | |
| "learning_rate": 0.0002, | |
| "loss": 0.8449, | |
| "step": 700 | |
| }, | |
| { | |
| "epoch": 0.93, | |
| "eval_loss": 1.012062430381775, | |
| "eval_runtime": 728.824, | |
| "eval_samples_per_second": 2.648, | |
| "eval_steps_per_second": 0.332, | |
| "step": 700 | |
| }, | |
| { | |
| "epoch": 1.07, | |
| "learning_rate": 0.0002, | |
| "loss": 0.8282, | |
| "step": 800 | |
| }, | |
| { | |
| "epoch": 1.07, | |
| "eval_loss": 1.001969337463379, | |
| "eval_runtime": 726.4944, | |
| "eval_samples_per_second": 2.657, | |
| "eval_steps_per_second": 0.333, | |
| "step": 800 | |
| }, | |
| { | |
| "epoch": 1.2, | |
| "learning_rate": 0.0002, | |
| "loss": 0.8145, | |
| "step": 900 | |
| }, | |
| { | |
| "epoch": 1.2, | |
| "eval_loss": 0.9950811266899109, | |
| "eval_runtime": 725.592, | |
| "eval_samples_per_second": 2.66, | |
| "eval_steps_per_second": 0.334, | |
| "step": 900 | |
| }, | |
| { | |
| "epoch": 1.33, | |
| "learning_rate": 0.0002, | |
| "loss": 0.8091, | |
| "step": 1000 | |
| }, | |
| { | |
| "epoch": 1.33, | |
| "eval_loss": 0.9874381422996521, | |
| "eval_runtime": 724.5381, | |
| "eval_samples_per_second": 2.664, | |
| "eval_steps_per_second": 0.334, | |
| "step": 1000 | |
| }, | |
| { | |
| "epoch": 1.47, | |
| "learning_rate": 0.0002, | |
| "loss": 0.8156, | |
| "step": 1100 | |
| }, | |
| { | |
| "epoch": 1.47, | |
| "eval_loss": 0.9812790751457214, | |
| "eval_runtime": 726.6666, | |
| "eval_samples_per_second": 2.656, | |
| "eval_steps_per_second": 0.333, | |
| "step": 1100 | |
| }, | |
| { | |
| "epoch": 1.6, | |
| "learning_rate": 0.0002, | |
| "loss": 0.8055, | |
| "step": 1200 | |
| }, | |
| { | |
| "epoch": 1.6, | |
| "eval_loss": 0.9755277633666992, | |
| "eval_runtime": 725.2603, | |
| "eval_samples_per_second": 2.661, | |
| "eval_steps_per_second": 0.334, | |
| "step": 1200 | |
| }, | |
| { | |
| "epoch": 1.73, | |
| "learning_rate": 0.0002, | |
| "loss": 0.7951, | |
| "step": 1300 | |
| }, | |
| { | |
| "epoch": 1.73, | |
| "eval_loss": 0.9684687852859497, | |
| "eval_runtime": 731.5971, | |
| "eval_samples_per_second": 2.638, | |
| "eval_steps_per_second": 0.331, | |
| "step": 1300 | |
| }, | |
| { | |
| "epoch": 1.87, | |
| "learning_rate": 0.0002, | |
| "loss": 0.7943, | |
| "step": 1400 | |
| }, | |
| { | |
| "epoch": 1.87, | |
| "eval_loss": 0.9623979330062866, | |
| "eval_runtime": 731.9495, | |
| "eval_samples_per_second": 2.637, | |
| "eval_steps_per_second": 0.331, | |
| "step": 1400 | |
| }, | |
| { | |
| "epoch": 2.0, | |
| "learning_rate": 0.0002, | |
| "loss": 0.7869, | |
| "step": 1500 | |
| }, | |
| { | |
| "epoch": 2.0, | |
| "eval_loss": 0.9567301869392395, | |
| "eval_runtime": 732.3773, | |
| "eval_samples_per_second": 2.635, | |
| "eval_steps_per_second": 0.33, | |
| "step": 1500 | |
| }, | |
| { | |
| "epoch": 2.13, | |
| "learning_rate": 0.0002, | |
| "loss": 0.7669, | |
| "step": 1600 | |
| }, | |
| { | |
| "epoch": 2.13, | |
| "eval_loss": 0.9550666213035583, | |
| "eval_runtime": 729.6316, | |
| "eval_samples_per_second": 2.645, | |
| "eval_steps_per_second": 0.332, | |
| "step": 1600 | |
| }, | |
| { | |
| "epoch": 2.27, | |
| "learning_rate": 0.0002, | |
| "loss": 0.7576, | |
| "step": 1700 | |
| }, | |
| { | |
| "epoch": 2.27, | |
| "eval_loss": 0.9503656029701233, | |
| "eval_runtime": 913.453, | |
| "eval_samples_per_second": 2.113, | |
| "eval_steps_per_second": 0.265, | |
| "step": 1700 | |
| }, | |
| { | |
| "epoch": 2.4, | |
| "learning_rate": 0.0002, | |
| "loss": 0.7579, | |
| "step": 1800 | |
| }, | |
| { | |
| "epoch": 2.4, | |
| "eval_loss": 0.9474750757217407, | |
| "eval_runtime": 916.0603, | |
| "eval_samples_per_second": 2.107, | |
| "eval_steps_per_second": 0.264, | |
| "step": 1800 | |
| }, | |
| { | |
| "epoch": 2.53, | |
| "learning_rate": 0.0002, | |
| "loss": 0.7705, | |
| "step": 1900 | |
| }, | |
| { | |
| "epoch": 2.53, | |
| "eval_loss": 0.9419649839401245, | |
| "eval_runtime": 913.3902, | |
| "eval_samples_per_second": 2.113, | |
| "eval_steps_per_second": 0.265, | |
| "step": 1900 | |
| }, | |
| { | |
| "epoch": 2.67, | |
| "learning_rate": 0.0002, | |
| "loss": 0.7581, | |
| "step": 2000 | |
| }, | |
| { | |
| "epoch": 2.67, | |
| "eval_loss": 0.9371306300163269, | |
| "eval_runtime": 913.4597, | |
| "eval_samples_per_second": 2.113, | |
| "eval_steps_per_second": 0.265, | |
| "step": 2000 | |
| }, | |
| { | |
| "epoch": 2.8, | |
| "learning_rate": 0.0002, | |
| "loss": 0.7534, | |
| "step": 2100 | |
| }, | |
| { | |
| "epoch": 2.8, | |
| "eval_loss": 0.9311172366142273, | |
| "eval_runtime": 913.8258, | |
| "eval_samples_per_second": 2.112, | |
| "eval_steps_per_second": 0.265, | |
| "step": 2100 | |
| } | |
| ], | |
| "logging_steps": 100, | |
| "max_steps": 2250, | |
| "num_input_tokens_seen": 0, | |
| "num_train_epochs": 3, | |
| "save_steps": 100, | |
| "total_flos": 5.4402852955938816e+17, | |
| "trial_name": null, | |
| "trial_params": null | |
| } | |