| { | |
| "best_metric": 0.5989616200729698, | |
| "best_model_checkpoint": "models_intermediate/intermediate_it5-small/checkpoint-50000", | |
| "epoch": 25.0, | |
| "eval_steps": 10000, | |
| "global_step": 50000, | |
| "is_hyper_param_search": false, | |
| "is_local_process_zero": true, | |
| "is_world_process_zero": true, | |
| "log_history": [ | |
| { | |
| "epoch": 0.5, | |
| "learning_rate": 3.9200000000000004e-05, | |
| "loss": 4.7009, | |
| "step": 1000 | |
| }, | |
| { | |
| "epoch": 1.0, | |
| "learning_rate": 3.8400000000000005e-05, | |
| "loss": 1.7225, | |
| "step": 2000 | |
| }, | |
| { | |
| "epoch": 1.5, | |
| "learning_rate": 3.76e-05, | |
| "loss": 0.5391, | |
| "step": 3000 | |
| }, | |
| { | |
| "epoch": 2.0, | |
| "learning_rate": 3.680000000000001e-05, | |
| "loss": 0.4553, | |
| "step": 4000 | |
| }, | |
| { | |
| "epoch": 2.5, | |
| "learning_rate": 3.6e-05, | |
| "loss": 0.4145, | |
| "step": 5000 | |
| }, | |
| { | |
| "epoch": 3.0, | |
| "learning_rate": 3.52e-05, | |
| "loss": 0.39, | |
| "step": 6000 | |
| }, | |
| { | |
| "epoch": 3.5, | |
| "learning_rate": 3.44e-05, | |
| "loss": 0.3579, | |
| "step": 7000 | |
| }, | |
| { | |
| "epoch": 4.0, | |
| "learning_rate": 3.3600000000000004e-05, | |
| "loss": 0.3514, | |
| "step": 8000 | |
| }, | |
| { | |
| "epoch": 4.5, | |
| "learning_rate": 3.28e-05, | |
| "loss": 0.329, | |
| "step": 9000 | |
| }, | |
| { | |
| "epoch": 5.0, | |
| "learning_rate": 3.2000000000000005e-05, | |
| "loss": 0.3193, | |
| "step": 10000 | |
| }, | |
| { | |
| "epoch": 5.0, | |
| "eval_loss": 0.24682821333408356, | |
| "eval_p-value": 1.381553198766455e-280, | |
| "eval_runtime": 76.6844, | |
| "eval_samples_per_second": 52.162, | |
| "eval_spearman": 0.5236525970894446, | |
| "eval_steps_per_second": 6.52, | |
| "step": 10000 | |
| }, | |
| { | |
| "epoch": 5.5, | |
| "learning_rate": 3.1200000000000006e-05, | |
| "loss": 0.3036, | |
| "step": 11000 | |
| }, | |
| { | |
| "epoch": 6.0, | |
| "learning_rate": 3.0400000000000004e-05, | |
| "loss": 0.3049, | |
| "step": 12000 | |
| }, | |
| { | |
| "epoch": 6.5, | |
| "learning_rate": 2.96e-05, | |
| "loss": 0.2874, | |
| "step": 13000 | |
| }, | |
| { | |
| "epoch": 7.0, | |
| "learning_rate": 2.8800000000000002e-05, | |
| "loss": 0.2789, | |
| "step": 14000 | |
| }, | |
| { | |
| "epoch": 7.5, | |
| "learning_rate": 2.8e-05, | |
| "loss": 0.2758, | |
| "step": 15000 | |
| }, | |
| { | |
| "epoch": 8.0, | |
| "learning_rate": 2.7200000000000004e-05, | |
| "loss": 0.2708, | |
| "step": 16000 | |
| }, | |
| { | |
| "epoch": 8.5, | |
| "learning_rate": 2.6400000000000005e-05, | |
| "loss": 0.2677, | |
| "step": 17000 | |
| }, | |
| { | |
| "epoch": 9.0, | |
| "learning_rate": 2.5600000000000002e-05, | |
| "loss": 0.2547, | |
| "step": 18000 | |
| }, | |
| { | |
| "epoch": 9.5, | |
| "learning_rate": 2.4800000000000003e-05, | |
| "loss": 0.2597, | |
| "step": 19000 | |
| }, | |
| { | |
| "epoch": 10.0, | |
| "learning_rate": 2.4e-05, | |
| "loss": 0.2518, | |
| "step": 20000 | |
| }, | |
| { | |
| "epoch": 10.0, | |
| "eval_loss": 0.20981653034687042, | |
| "eval_p-value": 5.53e-322, | |
| "eval_runtime": 73.524, | |
| "eval_samples_per_second": 54.404, | |
| "eval_spearman": 0.5549641467300505, | |
| "eval_steps_per_second": 6.8, | |
| "step": 20000 | |
| }, | |
| { | |
| "epoch": 10.5, | |
| "learning_rate": 2.32e-05, | |
| "loss": 0.2497, | |
| "step": 21000 | |
| }, | |
| { | |
| "epoch": 11.0, | |
| "learning_rate": 2.2400000000000002e-05, | |
| "loss": 0.2435, | |
| "step": 22000 | |
| }, | |
| { | |
| "epoch": 11.5, | |
| "learning_rate": 2.1600000000000003e-05, | |
| "loss": 0.2442, | |
| "step": 23000 | |
| }, | |
| { | |
| "epoch": 12.0, | |
| "learning_rate": 2.08e-05, | |
| "loss": 0.2369, | |
| "step": 24000 | |
| }, | |
| { | |
| "epoch": 12.5, | |
| "learning_rate": 2e-05, | |
| "loss": 0.2412, | |
| "step": 25000 | |
| }, | |
| { | |
| "epoch": 13.0, | |
| "learning_rate": 1.9200000000000003e-05, | |
| "loss": 0.2298, | |
| "step": 26000 | |
| }, | |
| { | |
| "epoch": 13.5, | |
| "learning_rate": 1.8400000000000003e-05, | |
| "loss": 0.2282, | |
| "step": 27000 | |
| }, | |
| { | |
| "epoch": 14.0, | |
| "learning_rate": 1.76e-05, | |
| "loss": 0.232, | |
| "step": 28000 | |
| }, | |
| { | |
| "epoch": 14.5, | |
| "learning_rate": 1.6800000000000002e-05, | |
| "loss": 0.2302, | |
| "step": 29000 | |
| }, | |
| { | |
| "epoch": 15.0, | |
| "learning_rate": 1.6000000000000003e-05, | |
| "loss": 0.2264, | |
| "step": 30000 | |
| }, | |
| { | |
| "epoch": 15.0, | |
| "eval_loss": 0.19759084284305573, | |
| "eval_p-value": 0.0, | |
| "eval_runtime": 44.4809, | |
| "eval_samples_per_second": 89.926, | |
| "eval_spearman": 0.5808866463657348, | |
| "eval_steps_per_second": 11.241, | |
| "step": 30000 | |
| }, | |
| { | |
| "epoch": 15.5, | |
| "learning_rate": 1.5200000000000002e-05, | |
| "loss": 0.2223, | |
| "step": 31000 | |
| }, | |
| { | |
| "epoch": 16.0, | |
| "learning_rate": 1.4400000000000001e-05, | |
| "loss": 0.2232, | |
| "step": 32000 | |
| }, | |
| { | |
| "epoch": 16.5, | |
| "learning_rate": 1.3600000000000002e-05, | |
| "loss": 0.2188, | |
| "step": 33000 | |
| }, | |
| { | |
| "epoch": 17.0, | |
| "learning_rate": 1.2800000000000001e-05, | |
| "loss": 0.2257, | |
| "step": 34000 | |
| }, | |
| { | |
| "epoch": 17.5, | |
| "learning_rate": 1.2e-05, | |
| "loss": 0.2191, | |
| "step": 35000 | |
| }, | |
| { | |
| "epoch": 18.0, | |
| "learning_rate": 1.1200000000000001e-05, | |
| "loss": 0.2166, | |
| "step": 36000 | |
| }, | |
| { | |
| "epoch": 18.5, | |
| "learning_rate": 1.04e-05, | |
| "loss": 0.2177, | |
| "step": 37000 | |
| }, | |
| { | |
| "epoch": 19.0, | |
| "learning_rate": 9.600000000000001e-06, | |
| "loss": 0.2136, | |
| "step": 38000 | |
| }, | |
| { | |
| "epoch": 19.5, | |
| "learning_rate": 8.8e-06, | |
| "loss": 0.2133, | |
| "step": 39000 | |
| }, | |
| { | |
| "epoch": 20.0, | |
| "learning_rate": 8.000000000000001e-06, | |
| "loss": 0.2127, | |
| "step": 40000 | |
| }, | |
| { | |
| "epoch": 20.0, | |
| "eval_loss": 0.19654138386249542, | |
| "eval_p-value": 0.0, | |
| "eval_runtime": 44.5401, | |
| "eval_samples_per_second": 89.807, | |
| "eval_spearman": 0.5903544136329805, | |
| "eval_steps_per_second": 11.226, | |
| "step": 40000 | |
| }, | |
| { | |
| "epoch": 20.5, | |
| "learning_rate": 7.2000000000000005e-06, | |
| "loss": 0.2143, | |
| "step": 41000 | |
| }, | |
| { | |
| "epoch": 21.0, | |
| "learning_rate": 6.4000000000000006e-06, | |
| "loss": 0.2104, | |
| "step": 42000 | |
| }, | |
| { | |
| "epoch": 21.5, | |
| "learning_rate": 5.600000000000001e-06, | |
| "loss": 0.2067, | |
| "step": 43000 | |
| }, | |
| { | |
| "epoch": 22.0, | |
| "learning_rate": 4.800000000000001e-06, | |
| "loss": 0.213, | |
| "step": 44000 | |
| }, | |
| { | |
| "epoch": 22.5, | |
| "learning_rate": 4.000000000000001e-06, | |
| "loss": 0.2098, | |
| "step": 45000 | |
| }, | |
| { | |
| "epoch": 23.0, | |
| "learning_rate": 3.2000000000000003e-06, | |
| "loss": 0.2106, | |
| "step": 46000 | |
| }, | |
| { | |
| "epoch": 23.5, | |
| "learning_rate": 2.4000000000000003e-06, | |
| "loss": 0.2071, | |
| "step": 47000 | |
| }, | |
| { | |
| "epoch": 24.0, | |
| "learning_rate": 1.6000000000000001e-06, | |
| "loss": 0.2095, | |
| "step": 48000 | |
| }, | |
| { | |
| "epoch": 24.5, | |
| "learning_rate": 8.000000000000001e-07, | |
| "loss": 0.2053, | |
| "step": 49000 | |
| }, | |
| { | |
| "epoch": 25.0, | |
| "learning_rate": 0.0, | |
| "loss": 0.2119, | |
| "step": 50000 | |
| }, | |
| { | |
| "epoch": 25.0, | |
| "eval_loss": 0.19228921830654144, | |
| "eval_p-value": 0.0, | |
| "eval_runtime": 44.4561, | |
| "eval_samples_per_second": 89.976, | |
| "eval_spearman": 0.5989616200729698, | |
| "eval_steps_per_second": 11.247, | |
| "step": 50000 | |
| } | |
| ], | |
| "logging_steps": 1000, | |
| "max_steps": 50000, | |
| "num_input_tokens_seen": 0, | |
| "num_train_epochs": 25, | |
| "save_steps": 10000, | |
| "total_flos": 3.71702366208e+16, | |
| "train_batch_size": 8, | |
| "trial_name": null, | |
| "trial_params": null | |
| } | |