| { | |
| "best_metric": null, | |
| "best_model_checkpoint": null, | |
| "epoch": 0.42182968626417083, | |
| "eval_steps": 50, | |
| "global_step": 400, | |
| "is_hyper_param_search": false, | |
| "is_local_process_zero": true, | |
| "is_world_process_zero": true, | |
| "log_history": [ | |
| { | |
| "epoch": 0.010545742156604272, | |
| "grad_norm": 0.8949686288833618, | |
| "learning_rate": 2.0000000000000003e-06, | |
| "loss": 2.0803, | |
| "step": 10 | |
| }, | |
| { | |
| "epoch": 0.021091484313208543, | |
| "grad_norm": 0.8410115838050842, | |
| "learning_rate": 4.000000000000001e-06, | |
| "loss": 2.0584, | |
| "step": 20 | |
| }, | |
| { | |
| "epoch": 0.03163722646981281, | |
| "grad_norm": 0.8865960836410522, | |
| "learning_rate": 6e-06, | |
| "loss": 2.0496, | |
| "step": 30 | |
| }, | |
| { | |
| "epoch": 0.042182968626417086, | |
| "grad_norm": 0.7669851779937744, | |
| "learning_rate": 8.000000000000001e-06, | |
| "loss": 2.0115, | |
| "step": 40 | |
| }, | |
| { | |
| "epoch": 0.052728710783021354, | |
| "grad_norm": 0.7927244901657104, | |
| "learning_rate": 1e-05, | |
| "loss": 1.999, | |
| "step": 50 | |
| }, | |
| { | |
| "epoch": 0.052728710783021354, | |
| "eval_loss": 1.9660409688949585, | |
| "eval_runtime": 1845.8031, | |
| "eval_samples_per_second": 1.827, | |
| "eval_steps_per_second": 0.457, | |
| "step": 50 | |
| }, | |
| { | |
| "epoch": 0.06327445293962562, | |
| "grad_norm": 0.6458373069763184, | |
| "learning_rate": 1.2e-05, | |
| "loss": 1.9319, | |
| "step": 60 | |
| }, | |
| { | |
| "epoch": 0.07382019509622989, | |
| "grad_norm": 0.566554069519043, | |
| "learning_rate": 1.4000000000000001e-05, | |
| "loss": 1.9084, | |
| "step": 70 | |
| }, | |
| { | |
| "epoch": 0.08436593725283417, | |
| "grad_norm": 0.4990901052951813, | |
| "learning_rate": 1.6000000000000003e-05, | |
| "loss": 1.8971, | |
| "step": 80 | |
| }, | |
| { | |
| "epoch": 0.09491167940943844, | |
| "grad_norm": 0.4458348751068115, | |
| "learning_rate": 1.8e-05, | |
| "loss": 1.8216, | |
| "step": 90 | |
| }, | |
| { | |
| "epoch": 0.10545742156604271, | |
| "grad_norm": 0.4464481472969055, | |
| "learning_rate": 2e-05, | |
| "loss": 1.8433, | |
| "step": 100 | |
| }, | |
| { | |
| "epoch": 0.10545742156604271, | |
| "eval_loss": 1.8134711980819702, | |
| "eval_runtime": 1846.165, | |
| "eval_samples_per_second": 1.826, | |
| "eval_steps_per_second": 0.457, | |
| "step": 100 | |
| }, | |
| { | |
| "epoch": 0.11600316372264698, | |
| "grad_norm": 0.4543227255344391, | |
| "learning_rate": 2.2000000000000003e-05, | |
| "loss": 1.8193, | |
| "step": 110 | |
| }, | |
| { | |
| "epoch": 0.12654890587925124, | |
| "grad_norm": 0.4522910416126251, | |
| "learning_rate": 2.4e-05, | |
| "loss": 1.7737, | |
| "step": 120 | |
| }, | |
| { | |
| "epoch": 0.13709464803585553, | |
| "grad_norm": 0.4522746801376343, | |
| "learning_rate": 2.6000000000000002e-05, | |
| "loss": 1.7885, | |
| "step": 130 | |
| }, | |
| { | |
| "epoch": 0.14764039019245978, | |
| "grad_norm": 0.44632086157798767, | |
| "learning_rate": 2.8000000000000003e-05, | |
| "loss": 1.7669, | |
| "step": 140 | |
| }, | |
| { | |
| "epoch": 0.15818613234906406, | |
| "grad_norm": 0.4553607404232025, | |
| "learning_rate": 3e-05, | |
| "loss": 1.7605, | |
| "step": 150 | |
| }, | |
| { | |
| "epoch": 0.15818613234906406, | |
| "eval_loss": 1.7562175989151, | |
| "eval_runtime": 1836.9664, | |
| "eval_samples_per_second": 1.836, | |
| "eval_steps_per_second": 0.459, | |
| "step": 150 | |
| }, | |
| { | |
| "epoch": 0.16873187450566834, | |
| "grad_norm": 0.4852316975593567, | |
| "learning_rate": 3.2000000000000005e-05, | |
| "loss": 1.7367, | |
| "step": 160 | |
| }, | |
| { | |
| "epoch": 0.1792776166622726, | |
| "grad_norm": 0.49629154801368713, | |
| "learning_rate": 3.4000000000000007e-05, | |
| "loss": 1.7447, | |
| "step": 170 | |
| }, | |
| { | |
| "epoch": 0.18982335881887688, | |
| "grad_norm": 0.5253108739852905, | |
| "learning_rate": 3.6e-05, | |
| "loss": 1.7378, | |
| "step": 180 | |
| }, | |
| { | |
| "epoch": 0.20036910097548116, | |
| "grad_norm": 0.6104539036750793, | |
| "learning_rate": 3.8e-05, | |
| "loss": 1.737, | |
| "step": 190 | |
| }, | |
| { | |
| "epoch": 0.21091484313208542, | |
| "grad_norm": 0.564102828502655, | |
| "learning_rate": 4e-05, | |
| "loss": 1.7325, | |
| "step": 200 | |
| }, | |
| { | |
| "epoch": 0.21091484313208542, | |
| "eval_loss": 1.727364420890808, | |
| "eval_runtime": 1836.4408, | |
| "eval_samples_per_second": 1.836, | |
| "eval_steps_per_second": 0.459, | |
| "step": 200 | |
| }, | |
| { | |
| "epoch": 0.2214605852886897, | |
| "grad_norm": 0.6158842444419861, | |
| "learning_rate": 4.2e-05, | |
| "loss": 1.7168, | |
| "step": 210 | |
| }, | |
| { | |
| "epoch": 0.23200632744529395, | |
| "grad_norm": 0.699800431728363, | |
| "learning_rate": 4.4000000000000006e-05, | |
| "loss": 1.738, | |
| "step": 220 | |
| }, | |
| { | |
| "epoch": 0.24255206960189823, | |
| "grad_norm": 0.6008381843566895, | |
| "learning_rate": 4.600000000000001e-05, | |
| "loss": 1.7256, | |
| "step": 230 | |
| }, | |
| { | |
| "epoch": 0.2530978117585025, | |
| "grad_norm": 0.633844792842865, | |
| "learning_rate": 4.8e-05, | |
| "loss": 1.7033, | |
| "step": 240 | |
| }, | |
| { | |
| "epoch": 0.2636435539151068, | |
| "grad_norm": 0.6631755232810974, | |
| "learning_rate": 5e-05, | |
| "loss": 1.6999, | |
| "step": 250 | |
| }, | |
| { | |
| "epoch": 0.2636435539151068, | |
| "eval_loss": 1.7032877206802368, | |
| "eval_runtime": 1836.0737, | |
| "eval_samples_per_second": 1.837, | |
| "eval_steps_per_second": 0.459, | |
| "step": 250 | |
| }, | |
| { | |
| "epoch": 0.27418929607171105, | |
| "grad_norm": 0.6652688980102539, | |
| "learning_rate": 5.2000000000000004e-05, | |
| "loss": 1.6866, | |
| "step": 260 | |
| }, | |
| { | |
| "epoch": 0.2847350382283153, | |
| "grad_norm": 0.6938503980636597, | |
| "learning_rate": 5.4000000000000005e-05, | |
| "loss": 1.7029, | |
| "step": 270 | |
| }, | |
| { | |
| "epoch": 0.29528078038491956, | |
| "grad_norm": 0.686392605304718, | |
| "learning_rate": 5.6000000000000006e-05, | |
| "loss": 1.6989, | |
| "step": 280 | |
| }, | |
| { | |
| "epoch": 0.30582652254152387, | |
| "grad_norm": 0.7344717979431152, | |
| "learning_rate": 5.8e-05, | |
| "loss": 1.6685, | |
| "step": 290 | |
| }, | |
| { | |
| "epoch": 0.3163722646981281, | |
| "grad_norm": 0.6960188150405884, | |
| "learning_rate": 6e-05, | |
| "loss": 1.6705, | |
| "step": 300 | |
| }, | |
| { | |
| "epoch": 0.3163722646981281, | |
| "eval_loss": 1.6844133138656616, | |
| "eval_runtime": 1836.0829, | |
| "eval_samples_per_second": 1.837, | |
| "eval_steps_per_second": 0.459, | |
| "step": 300 | |
| }, | |
| { | |
| "epoch": 0.3269180068547324, | |
| "grad_norm": 0.6903261542320251, | |
| "learning_rate": 6.2e-05, | |
| "loss": 1.6672, | |
| "step": 310 | |
| }, | |
| { | |
| "epoch": 0.3374637490113367, | |
| "grad_norm": 0.7028161883354187, | |
| "learning_rate": 6.400000000000001e-05, | |
| "loss": 1.6934, | |
| "step": 320 | |
| }, | |
| { | |
| "epoch": 0.34800949116794094, | |
| "grad_norm": 0.7435672879219055, | |
| "learning_rate": 6.6e-05, | |
| "loss": 1.6712, | |
| "step": 330 | |
| }, | |
| { | |
| "epoch": 0.3585552333245452, | |
| "grad_norm": 0.7246424555778503, | |
| "learning_rate": 6.800000000000001e-05, | |
| "loss": 1.6608, | |
| "step": 340 | |
| }, | |
| { | |
| "epoch": 0.3691009754811495, | |
| "grad_norm": 0.6919660568237305, | |
| "learning_rate": 7e-05, | |
| "loss": 1.6568, | |
| "step": 350 | |
| }, | |
| { | |
| "epoch": 0.3691009754811495, | |
| "eval_loss": 1.6683924198150635, | |
| "eval_runtime": 1838.3493, | |
| "eval_samples_per_second": 1.834, | |
| "eval_steps_per_second": 0.459, | |
| "step": 350 | |
| }, | |
| { | |
| "epoch": 0.37964671763775376, | |
| "grad_norm": 0.7590942978858948, | |
| "learning_rate": 7.2e-05, | |
| "loss": 1.6742, | |
| "step": 360 | |
| }, | |
| { | |
| "epoch": 0.390192459794358, | |
| "grad_norm": 0.6980053186416626, | |
| "learning_rate": 7.4e-05, | |
| "loss": 1.6744, | |
| "step": 370 | |
| }, | |
| { | |
| "epoch": 0.4007382019509623, | |
| "grad_norm": 0.7112457752227783, | |
| "learning_rate": 7.6e-05, | |
| "loss": 1.648, | |
| "step": 380 | |
| }, | |
| { | |
| "epoch": 0.4112839441075666, | |
| "grad_norm": 0.7157771587371826, | |
| "learning_rate": 7.800000000000001e-05, | |
| "loss": 1.6634, | |
| "step": 390 | |
| }, | |
| { | |
| "epoch": 0.42182968626417083, | |
| "grad_norm": 0.7440850138664246, | |
| "learning_rate": 8e-05, | |
| "loss": 1.6567, | |
| "step": 400 | |
| }, | |
| { | |
| "epoch": 0.42182968626417083, | |
| "eval_loss": 1.6561506986618042, | |
| "eval_runtime": 1822.2187, | |
| "eval_samples_per_second": 1.85, | |
| "eval_steps_per_second": 0.463, | |
| "step": 400 | |
| } | |
| ], | |
| "logging_steps": 10, | |
| "max_steps": 2844, | |
| "num_input_tokens_seen": 0, | |
| "num_train_epochs": 3, | |
| "save_steps": 50, | |
| "stateful_callbacks": { | |
| "TrainerControl": { | |
| "args": { | |
| "should_epoch_stop": false, | |
| "should_evaluate": false, | |
| "should_log": false, | |
| "should_save": true, | |
| "should_training_stop": false | |
| }, | |
| "attributes": {} | |
| } | |
| }, | |
| "total_flos": 3.274968196448256e+17, | |
| "train_batch_size": 4, | |
| "trial_name": null, | |
| "trial_params": null | |
| } | |