| { | |
| "best_metric": null, | |
| "best_model_checkpoint": null, | |
| "epoch": 1.029966703662597, | |
| "eval_steps": 8, | |
| "global_step": 29, | |
| "is_hyper_param_search": false, | |
| "is_local_process_zero": true, | |
| "is_world_process_zero": true, | |
| "log_history": [ | |
| { | |
| "epoch": 0.03551609322974473, | |
| "grad_norm": 0.3238994777202606, | |
| "learning_rate": 5e-05, | |
| "loss": 1.1629, | |
| "step": 1 | |
| }, | |
| { | |
| "epoch": 0.03551609322974473, | |
| "eval_loss": 1.148727536201477, | |
| "eval_runtime": 17.5824, | |
| "eval_samples_per_second": 10.806, | |
| "eval_steps_per_second": 2.73, | |
| "step": 1 | |
| }, | |
| { | |
| "epoch": 0.07103218645948946, | |
| "grad_norm": 0.30336055159568787, | |
| "learning_rate": 0.0001, | |
| "loss": 1.08, | |
| "step": 2 | |
| }, | |
| { | |
| "epoch": 0.10654827968923418, | |
| "grad_norm": 0.32345050573349, | |
| "learning_rate": 9.966191788709716e-05, | |
| "loss": 1.1277, | |
| "step": 3 | |
| }, | |
| { | |
| "epoch": 0.14206437291897892, | |
| "grad_norm": 0.2949381172657013, | |
| "learning_rate": 9.865224352899119e-05, | |
| "loss": 1.0632, | |
| "step": 4 | |
| }, | |
| { | |
| "epoch": 0.17758046614872364, | |
| "grad_norm": 0.28073155879974365, | |
| "learning_rate": 9.698463103929542e-05, | |
| "loss": 1.0568, | |
| "step": 5 | |
| }, | |
| { | |
| "epoch": 0.21309655937846836, | |
| "grad_norm": 0.3517192304134369, | |
| "learning_rate": 9.468163201617062e-05, | |
| "loss": 1.0366, | |
| "step": 6 | |
| }, | |
| { | |
| "epoch": 0.2486126526082131, | |
| "grad_norm": 0.37389102578163147, | |
| "learning_rate": 9.177439057064683e-05, | |
| "loss": 0.9688, | |
| "step": 7 | |
| }, | |
| { | |
| "epoch": 0.28412874583795783, | |
| "grad_norm": 0.3350629508495331, | |
| "learning_rate": 8.83022221559489e-05, | |
| "loss": 1.0059, | |
| "step": 8 | |
| }, | |
| { | |
| "epoch": 0.28412874583795783, | |
| "eval_loss": 0.9954966306686401, | |
| "eval_runtime": 17.632, | |
| "eval_samples_per_second": 10.776, | |
| "eval_steps_per_second": 2.722, | |
| "step": 8 | |
| }, | |
| { | |
| "epoch": 0.3196448390677026, | |
| "grad_norm": 0.2728235721588135, | |
| "learning_rate": 8.43120818934367e-05, | |
| "loss": 0.9818, | |
| "step": 9 | |
| }, | |
| { | |
| "epoch": 0.3551609322974473, | |
| "grad_norm": 0.2388256937265396, | |
| "learning_rate": 7.985792958513931e-05, | |
| "loss": 0.9472, | |
| "step": 10 | |
| }, | |
| { | |
| "epoch": 0.390677025527192, | |
| "grad_norm": 0.23059608042240143, | |
| "learning_rate": 7.500000000000001e-05, | |
| "loss": 0.9187, | |
| "step": 11 | |
| }, | |
| { | |
| "epoch": 0.4261931187569367, | |
| "grad_norm": 0.24368488788604736, | |
| "learning_rate": 6.980398830195785e-05, | |
| "loss": 0.9736, | |
| "step": 12 | |
| }, | |
| { | |
| "epoch": 0.46170921198668147, | |
| "grad_norm": 0.24262012541294098, | |
| "learning_rate": 6.434016163555452e-05, | |
| "loss": 0.9481, | |
| "step": 13 | |
| }, | |
| { | |
| "epoch": 0.4972253052164262, | |
| "grad_norm": 0.24401485919952393, | |
| "learning_rate": 5.868240888334653e-05, | |
| "loss": 0.9301, | |
| "step": 14 | |
| }, | |
| { | |
| "epoch": 0.532741398446171, | |
| "grad_norm": 0.22144131362438202, | |
| "learning_rate": 5.290724144552379e-05, | |
| "loss": 0.886, | |
| "step": 15 | |
| }, | |
| { | |
| "epoch": 0.5682574916759157, | |
| "grad_norm": 0.2215518355369568, | |
| "learning_rate": 4.709275855447621e-05, | |
| "loss": 0.9087, | |
| "step": 16 | |
| }, | |
| { | |
| "epoch": 0.5682574916759157, | |
| "eval_loss": 0.933112382888794, | |
| "eval_runtime": 17.6126, | |
| "eval_samples_per_second": 10.788, | |
| "eval_steps_per_second": 2.725, | |
| "step": 16 | |
| }, | |
| { | |
| "epoch": 0.6037735849056604, | |
| "grad_norm": 0.23342889547348022, | |
| "learning_rate": 4.131759111665349e-05, | |
| "loss": 0.8903, | |
| "step": 17 | |
| }, | |
| { | |
| "epoch": 0.6392896781354052, | |
| "grad_norm": 0.21479082107543945, | |
| "learning_rate": 3.5659838364445505e-05, | |
| "loss": 0.9049, | |
| "step": 18 | |
| }, | |
| { | |
| "epoch": 0.6748057713651499, | |
| "grad_norm": 0.2083897590637207, | |
| "learning_rate": 3.019601169804216e-05, | |
| "loss": 0.9096, | |
| "step": 19 | |
| }, | |
| { | |
| "epoch": 0.7103218645948945, | |
| "grad_norm": 0.2214074432849884, | |
| "learning_rate": 2.500000000000001e-05, | |
| "loss": 0.9017, | |
| "step": 20 | |
| }, | |
| { | |
| "epoch": 0.7458379578246392, | |
| "grad_norm": 0.21316947042942047, | |
| "learning_rate": 2.0142070414860704e-05, | |
| "loss": 0.8555, | |
| "step": 21 | |
| }, | |
| { | |
| "epoch": 0.781354051054384, | |
| "grad_norm": 0.19444766640663147, | |
| "learning_rate": 1.5687918106563326e-05, | |
| "loss": 0.8838, | |
| "step": 22 | |
| }, | |
| { | |
| "epoch": 0.8168701442841287, | |
| "grad_norm": 0.24460318684577942, | |
| "learning_rate": 1.1697777844051105e-05, | |
| "loss": 0.9097, | |
| "step": 23 | |
| }, | |
| { | |
| "epoch": 0.8523862375138734, | |
| "grad_norm": 0.20731952786445618, | |
| "learning_rate": 8.225609429353187e-06, | |
| "loss": 0.8475, | |
| "step": 24 | |
| }, | |
| { | |
| "epoch": 0.8523862375138734, | |
| "eval_loss": 0.9121983647346497, | |
| "eval_runtime": 17.6276, | |
| "eval_samples_per_second": 10.779, | |
| "eval_steps_per_second": 2.723, | |
| "step": 24 | |
| }, | |
| { | |
| "epoch": 0.8879023307436182, | |
| "grad_norm": 0.209746316075325, | |
| "learning_rate": 5.318367983829392e-06, | |
| "loss": 0.8696, | |
| "step": 25 | |
| }, | |
| { | |
| "epoch": 0.9234184239733629, | |
| "grad_norm": 0.21727362275123596, | |
| "learning_rate": 3.0153689607045845e-06, | |
| "loss": 0.886, | |
| "step": 26 | |
| }, | |
| { | |
| "epoch": 0.9589345172031076, | |
| "grad_norm": 0.21419359743595123, | |
| "learning_rate": 1.3477564710088098e-06, | |
| "loss": 0.8859, | |
| "step": 27 | |
| }, | |
| { | |
| "epoch": 0.9944506104328524, | |
| "grad_norm": 0.22184455394744873, | |
| "learning_rate": 3.380821129028489e-07, | |
| "loss": 0.8763, | |
| "step": 28 | |
| }, | |
| { | |
| "epoch": 1.029966703662597, | |
| "grad_norm": 0.48768743872642517, | |
| "learning_rate": 0.0, | |
| "loss": 1.5605, | |
| "step": 29 | |
| } | |
| ], | |
| "logging_steps": 1, | |
| "max_steps": 29, | |
| "num_input_tokens_seen": 0, | |
| "num_train_epochs": 2, | |
| "save_steps": 8, | |
| "stateful_callbacks": { | |
| "TrainerControl": { | |
| "args": { | |
| "should_epoch_stop": false, | |
| "should_evaluate": false, | |
| "should_log": false, | |
| "should_save": true, | |
| "should_training_stop": true | |
| }, | |
| "attributes": {} | |
| } | |
| }, | |
| "total_flos": 6.884713991440957e+17, | |
| "train_batch_size": 2, | |
| "trial_name": null, | |
| "trial_params": null | |
| } | |