| { | |
| "best_metric": null, | |
| "best_model_checkpoint": null, | |
| "epoch": 0.19063959584405682, | |
| "eval_steps": 200, | |
| "global_step": 2000, | |
| "is_hyper_param_search": false, | |
| "is_local_process_zero": true, | |
| "is_world_process_zero": true, | |
| "log_history": [ | |
| { | |
| "epoch": 0.019063959584405682, | |
| "eval_accuracy": 0.6766402862189922, | |
| "eval_loss": 1.5877882242202759, | |
| "eval_runtime": 1128.2057, | |
| "eval_samples_per_second": 15.517, | |
| "eval_steps_per_second": 0.486, | |
| "step": 200 | |
| }, | |
| { | |
| "epoch": 0.038127919168811364, | |
| "eval_accuracy": 0.7249642161946639, | |
| "eval_loss": 1.3184435367584229, | |
| "eval_runtime": 1127.8694, | |
| "eval_samples_per_second": 15.521, | |
| "eval_steps_per_second": 0.486, | |
| "step": 400 | |
| }, | |
| { | |
| "epoch": 0.047659898961014205, | |
| "grad_norm": 2.379204750061035, | |
| "learning_rate": 4.920566835064976e-05, | |
| "loss": 1.8513, | |
| "step": 500 | |
| }, | |
| { | |
| "epoch": 0.057191878753217046, | |
| "eval_accuracy": 0.7487254293909154, | |
| "eval_loss": 1.188010334968567, | |
| "eval_runtime": 1128.193, | |
| "eval_samples_per_second": 15.517, | |
| "eval_steps_per_second": 0.486, | |
| "step": 600 | |
| }, | |
| { | |
| "epoch": 0.07625583833762273, | |
| "eval_accuracy": 0.7630523662181098, | |
| "eval_loss": 1.1084955930709839, | |
| "eval_runtime": 1128.3934, | |
| "eval_samples_per_second": 15.514, | |
| "eval_steps_per_second": 0.486, | |
| "step": 800 | |
| }, | |
| { | |
| "epoch": 0.09531979792202841, | |
| "grad_norm": 2.218456268310547, | |
| "learning_rate": 4.841133670129953e-05, | |
| "loss": 1.1387, | |
| "step": 1000 | |
| }, | |
| { | |
| "epoch": 0.09531979792202841, | |
| "eval_accuracy": 0.7720419453759662, | |
| "eval_loss": 1.057796597480774, | |
| "eval_runtime": 1128.1566, | |
| "eval_samples_per_second": 15.517, | |
| "eval_steps_per_second": 0.486, | |
| "step": 1000 | |
| }, | |
| { | |
| "epoch": 0.11438375750643409, | |
| "eval_accuracy": 0.7792805618298705, | |
| "eval_loss": 1.0198323726654053, | |
| "eval_runtime": 1128.4409, | |
| "eval_samples_per_second": 15.513, | |
| "eval_steps_per_second": 0.486, | |
| "step": 1200 | |
| }, | |
| { | |
| "epoch": 0.13344771709083977, | |
| "eval_accuracy": 0.7836537788805386, | |
| "eval_loss": 0.9978336691856384, | |
| "eval_runtime": 1127.624, | |
| "eval_samples_per_second": 15.525, | |
| "eval_steps_per_second": 0.486, | |
| "step": 1400 | |
| }, | |
| { | |
| "epoch": 0.1429796968830426, | |
| "grad_norm": 2.3090548515319824, | |
| "learning_rate": 4.761700505194929e-05, | |
| "loss": 1.0151, | |
| "step": 1500 | |
| }, | |
| { | |
| "epoch": 0.15251167667524546, | |
| "eval_accuracy": 0.7889919141146797, | |
| "eval_loss": 0.9722086191177368, | |
| "eval_runtime": 1128.0241, | |
| "eval_samples_per_second": 15.519, | |
| "eval_steps_per_second": 0.486, | |
| "step": 1600 | |
| }, | |
| { | |
| "epoch": 0.17157563625965114, | |
| "eval_accuracy": 0.7922293997022891, | |
| "eval_loss": 0.9535930156707764, | |
| "eval_runtime": 1128.3297, | |
| "eval_samples_per_second": 15.515, | |
| "eval_steps_per_second": 0.486, | |
| "step": 1800 | |
| }, | |
| { | |
| "epoch": 0.19063959584405682, | |
| "grad_norm": 2.6724276542663574, | |
| "learning_rate": 4.682267340259906e-05, | |
| "loss": 0.9603, | |
| "step": 2000 | |
| }, | |
| { | |
| "epoch": 0.19063959584405682, | |
| "eval_accuracy": 0.7961234680845941, | |
| "eval_loss": 0.9353302121162415, | |
| "eval_runtime": 1128.8643, | |
| "eval_samples_per_second": 15.508, | |
| "eval_steps_per_second": 0.485, | |
| "step": 2000 | |
| } | |
| ], | |
| "logging_steps": 500, | |
| "max_steps": 31473, | |
| "num_input_tokens_seen": 0, | |
| "num_train_epochs": 3, | |
| "save_steps": 1000, | |
| "total_flos": 1.483774567120896e+18, | |
| "train_batch_size": 32, | |
| "trial_name": null, | |
| "trial_params": null | |
| } | |