| { | |
| "best_metric": null, | |
| "best_model_checkpoint": null, | |
| "epoch": 0.007985480943738656, | |
| "eval_steps": 5, | |
| "global_step": 33, | |
| "is_hyper_param_search": false, | |
| "is_local_process_zero": true, | |
| "is_world_process_zero": true, | |
| "log_history": [ | |
| { | |
| "epoch": 0.00024198427102238354, | |
| "eval_loss": 1.8433265686035156, | |
| "eval_runtime": 111.1181, | |
| "eval_samples_per_second": 15.659, | |
| "eval_steps_per_second": 7.83, | |
| "step": 1 | |
| }, | |
| { | |
| "epoch": 0.0007259528130671506, | |
| "grad_norm": 2.1982052326202393, | |
| "learning_rate": 3.529411764705883e-05, | |
| "loss": 7.0831, | |
| "step": 3 | |
| }, | |
| { | |
| "epoch": 0.0012099213551119178, | |
| "eval_loss": 1.830064058303833, | |
| "eval_runtime": 112.2893, | |
| "eval_samples_per_second": 15.496, | |
| "eval_steps_per_second": 7.748, | |
| "step": 5 | |
| }, | |
| { | |
| "epoch": 0.0014519056261343012, | |
| "grad_norm": 2.6449875831604004, | |
| "learning_rate": 7.058823529411765e-05, | |
| "loss": 6.9298, | |
| "step": 6 | |
| }, | |
| { | |
| "epoch": 0.002177858439201452, | |
| "grad_norm": 3.1286723613739014, | |
| "learning_rate": 0.00010588235294117647, | |
| "loss": 7.3702, | |
| "step": 9 | |
| }, | |
| { | |
| "epoch": 0.0024198427102238356, | |
| "eval_loss": 1.7329907417297363, | |
| "eval_runtime": 112.2576, | |
| "eval_samples_per_second": 15.5, | |
| "eval_steps_per_second": 7.75, | |
| "step": 10 | |
| }, | |
| { | |
| "epoch": 0.0029038112522686023, | |
| "grad_norm": 3.319622755050659, | |
| "learning_rate": 0.0001411764705882353, | |
| "loss": 6.2615, | |
| "step": 12 | |
| }, | |
| { | |
| "epoch": 0.003629764065335753, | |
| "grad_norm": 3.7298972606658936, | |
| "learning_rate": 0.00017647058823529413, | |
| "loss": 7.1705, | |
| "step": 15 | |
| }, | |
| { | |
| "epoch": 0.003629764065335753, | |
| "eval_loss": 1.6584177017211914, | |
| "eval_runtime": 112.3477, | |
| "eval_samples_per_second": 15.488, | |
| "eval_steps_per_second": 7.744, | |
| "step": 15 | |
| }, | |
| { | |
| "epoch": 0.004355716878402904, | |
| "grad_norm": 3.502082586288452, | |
| "learning_rate": 0.00019807852804032305, | |
| "loss": 6.9885, | |
| "step": 18 | |
| }, | |
| { | |
| "epoch": 0.004839685420447671, | |
| "eval_loss": 1.6045597791671753, | |
| "eval_runtime": 112.3898, | |
| "eval_samples_per_second": 15.482, | |
| "eval_steps_per_second": 7.741, | |
| "step": 20 | |
| }, | |
| { | |
| "epoch": 0.005081669691470054, | |
| "grad_norm": 4.299140930175781, | |
| "learning_rate": 0.00017071067811865476, | |
| "loss": 6.4409, | |
| "step": 21 | |
| }, | |
| { | |
| "epoch": 0.005807622504537205, | |
| "grad_norm": 3.838430643081665, | |
| "learning_rate": 0.00011950903220161285, | |
| "loss": 5.5058, | |
| "step": 24 | |
| }, | |
| { | |
| "epoch": 0.006049606775559589, | |
| "eval_loss": 1.5623209476470947, | |
| "eval_runtime": 112.5246, | |
| "eval_samples_per_second": 15.463, | |
| "eval_steps_per_second": 7.732, | |
| "step": 25 | |
| }, | |
| { | |
| "epoch": 0.006533575317604356, | |
| "grad_norm": 3.390537977218628, | |
| "learning_rate": 6.173165676349103e-05, | |
| "loss": 6.1822, | |
| "step": 27 | |
| }, | |
| { | |
| "epoch": 0.007259528130671506, | |
| "grad_norm": 3.009563684463501, | |
| "learning_rate": 1.6853038769745467e-05, | |
| "loss": 5.9296, | |
| "step": 30 | |
| }, | |
| { | |
| "epoch": 0.007259528130671506, | |
| "eval_loss": 1.550625205039978, | |
| "eval_runtime": 112.422, | |
| "eval_samples_per_second": 15.477, | |
| "eval_steps_per_second": 7.739, | |
| "step": 30 | |
| }, | |
| { | |
| "epoch": 0.007985480943738656, | |
| "grad_norm": 3.4622366428375244, | |
| "learning_rate": 0.0, | |
| "loss": 6.3164, | |
| "step": 33 | |
| } | |
| ], | |
| "logging_steps": 3, | |
| "max_steps": 33, | |
| "num_input_tokens_seen": 0, | |
| "num_train_epochs": 1, | |
| "save_steps": 17, | |
| "stateful_callbacks": { | |
| "TrainerControl": { | |
| "args": { | |
| "should_epoch_stop": false, | |
| "should_evaluate": false, | |
| "should_log": false, | |
| "should_save": true, | |
| "should_training_stop": true | |
| }, | |
| "attributes": {} | |
| } | |
| }, | |
| "total_flos": 2115835829157888.0, | |
| "train_batch_size": 2, | |
| "trial_name": null, | |
| "trial_params": null | |
| } | |