| { | |
| "best_metric": 0.8248077630996704, | |
| "best_model_checkpoint": "./out/checkpoint-10", | |
| "epoch": 3.217391304347826, | |
| "eval_steps": 2, | |
| "global_step": 20, | |
| "is_hyper_param_search": false, | |
| "is_local_process_zero": true, | |
| "is_world_process_zero": true, | |
| "log_history": [ | |
| { | |
| "epoch": 0.17, | |
| "learning_rate": 5.000000000000001e-07, | |
| "loss": 0.9984, | |
| "step": 1 | |
| }, | |
| { | |
| "epoch": 0.17, | |
| "eval_loss": 1.09337317943573, | |
| "eval_runtime": 65.1601, | |
| "eval_samples_per_second": 1.535, | |
| "eval_steps_per_second": 0.767, | |
| "step": 1 | |
| }, | |
| { | |
| "epoch": 0.35, | |
| "learning_rate": 1.0000000000000002e-06, | |
| "loss": 0.9864, | |
| "step": 2 | |
| }, | |
| { | |
| "epoch": 0.35, | |
| "eval_loss": 1.0603257417678833, | |
| "eval_runtime": 66.8005, | |
| "eval_samples_per_second": 1.497, | |
| "eval_steps_per_second": 0.748, | |
| "step": 2 | |
| }, | |
| { | |
| "epoch": 0.52, | |
| "learning_rate": 1.5e-06, | |
| "loss": 0.9692, | |
| "step": 3 | |
| }, | |
| { | |
| "epoch": 0.7, | |
| "learning_rate": 2.0000000000000003e-06, | |
| "loss": 0.9181, | |
| "step": 4 | |
| }, | |
| { | |
| "epoch": 0.7, | |
| "eval_loss": 0.9132175445556641, | |
| "eval_runtime": 65.6994, | |
| "eval_samples_per_second": 1.522, | |
| "eval_steps_per_second": 0.761, | |
| "step": 4 | |
| }, | |
| { | |
| "epoch": 0.87, | |
| "learning_rate": 2.5e-06, | |
| "loss": 0.9237, | |
| "step": 5 | |
| }, | |
| { | |
| "epoch": 1.04, | |
| "learning_rate": 3e-06, | |
| "loss": 0.8843, | |
| "step": 6 | |
| }, | |
| { | |
| "epoch": 1.04, | |
| "eval_loss": 0.8623057007789612, | |
| "eval_runtime": 64.7101, | |
| "eval_samples_per_second": 1.545, | |
| "eval_steps_per_second": 0.773, | |
| "step": 6 | |
| }, | |
| { | |
| "epoch": 1.13, | |
| "learning_rate": 3.5e-06, | |
| "loss": 0.8434, | |
| "step": 7 | |
| }, | |
| { | |
| "epoch": 1.3, | |
| "learning_rate": 4.000000000000001e-06, | |
| "loss": 0.8513, | |
| "step": 8 | |
| }, | |
| { | |
| "epoch": 1.3, | |
| "eval_loss": 0.8309622406959534, | |
| "eval_runtime": 64.996, | |
| "eval_samples_per_second": 1.539, | |
| "eval_steps_per_second": 0.769, | |
| "step": 8 | |
| }, | |
| { | |
| "epoch": 1.48, | |
| "learning_rate": 4.5e-06, | |
| "loss": 0.8174, | |
| "step": 9 | |
| }, | |
| { | |
| "epoch": 1.65, | |
| "learning_rate": 5e-06, | |
| "loss": 0.7957, | |
| "step": 10 | |
| }, | |
| { | |
| "epoch": 1.65, | |
| "eval_loss": 0.8248077630996704, | |
| "eval_runtime": 65.0362, | |
| "eval_samples_per_second": 1.538, | |
| "eval_steps_per_second": 0.769, | |
| "step": 10 | |
| }, | |
| { | |
| "epoch": 1.83, | |
| "learning_rate": 4.8776412907378845e-06, | |
| "loss": 0.7513, | |
| "step": 11 | |
| }, | |
| { | |
| "epoch": 2.0, | |
| "learning_rate": 4.522542485937369e-06, | |
| "loss": 0.7823, | |
| "step": 12 | |
| }, | |
| { | |
| "epoch": 2.0, | |
| "eval_loss": 0.822129487991333, | |
| "eval_runtime": 64.6722, | |
| "eval_samples_per_second": 1.546, | |
| "eval_steps_per_second": 0.773, | |
| "step": 12 | |
| }, | |
| { | |
| "epoch": 2.09, | |
| "learning_rate": 3.969463130731183e-06, | |
| "loss": 0.681, | |
| "step": 13 | |
| }, | |
| { | |
| "epoch": 2.26, | |
| "learning_rate": 3.272542485937369e-06, | |
| "loss": 0.5977, | |
| "step": 14 | |
| }, | |
| { | |
| "epoch": 2.26, | |
| "eval_loss": 0.845899224281311, | |
| "eval_runtime": 64.9298, | |
| "eval_samples_per_second": 1.54, | |
| "eval_steps_per_second": 0.77, | |
| "step": 14 | |
| }, | |
| { | |
| "epoch": 2.43, | |
| "learning_rate": 2.5e-06, | |
| "loss": 0.5913, | |
| "step": 15 | |
| }, | |
| { | |
| "epoch": 2.61, | |
| "learning_rate": 1.7274575140626318e-06, | |
| "loss": 0.5766, | |
| "step": 16 | |
| }, | |
| { | |
| "epoch": 2.61, | |
| "eval_loss": 0.8647727966308594, | |
| "eval_runtime": 64.5733, | |
| "eval_samples_per_second": 1.549, | |
| "eval_steps_per_second": 0.774, | |
| "step": 16 | |
| }, | |
| { | |
| "epoch": 2.78, | |
| "learning_rate": 1.0305368692688175e-06, | |
| "loss": 0.5096, | |
| "step": 17 | |
| }, | |
| { | |
| "epoch": 2.96, | |
| "learning_rate": 4.774575140626317e-07, | |
| "loss": 0.546, | |
| "step": 18 | |
| }, | |
| { | |
| "epoch": 2.96, | |
| "eval_loss": 0.863722026348114, | |
| "eval_runtime": 64.9313, | |
| "eval_samples_per_second": 1.54, | |
| "eval_steps_per_second": 0.77, | |
| "step": 18 | |
| }, | |
| { | |
| "epoch": 3.04, | |
| "learning_rate": 1.223587092621162e-07, | |
| "loss": 0.524, | |
| "step": 19 | |
| }, | |
| { | |
| "epoch": 3.22, | |
| "learning_rate": 0.0, | |
| "loss": 0.5024, | |
| "step": 20 | |
| }, | |
| { | |
| "epoch": 3.22, | |
| "eval_loss": 0.863263726234436, | |
| "eval_runtime": 65.767, | |
| "eval_samples_per_second": 1.521, | |
| "eval_steps_per_second": 0.76, | |
| "step": 20 | |
| } | |
| ], | |
| "logging_steps": 1, | |
| "max_steps": 20, | |
| "num_input_tokens_seen": 0, | |
| "num_train_epochs": 4, | |
| "save_steps": 5, | |
| "total_flos": 5.592050640617472e+16, | |
| "train_batch_size": 2, | |
| "trial_name": null, | |
| "trial_params": null | |
| } | |