| { | |
| "best_metric": null, | |
| "best_model_checkpoint": null, | |
| "epoch": 2.6859016393442623, | |
| "global_step": 49152, | |
| "is_hyper_param_search": false, | |
| "is_local_process_zero": true, | |
| "is_world_process_zero": true, | |
| "log_history": [ | |
| { | |
| "epoch": 0.22, | |
| "learning_rate": 0.0004984426897459585, | |
| "loss": 0.5454, | |
| "step": 4096 | |
| }, | |
| { | |
| "epoch": 0.22, | |
| "eval_loss": 0.5005695223808289, | |
| "eval_runtime": 32.9622, | |
| "eval_samples_per_second": 104.271, | |
| "eval_steps_per_second": 6.523, | |
| "step": 4096 | |
| }, | |
| { | |
| "epoch": 0.45, | |
| "learning_rate": 0.0004934687023955681, | |
| "loss": 0.4849, | |
| "step": 8192 | |
| }, | |
| { | |
| "epoch": 0.45, | |
| "eval_loss": 0.480915367603302, | |
| "eval_runtime": 32.9785, | |
| "eval_samples_per_second": 104.219, | |
| "eval_steps_per_second": 6.519, | |
| "step": 8192 | |
| }, | |
| { | |
| "epoch": 0.67, | |
| "learning_rate": 0.000485144849673373, | |
| "loss": 0.4518, | |
| "step": 12288 | |
| }, | |
| { | |
| "epoch": 0.67, | |
| "eval_loss": 0.46660953760147095, | |
| "eval_runtime": 32.8208, | |
| "eval_samples_per_second": 104.72, | |
| "eval_steps_per_second": 6.551, | |
| "step": 12288 | |
| }, | |
| { | |
| "epoch": 0.9, | |
| "learning_rate": 0.0004735848873631612, | |
| "loss": 0.4282, | |
| "step": 16384 | |
| }, | |
| { | |
| "epoch": 0.9, | |
| "eval_loss": 0.46097490191459656, | |
| "eval_runtime": 32.8311, | |
| "eval_samples_per_second": 104.687, | |
| "eval_steps_per_second": 6.549, | |
| "step": 16384 | |
| }, | |
| { | |
| "epoch": 1.12, | |
| "learning_rate": 0.0004589518403420676, | |
| "loss": 0.4145, | |
| "step": 20480 | |
| }, | |
| { | |
| "epoch": 1.12, | |
| "eval_loss": 0.45063599944114685, | |
| "eval_runtime": 32.8912, | |
| "eval_samples_per_second": 104.496, | |
| "eval_steps_per_second": 6.537, | |
| "step": 20480 | |
| }, | |
| { | |
| "epoch": 1.34, | |
| "learning_rate": 0.0004414445597486605, | |
| "loss": 0.399, | |
| "step": 24576 | |
| }, | |
| { | |
| "epoch": 1.34, | |
| "eval_loss": 0.44468095898628235, | |
| "eval_runtime": 32.706, | |
| "eval_samples_per_second": 105.088, | |
| "eval_steps_per_second": 6.574, | |
| "step": 24576 | |
| }, | |
| { | |
| "epoch": 1.57, | |
| "learning_rate": 0.00042130386669061293, | |
| "loss": 0.3882, | |
| "step": 28672 | |
| }, | |
| { | |
| "epoch": 1.57, | |
| "eval_loss": 0.44857361912727356, | |
| "eval_runtime": 32.9754, | |
| "eval_samples_per_second": 104.229, | |
| "eval_steps_per_second": 6.52, | |
| "step": 28672 | |
| }, | |
| { | |
| "epoch": 1.79, | |
| "learning_rate": 0.0003988010477498867, | |
| "loss": 0.3767, | |
| "step": 32768 | |
| }, | |
| { | |
| "epoch": 1.79, | |
| "eval_loss": 0.44354742765426636, | |
| "eval_runtime": 32.8219, | |
| "eval_samples_per_second": 104.717, | |
| "eval_steps_per_second": 6.55, | |
| "step": 32768 | |
| }, | |
| { | |
| "epoch": 2.01, | |
| "learning_rate": 0.0003742566178542921, | |
| "loss": 0.3676, | |
| "step": 36864 | |
| }, | |
| { | |
| "epoch": 2.01, | |
| "eval_loss": 0.43940743803977966, | |
| "eval_runtime": 32.9279, | |
| "eval_samples_per_second": 104.38, | |
| "eval_steps_per_second": 6.529, | |
| "step": 36864 | |
| }, | |
| { | |
| "epoch": 2.24, | |
| "learning_rate": 0.0003479963856008823, | |
| "loss": 0.3577, | |
| "step": 40960 | |
| }, | |
| { | |
| "epoch": 2.24, | |
| "eval_loss": 0.4323909878730774, | |
| "eval_runtime": 33.0718, | |
| "eval_samples_per_second": 103.926, | |
| "eval_steps_per_second": 6.501, | |
| "step": 40960 | |
| }, | |
| { | |
| "epoch": 2.46, | |
| "learning_rate": 0.0003203943839704654, | |
| "loss": 0.3517, | |
| "step": 45056 | |
| }, | |
| { | |
| "epoch": 2.46, | |
| "eval_loss": 0.43262797594070435, | |
| "eval_runtime": 32.8978, | |
| "eval_samples_per_second": 104.475, | |
| "eval_steps_per_second": 6.535, | |
| "step": 45056 | |
| }, | |
| { | |
| "epoch": 2.69, | |
| "learning_rate": 0.000291817008494138, | |
| "loss": 0.3427, | |
| "step": 49152 | |
| }, | |
| { | |
| "epoch": 2.69, | |
| "eval_loss": 0.4303751289844513, | |
| "eval_runtime": 32.7306, | |
| "eval_samples_per_second": 105.009, | |
| "eval_steps_per_second": 6.569, | |
| "step": 49152 | |
| } | |
| ], | |
| "max_steps": 109800, | |
| "num_train_epochs": 6, | |
| "total_flos": 2.28488651552981e+17, | |
| "trial_name": null, | |
| "trial_params": null | |
| } | |