| { | |
| "best_metric": 0.49536439776420593, | |
| "best_model_checkpoint": "output/llama2-7b-chat-hf/checkpoint-12000", | |
| "epoch": 1.8354236769654328, | |
| "eval_steps": 1000, | |
| "global_step": 12000, | |
| "is_hyper_param_search": false, | |
| "is_local_process_zero": true, | |
| "is_world_process_zero": true, | |
| "log_history": [ | |
| { | |
| "epoch": 0.08, | |
| "learning_rate": 4.923524013459774e-05, | |
| "loss": 0.9566, | |
| "step": 500 | |
| }, | |
| { | |
| "epoch": 0.15, | |
| "learning_rate": 4.8470480269195475e-05, | |
| "loss": 0.2966, | |
| "step": 1000 | |
| }, | |
| { | |
| "epoch": 0.15, | |
| "eval_loss": 0.5167073011398315, | |
| "eval_runtime": 38.9684, | |
| "eval_samples_per_second": 12.831, | |
| "eval_steps_per_second": 1.617, | |
| "step": 1000 | |
| }, | |
| { | |
| "epoch": 0.23, | |
| "learning_rate": 4.770572040379321e-05, | |
| "loss": 0.2828, | |
| "step": 1500 | |
| }, | |
| { | |
| "epoch": 0.31, | |
| "learning_rate": 4.694096053839095e-05, | |
| "loss": 0.2869, | |
| "step": 2000 | |
| }, | |
| { | |
| "epoch": 0.31, | |
| "eval_loss": 0.5082407593727112, | |
| "eval_runtime": 38.9667, | |
| "eval_samples_per_second": 12.831, | |
| "eval_steps_per_second": 1.617, | |
| "step": 2000 | |
| }, | |
| { | |
| "epoch": 0.38, | |
| "learning_rate": 4.617620067298868e-05, | |
| "loss": 0.2855, | |
| "step": 2500 | |
| }, | |
| { | |
| "epoch": 0.46, | |
| "learning_rate": 4.541144080758642e-05, | |
| "loss": 0.2754, | |
| "step": 3000 | |
| }, | |
| { | |
| "epoch": 0.46, | |
| "eval_loss": 0.5053573250770569, | |
| "eval_runtime": 38.968, | |
| "eval_samples_per_second": 12.831, | |
| "eval_steps_per_second": 1.617, | |
| "step": 3000 | |
| }, | |
| { | |
| "epoch": 0.54, | |
| "learning_rate": 4.4646680942184155e-05, | |
| "loss": 0.2909, | |
| "step": 3500 | |
| }, | |
| { | |
| "epoch": 0.61, | |
| "learning_rate": 4.388192107678189e-05, | |
| "loss": 0.2904, | |
| "step": 4000 | |
| }, | |
| { | |
| "epoch": 0.61, | |
| "eval_loss": 0.5021488070487976, | |
| "eval_runtime": 38.9639, | |
| "eval_samples_per_second": 12.832, | |
| "eval_steps_per_second": 1.617, | |
| "step": 4000 | |
| }, | |
| { | |
| "epoch": 0.69, | |
| "learning_rate": 4.311716121137963e-05, | |
| "loss": 0.2871, | |
| "step": 4500 | |
| }, | |
| { | |
| "epoch": 0.76, | |
| "learning_rate": 4.2352401345977364e-05, | |
| "loss": 0.2837, | |
| "step": 5000 | |
| }, | |
| { | |
| "epoch": 0.76, | |
| "eval_loss": 0.5003100037574768, | |
| "eval_runtime": 38.9665, | |
| "eval_samples_per_second": 12.832, | |
| "eval_steps_per_second": 1.617, | |
| "step": 5000 | |
| }, | |
| { | |
| "epoch": 0.84, | |
| "learning_rate": 4.15876414805751e-05, | |
| "loss": 0.2775, | |
| "step": 5500 | |
| }, | |
| { | |
| "epoch": 0.92, | |
| "learning_rate": 4.0822881615172836e-05, | |
| "loss": 0.2842, | |
| "step": 6000 | |
| }, | |
| { | |
| "epoch": 0.92, | |
| "eval_loss": 0.5002090334892273, | |
| "eval_runtime": 38.9654, | |
| "eval_samples_per_second": 12.832, | |
| "eval_steps_per_second": 1.617, | |
| "step": 6000 | |
| }, | |
| { | |
| "epoch": 0.99, | |
| "learning_rate": 4.005812174977057e-05, | |
| "loss": 0.29, | |
| "step": 6500 | |
| }, | |
| { | |
| "epoch": 1.07, | |
| "learning_rate": 3.9293361884368315e-05, | |
| "loss": 0.2851, | |
| "step": 7000 | |
| }, | |
| { | |
| "epoch": 1.07, | |
| "eval_loss": 0.49814751744270325, | |
| "eval_runtime": 38.9615, | |
| "eval_samples_per_second": 12.833, | |
| "eval_steps_per_second": 1.617, | |
| "step": 7000 | |
| }, | |
| { | |
| "epoch": 1.15, | |
| "learning_rate": 3.8528602018966045e-05, | |
| "loss": 0.2758, | |
| "step": 7500 | |
| }, | |
| { | |
| "epoch": 1.22, | |
| "learning_rate": 3.776384215356378e-05, | |
| "loss": 0.2763, | |
| "step": 8000 | |
| }, | |
| { | |
| "epoch": 1.22, | |
| "eval_loss": 0.498801052570343, | |
| "eval_runtime": 38.9636, | |
| "eval_samples_per_second": 12.832, | |
| "eval_steps_per_second": 1.617, | |
| "step": 8000 | |
| }, | |
| { | |
| "epoch": 1.3, | |
| "learning_rate": 3.699908228816152e-05, | |
| "loss": 0.2826, | |
| "step": 8500 | |
| }, | |
| { | |
| "epoch": 1.38, | |
| "learning_rate": 3.623432242275925e-05, | |
| "loss": 0.2712, | |
| "step": 9000 | |
| }, | |
| { | |
| "epoch": 1.38, | |
| "eval_loss": 0.49798327684402466, | |
| "eval_runtime": 38.9634, | |
| "eval_samples_per_second": 12.833, | |
| "eval_steps_per_second": 1.617, | |
| "step": 9000 | |
| }, | |
| { | |
| "epoch": 1.45, | |
| "learning_rate": 3.546956255735699e-05, | |
| "loss": 0.2775, | |
| "step": 9500 | |
| }, | |
| { | |
| "epoch": 1.53, | |
| "learning_rate": 3.470480269195473e-05, | |
| "loss": 0.2727, | |
| "step": 10000 | |
| }, | |
| { | |
| "epoch": 1.53, | |
| "eval_loss": 0.49744608998298645, | |
| "eval_runtime": 38.9669, | |
| "eval_samples_per_second": 12.831, | |
| "eval_steps_per_second": 1.617, | |
| "step": 10000 | |
| }, | |
| { | |
| "epoch": 1.61, | |
| "learning_rate": 3.394004282655247e-05, | |
| "loss": 0.2747, | |
| "step": 10500 | |
| }, | |
| { | |
| "epoch": 1.68, | |
| "learning_rate": 3.31752829611502e-05, | |
| "loss": 0.2711, | |
| "step": 11000 | |
| }, | |
| { | |
| "epoch": 1.68, | |
| "eval_loss": 0.49606311321258545, | |
| "eval_runtime": 38.9636, | |
| "eval_samples_per_second": 12.832, | |
| "eval_steps_per_second": 1.617, | |
| "step": 11000 | |
| }, | |
| { | |
| "epoch": 1.76, | |
| "learning_rate": 3.2410523095747934e-05, | |
| "loss": 0.2766, | |
| "step": 11500 | |
| }, | |
| { | |
| "epoch": 1.84, | |
| "learning_rate": 3.164576323034567e-05, | |
| "loss": 0.2785, | |
| "step": 12000 | |
| }, | |
| { | |
| "epoch": 1.84, | |
| "eval_loss": 0.49536439776420593, | |
| "eval_runtime": 38.9645, | |
| "eval_samples_per_second": 12.832, | |
| "eval_steps_per_second": 1.617, | |
| "step": 12000 | |
| } | |
| ], | |
| "logging_steps": 500, | |
| "max_steps": 32690, | |
| "num_train_epochs": 5, | |
| "save_steps": 1000, | |
| "total_flos": 2.1886994939904e+17, | |
| "trial_name": null, | |
| "trial_params": null | |
| } | |