| { | |
| "best_metric": null, | |
| "best_model_checkpoint": null, | |
| "epoch": 0.9920199230667882, | |
| "global_step": 24000, | |
| "is_hyper_param_search": false, | |
| "is_local_process_zero": true, | |
| "is_world_process_zero": true, | |
| "log_history": [ | |
| { | |
| "epoch": 0.02, | |
| "learning_rate": 9.999793741991486e-06, | |
| "loss": 1.742, | |
| "step": 500 | |
| }, | |
| { | |
| "epoch": 0.04, | |
| "learning_rate": 9.999587070640269e-06, | |
| "loss": 1.7307, | |
| "step": 1000 | |
| }, | |
| { | |
| "epoch": 0.06, | |
| "learning_rate": 9.999380399289051e-06, | |
| "loss": 1.7414, | |
| "step": 1500 | |
| }, | |
| { | |
| "epoch": 0.08, | |
| "learning_rate": 9.999173727937835e-06, | |
| "loss": 1.7396, | |
| "step": 2000 | |
| }, | |
| { | |
| "epoch": 0.08, | |
| "eval_accuracy": 0.6747098550914694, | |
| "eval_loss": 1.6244006156921387, | |
| "eval_runtime": 13.0499, | |
| "eval_samples_per_second": 306.516, | |
| "eval_steps_per_second": 0.766, | |
| "step": 2000 | |
| }, | |
| { | |
| "epoch": 0.1, | |
| "learning_rate": 9.998967056586617e-06, | |
| "loss": 1.738, | |
| "step": 2500 | |
| }, | |
| { | |
| "epoch": 0.12, | |
| "learning_rate": 9.9987603852354e-06, | |
| "loss": 1.7357, | |
| "step": 3000 | |
| }, | |
| { | |
| "epoch": 0.14, | |
| "learning_rate": 9.998553713884183e-06, | |
| "loss": 1.7349, | |
| "step": 3500 | |
| }, | |
| { | |
| "epoch": 0.17, | |
| "learning_rate": 9.998347042532966e-06, | |
| "loss": 1.7309, | |
| "step": 4000 | |
| }, | |
| { | |
| "epoch": 0.17, | |
| "eval_accuracy": 0.6821878279118573, | |
| "eval_loss": 1.6006274223327637, | |
| "eval_runtime": 13.0815, | |
| "eval_samples_per_second": 305.775, | |
| "eval_steps_per_second": 0.764, | |
| "step": 4000 | |
| }, | |
| { | |
| "epoch": 0.19, | |
| "learning_rate": 9.998140371181748e-06, | |
| "loss": 1.7322, | |
| "step": 4500 | |
| }, | |
| { | |
| "epoch": 0.21, | |
| "learning_rate": 9.99793369983053e-06, | |
| "loss": 1.7302, | |
| "step": 5000 | |
| }, | |
| { | |
| "epoch": 0.23, | |
| "learning_rate": 9.997727028479312e-06, | |
| "loss": 1.7281, | |
| "step": 5500 | |
| }, | |
| { | |
| "epoch": 0.25, | |
| "learning_rate": 9.997520357128095e-06, | |
| "loss": 1.7273, | |
| "step": 6000 | |
| }, | |
| { | |
| "epoch": 0.25, | |
| "eval_accuracy": 0.6896347482724581, | |
| "eval_loss": 1.5701262950897217, | |
| "eval_runtime": 13.0657, | |
| "eval_samples_per_second": 306.145, | |
| "eval_steps_per_second": 0.765, | |
| "step": 6000 | |
| }, | |
| { | |
| "epoch": 0.27, | |
| "learning_rate": 9.997313685776879e-06, | |
| "loss": 1.7252, | |
| "step": 6500 | |
| }, | |
| { | |
| "epoch": 0.29, | |
| "learning_rate": 9.997107427768363e-06, | |
| "loss": 1.7231, | |
| "step": 7000 | |
| }, | |
| { | |
| "epoch": 0.31, | |
| "learning_rate": 9.996901169759848e-06, | |
| "loss": 1.7223, | |
| "step": 7500 | |
| }, | |
| { | |
| "epoch": 0.33, | |
| "learning_rate": 9.99669449840863e-06, | |
| "loss": 1.7215, | |
| "step": 8000 | |
| }, | |
| { | |
| "epoch": 0.33, | |
| "eval_accuracy": 0.6799366211130917, | |
| "eval_loss": 1.6207975149154663, | |
| "eval_runtime": 11.6877, | |
| "eval_samples_per_second": 342.241, | |
| "eval_steps_per_second": 0.856, | |
| "step": 8000 | |
| }, | |
| { | |
| "epoch": 0.35, | |
| "learning_rate": 9.996487827057414e-06, | |
| "loss": 1.7182, | |
| "step": 8500 | |
| }, | |
| { | |
| "epoch": 0.37, | |
| "learning_rate": 9.996281155706197e-06, | |
| "loss": 1.7189, | |
| "step": 9000 | |
| }, | |
| { | |
| "epoch": 0.39, | |
| "learning_rate": 9.996074484354979e-06, | |
| "loss": 1.7169, | |
| "step": 9500 | |
| }, | |
| { | |
| "epoch": 0.41, | |
| "learning_rate": 9.995868226346464e-06, | |
| "loss": 1.7173, | |
| "step": 10000 | |
| }, | |
| { | |
| "epoch": 0.41, | |
| "eval_accuracy": 0.6820005349023803, | |
| "eval_loss": 1.6004101037979126, | |
| "eval_runtime": 13.0596, | |
| "eval_samples_per_second": 306.287, | |
| "eval_steps_per_second": 0.766, | |
| "step": 10000 | |
| }, | |
| { | |
| "epoch": 0.43, | |
| "learning_rate": 9.995661554995247e-06, | |
| "loss": 1.7152, | |
| "step": 10500 | |
| }, | |
| { | |
| "epoch": 0.45, | |
| "learning_rate": 9.99545488364403e-06, | |
| "loss": 1.7126, | |
| "step": 11000 | |
| }, | |
| { | |
| "epoch": 0.48, | |
| "learning_rate": 9.995248212292813e-06, | |
| "loss": 1.7114, | |
| "step": 11500 | |
| }, | |
| { | |
| "epoch": 0.5, | |
| "learning_rate": 9.995041954284299e-06, | |
| "loss": 1.7098, | |
| "step": 12000 | |
| }, | |
| { | |
| "epoch": 0.5, | |
| "eval_accuracy": 0.6854211592498203, | |
| "eval_loss": 1.6027569770812988, | |
| "eval_runtime": 13.0635, | |
| "eval_samples_per_second": 306.196, | |
| "eval_steps_per_second": 0.765, | |
| "step": 12000 | |
| }, | |
| { | |
| "epoch": 0.52, | |
| "learning_rate": 9.99483528293308e-06, | |
| "loss": 1.7092, | |
| "step": 12500 | |
| }, | |
| { | |
| "epoch": 0.54, | |
| "learning_rate": 9.994628611581863e-06, | |
| "loss": 1.7081, | |
| "step": 13000 | |
| }, | |
| { | |
| "epoch": 0.56, | |
| "learning_rate": 9.994421940230647e-06, | |
| "loss": 1.7071, | |
| "step": 13500 | |
| }, | |
| { | |
| "epoch": 0.58, | |
| "learning_rate": 9.99421526887943e-06, | |
| "loss": 1.7066, | |
| "step": 14000 | |
| }, | |
| { | |
| "epoch": 0.58, | |
| "eval_accuracy": 0.6892430278884463, | |
| "eval_loss": 1.568811058998108, | |
| "eval_runtime": 11.6681, | |
| "eval_samples_per_second": 342.815, | |
| "eval_steps_per_second": 0.857, | |
| "step": 14000 | |
| }, | |
| { | |
| "epoch": 0.6, | |
| "learning_rate": 9.994008597528212e-06, | |
| "loss": 1.7062, | |
| "step": 14500 | |
| }, | |
| { | |
| "epoch": 0.62, | |
| "learning_rate": 9.993801926176994e-06, | |
| "loss": 1.7041, | |
| "step": 15000 | |
| }, | |
| { | |
| "epoch": 0.64, | |
| "learning_rate": 9.993595254825776e-06, | |
| "loss": 1.7028, | |
| "step": 15500 | |
| }, | |
| { | |
| "epoch": 0.66, | |
| "learning_rate": 9.99338858347456e-06, | |
| "loss": 1.702, | |
| "step": 16000 | |
| }, | |
| { | |
| "epoch": 0.66, | |
| "eval_accuracy": 0.689404839764552, | |
| "eval_loss": 1.5777287483215332, | |
| "eval_runtime": 13.0408, | |
| "eval_samples_per_second": 306.729, | |
| "eval_steps_per_second": 0.767, | |
| "step": 16000 | |
| }, | |
| { | |
| "epoch": 0.68, | |
| "learning_rate": 9.993181912123342e-06, | |
| "loss": 1.6984, | |
| "step": 16500 | |
| }, | |
| { | |
| "epoch": 0.7, | |
| "learning_rate": 9.992975240772125e-06, | |
| "loss": 1.6999, | |
| "step": 17000 | |
| }, | |
| { | |
| "epoch": 0.72, | |
| "learning_rate": 9.99276898276361e-06, | |
| "loss": 1.6983, | |
| "step": 17500 | |
| }, | |
| { | |
| "epoch": 0.74, | |
| "learning_rate": 9.992562724755096e-06, | |
| "loss": 1.6972, | |
| "step": 18000 | |
| }, | |
| { | |
| "epoch": 0.74, | |
| "eval_accuracy": 0.6915187376725839, | |
| "eval_loss": 1.5638251304626465, | |
| "eval_runtime": 11.6615, | |
| "eval_samples_per_second": 343.008, | |
| "eval_steps_per_second": 0.858, | |
| "step": 18000 | |
| }, | |
| { | |
| "epoch": 0.76, | |
| "learning_rate": 9.992356053403878e-06, | |
| "loss": 1.694, | |
| "step": 18500 | |
| }, | |
| { | |
| "epoch": 0.79, | |
| "learning_rate": 9.992149795395364e-06, | |
| "loss": 1.6954, | |
| "step": 19000 | |
| }, | |
| { | |
| "epoch": 0.81, | |
| "learning_rate": 9.991943124044146e-06, | |
| "loss": 1.6946, | |
| "step": 19500 | |
| }, | |
| { | |
| "epoch": 0.83, | |
| "learning_rate": 9.991736452692928e-06, | |
| "loss": 1.6939, | |
| "step": 20000 | |
| }, | |
| { | |
| "epoch": 0.83, | |
| "eval_accuracy": 0.6928982725527831, | |
| "eval_loss": 1.5452697277069092, | |
| "eval_runtime": 13.0226, | |
| "eval_samples_per_second": 307.159, | |
| "eval_steps_per_second": 0.768, | |
| "step": 20000 | |
| }, | |
| { | |
| "epoch": 0.85, | |
| "learning_rate": 9.99152978134171e-06, | |
| "loss": 1.6927, | |
| "step": 20500 | |
| }, | |
| { | |
| "epoch": 0.87, | |
| "learning_rate": 9.991323109990494e-06, | |
| "loss": 1.6901, | |
| "step": 21000 | |
| }, | |
| { | |
| "epoch": 0.89, | |
| "learning_rate": 9.991116438639277e-06, | |
| "loss": 1.6892, | |
| "step": 21500 | |
| }, | |
| { | |
| "epoch": 0.91, | |
| "learning_rate": 9.990909767288059e-06, | |
| "loss": 1.6898, | |
| "step": 22000 | |
| }, | |
| { | |
| "epoch": 0.91, | |
| "eval_accuracy": 0.683527801316733, | |
| "eval_loss": 1.5818458795547485, | |
| "eval_runtime": 13.0663, | |
| "eval_samples_per_second": 306.132, | |
| "eval_steps_per_second": 0.765, | |
| "step": 22000 | |
| }, | |
| { | |
| "epoch": 0.93, | |
| "learning_rate": 9.990703095936843e-06, | |
| "loss": 1.6889, | |
| "step": 22500 | |
| }, | |
| { | |
| "epoch": 0.95, | |
| "learning_rate": 9.990496424585625e-06, | |
| "loss": 1.6855, | |
| "step": 23000 | |
| }, | |
| { | |
| "epoch": 0.97, | |
| "learning_rate": 9.990289753234407e-06, | |
| "loss": 1.6853, | |
| "step": 23500 | |
| }, | |
| { | |
| "epoch": 0.99, | |
| "learning_rate": 9.990083908568595e-06, | |
| "loss": 1.6856, | |
| "step": 24000 | |
| }, | |
| { | |
| "epoch": 0.99, | |
| "eval_accuracy": 0.6877091672681934, | |
| "eval_loss": 1.5628535747528076, | |
| "eval_runtime": 13.0163, | |
| "eval_samples_per_second": 307.306, | |
| "eval_steps_per_second": 0.768, | |
| "step": 24000 | |
| } | |
| ], | |
| "max_steps": 24193000, | |
| "num_train_epochs": 1000, | |
| "total_flos": 1.78960468082688e+19, | |
| "trial_name": null, | |
| "trial_params": null | |
| } | |