| { | |
| "best_metric": null, | |
| "best_model_checkpoint": null, | |
| "epoch": 3.0, | |
| "eval_steps": 10, | |
| "global_step": 81, | |
| "is_hyper_param_search": false, | |
| "is_local_process_zero": true, | |
| "is_world_process_zero": true, | |
| "log_history": [ | |
| { | |
| "epoch": 0.04, | |
| "learning_rate": 0.0, | |
| "loss": 1.0172, | |
| "step": 1 | |
| }, | |
| { | |
| "epoch": 0.07, | |
| "learning_rate": 0.00018927892607143717, | |
| "loss": 0.9587, | |
| "step": 2 | |
| }, | |
| { | |
| "epoch": 0.11, | |
| "learning_rate": 0.0003, | |
| "loss": 0.9422, | |
| "step": 3 | |
| }, | |
| { | |
| "epoch": 0.15, | |
| "learning_rate": 0.0003, | |
| "loss": 0.9651, | |
| "step": 4 | |
| }, | |
| { | |
| "epoch": 0.19, | |
| "learning_rate": 0.0003, | |
| "loss": 1.0197, | |
| "step": 5 | |
| }, | |
| { | |
| "epoch": 0.22, | |
| "learning_rate": 0.0003, | |
| "loss": 0.9455, | |
| "step": 6 | |
| }, | |
| { | |
| "epoch": 0.26, | |
| "learning_rate": 0.0003, | |
| "loss": 0.9029, | |
| "step": 7 | |
| }, | |
| { | |
| "epoch": 0.3, | |
| "learning_rate": 0.0003, | |
| "loss": 0.9045, | |
| "step": 8 | |
| }, | |
| { | |
| "epoch": 0.33, | |
| "learning_rate": 0.0003, | |
| "loss": 0.8424, | |
| "step": 9 | |
| }, | |
| { | |
| "epoch": 0.37, | |
| "learning_rate": 0.0003, | |
| "loss": 0.882, | |
| "step": 10 | |
| }, | |
| { | |
| "epoch": 0.37, | |
| "eval_accuracy": 0.22187782513237764, | |
| "eval_loss": 0.8447974324226379, | |
| "eval_runtime": 44.6213, | |
| "eval_samples_per_second": 0.807, | |
| "eval_steps_per_second": 0.045, | |
| "step": 10 | |
| }, | |
| { | |
| "epoch": 0.41, | |
| "learning_rate": 0.0003, | |
| "loss": 0.8986, | |
| "step": 11 | |
| }, | |
| { | |
| "epoch": 0.44, | |
| "learning_rate": 0.0003, | |
| "loss": 0.9005, | |
| "step": 12 | |
| }, | |
| { | |
| "epoch": 0.48, | |
| "learning_rate": 0.0003, | |
| "loss": 0.8788, | |
| "step": 13 | |
| }, | |
| { | |
| "epoch": 0.52, | |
| "learning_rate": 0.0003, | |
| "loss": 0.8391, | |
| "step": 14 | |
| }, | |
| { | |
| "epoch": 0.56, | |
| "learning_rate": 0.0003, | |
| "loss": 0.8564, | |
| "step": 15 | |
| }, | |
| { | |
| "epoch": 0.59, | |
| "learning_rate": 0.0003, | |
| "loss": 0.84, | |
| "step": 16 | |
| }, | |
| { | |
| "epoch": 0.63, | |
| "learning_rate": 0.0003, | |
| "loss": 0.8479, | |
| "step": 17 | |
| }, | |
| { | |
| "epoch": 0.67, | |
| "learning_rate": 0.0003, | |
| "loss": 0.8174, | |
| "step": 18 | |
| }, | |
| { | |
| "epoch": 0.7, | |
| "learning_rate": 0.0003, | |
| "loss": 0.8431, | |
| "step": 19 | |
| }, | |
| { | |
| "epoch": 0.74, | |
| "learning_rate": 0.0003, | |
| "loss": 0.8456, | |
| "step": 20 | |
| }, | |
| { | |
| "epoch": 0.74, | |
| "eval_accuracy": 0.2230832149468337, | |
| "eval_loss": 0.8235660195350647, | |
| "eval_runtime": 43.3512, | |
| "eval_samples_per_second": 0.83, | |
| "eval_steps_per_second": 0.046, | |
| "step": 20 | |
| }, | |
| { | |
| "epoch": 0.78, | |
| "learning_rate": 0.0003, | |
| "loss": 0.8614, | |
| "step": 21 | |
| }, | |
| { | |
| "epoch": 0.81, | |
| "learning_rate": 0.0003, | |
| "loss": 0.8756, | |
| "step": 22 | |
| }, | |
| { | |
| "epoch": 0.85, | |
| "learning_rate": 0.0003, | |
| "loss": 0.7752, | |
| "step": 23 | |
| }, | |
| { | |
| "epoch": 0.89, | |
| "learning_rate": 0.0003, | |
| "loss": 0.7891, | |
| "step": 24 | |
| }, | |
| { | |
| "epoch": 0.93, | |
| "learning_rate": 0.0003, | |
| "loss": 0.8094, | |
| "step": 25 | |
| }, | |
| { | |
| "epoch": 0.96, | |
| "learning_rate": 0.0003, | |
| "loss": 0.8546, | |
| "step": 26 | |
| }, | |
| { | |
| "epoch": 1.0, | |
| "learning_rate": 0.0003, | |
| "loss": 0.812, | |
| "step": 27 | |
| }, | |
| { | |
| "epoch": 1.04, | |
| "learning_rate": 0.0003, | |
| "loss": 0.769, | |
| "step": 28 | |
| }, | |
| { | |
| "epoch": 1.07, | |
| "learning_rate": 0.0003, | |
| "loss": 0.7538, | |
| "step": 29 | |
| }, | |
| { | |
| "epoch": 1.11, | |
| "learning_rate": 0.0003, | |
| "loss": 0.7431, | |
| "step": 30 | |
| }, | |
| { | |
| "epoch": 1.11, | |
| "eval_accuracy": 0.2239011580352146, | |
| "eval_loss": 0.8174943923950195, | |
| "eval_runtime": 43.2867, | |
| "eval_samples_per_second": 0.832, | |
| "eval_steps_per_second": 0.046, | |
| "step": 30 | |
| }, | |
| { | |
| "epoch": 1.15, | |
| "learning_rate": 0.0003, | |
| "loss": 0.812, | |
| "step": 31 | |
| }, | |
| { | |
| "epoch": 1.19, | |
| "learning_rate": 0.0003, | |
| "loss": 0.8473, | |
| "step": 32 | |
| }, | |
| { | |
| "epoch": 1.22, | |
| "learning_rate": 0.0003, | |
| "loss": 0.7627, | |
| "step": 33 | |
| }, | |
| { | |
| "epoch": 1.26, | |
| "learning_rate": 0.0003, | |
| "loss": 0.7231, | |
| "step": 34 | |
| }, | |
| { | |
| "epoch": 1.3, | |
| "learning_rate": 0.0003, | |
| "loss": 0.7546, | |
| "step": 35 | |
| }, | |
| { | |
| "epoch": 1.33, | |
| "learning_rate": 0.0003, | |
| "loss": 0.6798, | |
| "step": 36 | |
| }, | |
| { | |
| "epoch": 1.37, | |
| "learning_rate": 0.0003, | |
| "loss": 0.7154, | |
| "step": 37 | |
| }, | |
| { | |
| "epoch": 1.41, | |
| "learning_rate": 0.0003, | |
| "loss": 0.8383, | |
| "step": 38 | |
| }, | |
| { | |
| "epoch": 1.44, | |
| "learning_rate": 0.0003, | |
| "loss": 0.7747, | |
| "step": 39 | |
| }, | |
| { | |
| "epoch": 1.48, | |
| "learning_rate": 0.0003, | |
| "loss": 0.7121, | |
| "step": 40 | |
| }, | |
| { | |
| "epoch": 1.48, | |
| "eval_accuracy": 0.22361057299065823, | |
| "eval_loss": 0.830994188785553, | |
| "eval_runtime": 43.2574, | |
| "eval_samples_per_second": 0.832, | |
| "eval_steps_per_second": 0.046, | |
| "step": 40 | |
| }, | |
| { | |
| "epoch": 1.52, | |
| "learning_rate": 0.0003, | |
| "loss": 0.7714, | |
| "step": 41 | |
| }, | |
| { | |
| "epoch": 1.56, | |
| "learning_rate": 0.0003, | |
| "loss": 0.7679, | |
| "step": 42 | |
| }, | |
| { | |
| "epoch": 1.59, | |
| "learning_rate": 0.0003, | |
| "loss": 0.6597, | |
| "step": 43 | |
| }, | |
| { | |
| "epoch": 1.63, | |
| "learning_rate": 0.0003, | |
| "loss": 0.73, | |
| "step": 44 | |
| }, | |
| { | |
| "epoch": 1.67, | |
| "learning_rate": 0.0003, | |
| "loss": 0.7872, | |
| "step": 45 | |
| }, | |
| { | |
| "epoch": 1.7, | |
| "learning_rate": 0.0003, | |
| "loss": 0.7107, | |
| "step": 46 | |
| }, | |
| { | |
| "epoch": 1.74, | |
| "learning_rate": 0.0003, | |
| "loss": 0.6862, | |
| "step": 47 | |
| }, | |
| { | |
| "epoch": 1.78, | |
| "learning_rate": 0.0003, | |
| "loss": 0.6835, | |
| "step": 48 | |
| }, | |
| { | |
| "epoch": 1.81, | |
| "learning_rate": 0.0003, | |
| "loss": 0.8554, | |
| "step": 49 | |
| }, | |
| { | |
| "epoch": 1.85, | |
| "learning_rate": 0.0003, | |
| "loss": 0.6281, | |
| "step": 50 | |
| }, | |
| { | |
| "epoch": 1.85, | |
| "eval_accuracy": 0.22325541349175598, | |
| "eval_loss": 0.8365404605865479, | |
| "eval_runtime": 43.3797, | |
| "eval_samples_per_second": 0.83, | |
| "eval_steps_per_second": 0.046, | |
| "step": 50 | |
| }, | |
| { | |
| "epoch": 1.89, | |
| "learning_rate": 0.0003, | |
| "loss": 0.6688, | |
| "step": 51 | |
| }, | |
| { | |
| "epoch": 1.93, | |
| "learning_rate": 0.0003, | |
| "loss": 0.7317, | |
| "step": 52 | |
| }, | |
| { | |
| "epoch": 1.96, | |
| "learning_rate": 0.0003, | |
| "loss": 0.7165, | |
| "step": 53 | |
| }, | |
| { | |
| "epoch": 2.0, | |
| "learning_rate": 0.0003, | |
| "loss": 0.5791, | |
| "step": 54 | |
| }, | |
| { | |
| "epoch": 2.04, | |
| "learning_rate": 0.0003, | |
| "loss": 0.5359, | |
| "step": 55 | |
| }, | |
| { | |
| "epoch": 2.07, | |
| "learning_rate": 0.0003, | |
| "loss": 0.636, | |
| "step": 56 | |
| }, | |
| { | |
| "epoch": 2.11, | |
| "learning_rate": 0.0003, | |
| "loss": 0.5707, | |
| "step": 57 | |
| }, | |
| { | |
| "epoch": 2.15, | |
| "learning_rate": 0.0003, | |
| "loss": 0.7385, | |
| "step": 58 | |
| }, | |
| { | |
| "epoch": 2.19, | |
| "learning_rate": 0.0003, | |
| "loss": 0.6318, | |
| "step": 59 | |
| }, | |
| { | |
| "epoch": 2.22, | |
| "learning_rate": 0.0003, | |
| "loss": 0.6325, | |
| "step": 60 | |
| }, | |
| { | |
| "epoch": 2.22, | |
| "eval_accuracy": 0.22314778940117955, | |
| "eval_loss": 0.8458040952682495, | |
| "eval_runtime": 43.5251, | |
| "eval_samples_per_second": 0.827, | |
| "eval_steps_per_second": 0.046, | |
| "step": 60 | |
| }, | |
| { | |
| "epoch": 2.26, | |
| "learning_rate": 0.0003, | |
| "loss": 0.6493, | |
| "step": 61 | |
| }, | |
| { | |
| "epoch": 2.3, | |
| "learning_rate": 0.0003, | |
| "loss": 0.5792, | |
| "step": 62 | |
| }, | |
| { | |
| "epoch": 2.33, | |
| "learning_rate": 0.0003, | |
| "loss": 0.5246, | |
| "step": 63 | |
| }, | |
| { | |
| "epoch": 2.37, | |
| "learning_rate": 0.0003, | |
| "loss": 0.5364, | |
| "step": 64 | |
| }, | |
| { | |
| "epoch": 2.41, | |
| "learning_rate": 0.0003, | |
| "loss": 0.829, | |
| "step": 65 | |
| }, | |
| { | |
| "epoch": 2.44, | |
| "learning_rate": 0.0003, | |
| "loss": 0.5891, | |
| "step": 66 | |
| }, | |
| { | |
| "epoch": 2.48, | |
| "learning_rate": 0.0003, | |
| "loss": 0.5092, | |
| "step": 67 | |
| }, | |
| { | |
| "epoch": 2.52, | |
| "learning_rate": 0.0003, | |
| "loss": 0.7564, | |
| "step": 68 | |
| }, | |
| { | |
| "epoch": 2.56, | |
| "learning_rate": 0.0003, | |
| "loss": 0.747, | |
| "step": 69 | |
| }, | |
| { | |
| "epoch": 2.59, | |
| "learning_rate": 0.0003, | |
| "loss": 0.6826, | |
| "step": 70 | |
| }, | |
| { | |
| "epoch": 2.59, | |
| "eval_accuracy": 0.222168410176934, | |
| "eval_loss": 0.8560571670532227, | |
| "eval_runtime": 43.3441, | |
| "eval_samples_per_second": 0.831, | |
| "eval_steps_per_second": 0.046, | |
| "step": 70 | |
| }, | |
| { | |
| "epoch": 2.63, | |
| "learning_rate": 0.0003, | |
| "loss": 0.7208, | |
| "step": 71 | |
| }, | |
| { | |
| "epoch": 2.67, | |
| "learning_rate": 0.0003, | |
| "loss": 0.7304, | |
| "step": 72 | |
| }, | |
| { | |
| "epoch": 2.7, | |
| "learning_rate": 0.0003, | |
| "loss": 0.5485, | |
| "step": 73 | |
| }, | |
| { | |
| "epoch": 2.74, | |
| "learning_rate": 0.0003, | |
| "loss": 0.5899, | |
| "step": 74 | |
| }, | |
| { | |
| "epoch": 2.78, | |
| "learning_rate": 0.0003, | |
| "loss": 0.5291, | |
| "step": 75 | |
| }, | |
| { | |
| "epoch": 2.81, | |
| "learning_rate": 0.0003, | |
| "loss": 0.6907, | |
| "step": 76 | |
| }, | |
| { | |
| "epoch": 2.85, | |
| "learning_rate": 0.0003, | |
| "loss": 0.5826, | |
| "step": 77 | |
| }, | |
| { | |
| "epoch": 2.89, | |
| "learning_rate": 0.0003, | |
| "loss": 0.5553, | |
| "step": 78 | |
| }, | |
| { | |
| "epoch": 2.93, | |
| "learning_rate": 0.0003, | |
| "loss": 0.6393, | |
| "step": 79 | |
| }, | |
| { | |
| "epoch": 2.96, | |
| "learning_rate": 0.0003, | |
| "loss": 0.5822, | |
| "step": 80 | |
| }, | |
| { | |
| "epoch": 2.96, | |
| "eval_accuracy": 0.2210275948168238, | |
| "eval_loss": 0.8713862895965576, | |
| "eval_runtime": 42.5816, | |
| "eval_samples_per_second": 0.845, | |
| "eval_steps_per_second": 0.047, | |
| "step": 80 | |
| }, | |
| { | |
| "epoch": 3.0, | |
| "learning_rate": 0.0003, | |
| "loss": 0.5126, | |
| "step": 81 | |
| } | |
| ], | |
| "logging_steps": 1.0, | |
| "max_steps": 81, | |
| "num_input_tokens_seen": 0, | |
| "num_train_epochs": 3, | |
| "save_steps": 500, | |
| "total_flos": 10840779079680.0, | |
| "train_batch_size": 2, | |
| "trial_name": null, | |
| "trial_params": null | |
| } | |