| { | |
| "best_metric": null, | |
| "best_model_checkpoint": null, | |
| "epoch": 10.0, | |
| "eval_steps": 500, | |
| "global_step": 90, | |
| "is_hyper_param_search": false, | |
| "is_local_process_zero": true, | |
| "is_world_process_zero": true, | |
| "log_history": [ | |
| { | |
| "epoch": 0.11, | |
| "learning_rate": 0.0, | |
| "loss": 1.8453, | |
| "step": 1 | |
| }, | |
| { | |
| "epoch": 0.22, | |
| "learning_rate": 0.00012618595071429146, | |
| "loss": 1.3368, | |
| "step": 2 | |
| }, | |
| { | |
| "epoch": 0.33, | |
| "learning_rate": 0.0002, | |
| "loss": 1.4472, | |
| "step": 3 | |
| }, | |
| { | |
| "epoch": 0.44, | |
| "learning_rate": 0.0002, | |
| "loss": 1.2461, | |
| "step": 4 | |
| }, | |
| { | |
| "epoch": 0.56, | |
| "learning_rate": 0.0002, | |
| "loss": 1.1358, | |
| "step": 5 | |
| }, | |
| { | |
| "epoch": 0.67, | |
| "learning_rate": 0.0002, | |
| "loss": 1.0415, | |
| "step": 6 | |
| }, | |
| { | |
| "epoch": 0.78, | |
| "learning_rate": 0.0002, | |
| "loss": 0.9974, | |
| "step": 7 | |
| }, | |
| { | |
| "epoch": 0.89, | |
| "learning_rate": 0.0002, | |
| "loss": 0.8689, | |
| "step": 8 | |
| }, | |
| { | |
| "epoch": 1.0, | |
| "learning_rate": 0.0002, | |
| "loss": 0.8996, | |
| "step": 9 | |
| }, | |
| { | |
| "epoch": 1.0, | |
| "eval_loss": 1.0149524211883545, | |
| "eval_runtime": 6.1319, | |
| "eval_samples_per_second": 8.154, | |
| "eval_steps_per_second": 0.326, | |
| "step": 9 | |
| }, | |
| { | |
| "epoch": 1.11, | |
| "learning_rate": 0.0002, | |
| "loss": 0.6935, | |
| "step": 10 | |
| }, | |
| { | |
| "epoch": 1.22, | |
| "learning_rate": 0.0002, | |
| "loss": 0.5113, | |
| "step": 11 | |
| }, | |
| { | |
| "epoch": 1.33, | |
| "learning_rate": 0.0002, | |
| "loss": 0.488, | |
| "step": 12 | |
| }, | |
| { | |
| "epoch": 1.44, | |
| "learning_rate": 0.0002, | |
| "loss": 0.5482, | |
| "step": 13 | |
| }, | |
| { | |
| "epoch": 1.56, | |
| "learning_rate": 0.0002, | |
| "loss": 0.4923, | |
| "step": 14 | |
| }, | |
| { | |
| "epoch": 1.67, | |
| "learning_rate": 0.0002, | |
| "loss": 0.4704, | |
| "step": 15 | |
| }, | |
| { | |
| "epoch": 1.78, | |
| "learning_rate": 0.0002, | |
| "loss": 0.3157, | |
| "step": 16 | |
| }, | |
| { | |
| "epoch": 1.89, | |
| "learning_rate": 0.0002, | |
| "loss": 0.2821, | |
| "step": 17 | |
| }, | |
| { | |
| "epoch": 2.0, | |
| "learning_rate": 0.0002, | |
| "loss": 0.2338, | |
| "step": 18 | |
| }, | |
| { | |
| "epoch": 2.0, | |
| "eval_loss": 1.1113728284835815, | |
| "eval_runtime": 6.1196, | |
| "eval_samples_per_second": 8.17, | |
| "eval_steps_per_second": 0.327, | |
| "step": 18 | |
| }, | |
| { | |
| "epoch": 2.11, | |
| "learning_rate": 0.0002, | |
| "loss": 0.171, | |
| "step": 19 | |
| }, | |
| { | |
| "epoch": 2.22, | |
| "learning_rate": 0.0002, | |
| "loss": 0.1126, | |
| "step": 20 | |
| }, | |
| { | |
| "epoch": 2.33, | |
| "learning_rate": 0.0002, | |
| "loss": 0.1116, | |
| "step": 21 | |
| }, | |
| { | |
| "epoch": 2.44, | |
| "learning_rate": 0.0002, | |
| "loss": 0.1327, | |
| "step": 22 | |
| }, | |
| { | |
| "epoch": 2.56, | |
| "learning_rate": 0.0002, | |
| "loss": 0.09, | |
| "step": 23 | |
| }, | |
| { | |
| "epoch": 2.67, | |
| "learning_rate": 0.0002, | |
| "loss": 0.0599, | |
| "step": 24 | |
| }, | |
| { | |
| "epoch": 2.78, | |
| "learning_rate": 0.0002, | |
| "loss": 0.1018, | |
| "step": 25 | |
| }, | |
| { | |
| "epoch": 2.89, | |
| "learning_rate": 0.0002, | |
| "loss": 0.0996, | |
| "step": 26 | |
| }, | |
| { | |
| "epoch": 3.0, | |
| "learning_rate": 0.0002, | |
| "loss": 0.0743, | |
| "step": 27 | |
| }, | |
| { | |
| "epoch": 3.0, | |
| "eval_loss": 1.183627724647522, | |
| "eval_runtime": 6.1208, | |
| "eval_samples_per_second": 8.169, | |
| "eval_steps_per_second": 0.327, | |
| "step": 27 | |
| }, | |
| { | |
| "epoch": 3.11, | |
| "learning_rate": 0.0002, | |
| "loss": 0.0415, | |
| "step": 28 | |
| }, | |
| { | |
| "epoch": 3.22, | |
| "learning_rate": 0.0002, | |
| "loss": 0.0361, | |
| "step": 29 | |
| }, | |
| { | |
| "epoch": 3.33, | |
| "learning_rate": 0.0002, | |
| "loss": 0.0414, | |
| "step": 30 | |
| }, | |
| { | |
| "epoch": 3.44, | |
| "learning_rate": 0.0002, | |
| "loss": 0.0251, | |
| "step": 31 | |
| }, | |
| { | |
| "epoch": 3.56, | |
| "learning_rate": 0.0002, | |
| "loss": 0.0348, | |
| "step": 32 | |
| }, | |
| { | |
| "epoch": 3.67, | |
| "learning_rate": 0.0002, | |
| "loss": 0.0364, | |
| "step": 33 | |
| }, | |
| { | |
| "epoch": 3.78, | |
| "learning_rate": 0.0002, | |
| "loss": 0.0334, | |
| "step": 34 | |
| }, | |
| { | |
| "epoch": 3.89, | |
| "learning_rate": 0.0002, | |
| "loss": 0.0443, | |
| "step": 35 | |
| }, | |
| { | |
| "epoch": 4.0, | |
| "learning_rate": 0.0002, | |
| "loss": 0.0278, | |
| "step": 36 | |
| }, | |
| { | |
| "epoch": 4.0, | |
| "eval_loss": 1.4033983945846558, | |
| "eval_runtime": 6.2763, | |
| "eval_samples_per_second": 7.966, | |
| "eval_steps_per_second": 0.319, | |
| "step": 36 | |
| }, | |
| { | |
| "epoch": 4.11, | |
| "learning_rate": 0.0002, | |
| "loss": 0.0107, | |
| "step": 37 | |
| }, | |
| { | |
| "epoch": 4.22, | |
| "learning_rate": 0.0002, | |
| "loss": 0.0149, | |
| "step": 38 | |
| }, | |
| { | |
| "epoch": 4.33, | |
| "learning_rate": 0.0002, | |
| "loss": 0.0131, | |
| "step": 39 | |
| }, | |
| { | |
| "epoch": 4.44, | |
| "learning_rate": 0.0002, | |
| "loss": 0.0121, | |
| "step": 40 | |
| }, | |
| { | |
| "epoch": 4.56, | |
| "learning_rate": 0.0002, | |
| "loss": 0.0107, | |
| "step": 41 | |
| }, | |
| { | |
| "epoch": 4.67, | |
| "learning_rate": 0.0002, | |
| "loss": 0.0186, | |
| "step": 42 | |
| }, | |
| { | |
| "epoch": 4.78, | |
| "learning_rate": 0.0002, | |
| "loss": 0.0241, | |
| "step": 43 | |
| }, | |
| { | |
| "epoch": 4.89, | |
| "learning_rate": 0.0002, | |
| "loss": 0.0122, | |
| "step": 44 | |
| }, | |
| { | |
| "epoch": 5.0, | |
| "learning_rate": 0.0002, | |
| "loss": 0.0122, | |
| "step": 45 | |
| }, | |
| { | |
| "epoch": 5.0, | |
| "eval_loss": 1.4663941860198975, | |
| "eval_runtime": 6.1899, | |
| "eval_samples_per_second": 8.078, | |
| "eval_steps_per_second": 0.323, | |
| "step": 45 | |
| }, | |
| { | |
| "epoch": 5.11, | |
| "learning_rate": 0.0002, | |
| "loss": 0.0065, | |
| "step": 46 | |
| }, | |
| { | |
| "epoch": 5.22, | |
| "learning_rate": 0.0002, | |
| "loss": 0.0114, | |
| "step": 47 | |
| }, | |
| { | |
| "epoch": 5.33, | |
| "learning_rate": 0.0002, | |
| "loss": 0.0099, | |
| "step": 48 | |
| }, | |
| { | |
| "epoch": 5.44, | |
| "learning_rate": 0.0002, | |
| "loss": 0.0098, | |
| "step": 49 | |
| }, | |
| { | |
| "epoch": 5.56, | |
| "learning_rate": 0.0002, | |
| "loss": 0.0099, | |
| "step": 50 | |
| }, | |
| { | |
| "epoch": 5.67, | |
| "learning_rate": 0.0002, | |
| "loss": 0.0163, | |
| "step": 51 | |
| }, | |
| { | |
| "epoch": 5.78, | |
| "learning_rate": 0.0002, | |
| "loss": 0.0134, | |
| "step": 52 | |
| }, | |
| { | |
| "epoch": 5.89, | |
| "learning_rate": 0.0002, | |
| "loss": 0.0075, | |
| "step": 53 | |
| }, | |
| { | |
| "epoch": 6.0, | |
| "learning_rate": 0.0002, | |
| "loss": 0.0129, | |
| "step": 54 | |
| }, | |
| { | |
| "epoch": 6.0, | |
| "eval_loss": 1.4691095352172852, | |
| "eval_runtime": 6.2165, | |
| "eval_samples_per_second": 8.043, | |
| "eval_steps_per_second": 0.322, | |
| "step": 54 | |
| }, | |
| { | |
| "epoch": 6.11, | |
| "learning_rate": 0.0002, | |
| "loss": 0.005, | |
| "step": 55 | |
| }, | |
| { | |
| "epoch": 6.22, | |
| "learning_rate": 0.0002, | |
| "loss": 0.0072, | |
| "step": 56 | |
| }, | |
| { | |
| "epoch": 6.33, | |
| "learning_rate": 0.0002, | |
| "loss": 0.0103, | |
| "step": 57 | |
| }, | |
| { | |
| "epoch": 6.44, | |
| "learning_rate": 0.0002, | |
| "loss": 0.0076, | |
| "step": 58 | |
| }, | |
| { | |
| "epoch": 6.56, | |
| "learning_rate": 0.0002, | |
| "loss": 0.0119, | |
| "step": 59 | |
| }, | |
| { | |
| "epoch": 6.67, | |
| "learning_rate": 0.0002, | |
| "loss": 0.0065, | |
| "step": 60 | |
| }, | |
| { | |
| "epoch": 6.78, | |
| "learning_rate": 0.0002, | |
| "loss": 0.0115, | |
| "step": 61 | |
| }, | |
| { | |
| "epoch": 6.89, | |
| "learning_rate": 0.0002, | |
| "loss": 0.0086, | |
| "step": 62 | |
| }, | |
| { | |
| "epoch": 7.0, | |
| "learning_rate": 0.0002, | |
| "loss": 0.0062, | |
| "step": 63 | |
| }, | |
| { | |
| "epoch": 7.0, | |
| "eval_loss": 1.4760642051696777, | |
| "eval_runtime": 6.1472, | |
| "eval_samples_per_second": 8.134, | |
| "eval_steps_per_second": 0.325, | |
| "step": 63 | |
| }, | |
| { | |
| "epoch": 7.11, | |
| "learning_rate": 0.0002, | |
| "loss": 0.0035, | |
| "step": 64 | |
| }, | |
| { | |
| "epoch": 7.22, | |
| "learning_rate": 0.0002, | |
| "loss": 0.0025, | |
| "step": 65 | |
| }, | |
| { | |
| "epoch": 7.33, | |
| "learning_rate": 0.0002, | |
| "loss": 0.0053, | |
| "step": 66 | |
| }, | |
| { | |
| "epoch": 7.44, | |
| "learning_rate": 0.0002, | |
| "loss": 0.0114, | |
| "step": 67 | |
| }, | |
| { | |
| "epoch": 7.56, | |
| "learning_rate": 0.0002, | |
| "loss": 0.0067, | |
| "step": 68 | |
| }, | |
| { | |
| "epoch": 7.67, | |
| "learning_rate": 0.0002, | |
| "loss": 0.007, | |
| "step": 69 | |
| }, | |
| { | |
| "epoch": 7.78, | |
| "learning_rate": 0.0002, | |
| "loss": 0.0096, | |
| "step": 70 | |
| }, | |
| { | |
| "epoch": 7.89, | |
| "learning_rate": 0.0002, | |
| "loss": 0.0066, | |
| "step": 71 | |
| }, | |
| { | |
| "epoch": 8.0, | |
| "learning_rate": 0.0002, | |
| "loss": 0.0026, | |
| "step": 72 | |
| }, | |
| { | |
| "epoch": 8.0, | |
| "eval_loss": 1.4730503559112549, | |
| "eval_runtime": 6.2469, | |
| "eval_samples_per_second": 8.004, | |
| "eval_steps_per_second": 0.32, | |
| "step": 72 | |
| }, | |
| { | |
| "epoch": 8.11, | |
| "learning_rate": 0.0002, | |
| "loss": 0.0077, | |
| "step": 73 | |
| }, | |
| { | |
| "epoch": 8.22, | |
| "learning_rate": 0.0002, | |
| "loss": 0.0019, | |
| "step": 74 | |
| }, | |
| { | |
| "epoch": 8.33, | |
| "learning_rate": 0.0002, | |
| "loss": 0.0033, | |
| "step": 75 | |
| }, | |
| { | |
| "epoch": 8.44, | |
| "learning_rate": 0.0002, | |
| "loss": 0.0038, | |
| "step": 76 | |
| }, | |
| { | |
| "epoch": 8.56, | |
| "learning_rate": 0.0002, | |
| "loss": 0.0034, | |
| "step": 77 | |
| }, | |
| { | |
| "epoch": 8.67, | |
| "learning_rate": 0.0002, | |
| "loss": 0.0021, | |
| "step": 78 | |
| }, | |
| { | |
| "epoch": 8.78, | |
| "learning_rate": 0.0002, | |
| "loss": 0.0054, | |
| "step": 79 | |
| }, | |
| { | |
| "epoch": 8.89, | |
| "learning_rate": 0.0002, | |
| "loss": 0.0067, | |
| "step": 80 | |
| }, | |
| { | |
| "epoch": 9.0, | |
| "learning_rate": 0.0002, | |
| "loss": 0.0018, | |
| "step": 81 | |
| }, | |
| { | |
| "epoch": 9.0, | |
| "eval_loss": 1.6150561571121216, | |
| "eval_runtime": 6.1573, | |
| "eval_samples_per_second": 8.12, | |
| "eval_steps_per_second": 0.325, | |
| "step": 81 | |
| }, | |
| { | |
| "epoch": 9.11, | |
| "learning_rate": 0.0002, | |
| "loss": 0.0011, | |
| "step": 82 | |
| }, | |
| { | |
| "epoch": 9.22, | |
| "learning_rate": 0.0002, | |
| "loss": 0.0018, | |
| "step": 83 | |
| }, | |
| { | |
| "epoch": 9.33, | |
| "learning_rate": 0.0002, | |
| "loss": 0.001, | |
| "step": 84 | |
| }, | |
| { | |
| "epoch": 9.44, | |
| "learning_rate": 0.0002, | |
| "loss": 0.0029, | |
| "step": 85 | |
| }, | |
| { | |
| "epoch": 9.56, | |
| "learning_rate": 0.0002, | |
| "loss": 0.0024, | |
| "step": 86 | |
| }, | |
| { | |
| "epoch": 9.67, | |
| "learning_rate": 0.0002, | |
| "loss": 0.0017, | |
| "step": 87 | |
| }, | |
| { | |
| "epoch": 9.78, | |
| "learning_rate": 0.0002, | |
| "loss": 0.0009, | |
| "step": 88 | |
| }, | |
| { | |
| "epoch": 9.89, | |
| "learning_rate": 0.0002, | |
| "loss": 0.0075, | |
| "step": 89 | |
| }, | |
| { | |
| "epoch": 10.0, | |
| "learning_rate": 0.0002, | |
| "loss": 0.0012, | |
| "step": 90 | |
| }, | |
| { | |
| "epoch": 10.0, | |
| "eval_loss": 1.6820263862609863, | |
| "eval_runtime": 6.2532, | |
| "eval_samples_per_second": 7.996, | |
| "eval_steps_per_second": 0.32, | |
| "step": 90 | |
| }, | |
| { | |
| "epoch": 10.0, | |
| "step": 90, | |
| "total_flos": 1.6878235281679974e+17, | |
| "train_loss": 0.1837897605379112, | |
| "train_runtime": 624.5113, | |
| "train_samples_per_second": 4.532, | |
| "train_steps_per_second": 0.144 | |
| } | |
| ], | |
| "logging_steps": 1.0, | |
| "max_steps": 90, | |
| "num_input_tokens_seen": 0, | |
| "num_train_epochs": 10, | |
| "save_steps": 50000, | |
| "total_flos": 1.6878235281679974e+17, | |
| "train_batch_size": 4, | |
| "trial_name": null, | |
| "trial_params": null | |
| } | |