| { | |
| "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": 1.3352, | |
| "step": 2 | |
| }, | |
| { | |
| "epoch": 0.33, | |
| "learning_rate": 0.0002, | |
| "loss": 1.4468, | |
| "step": 3 | |
| }, | |
| { | |
| "epoch": 0.44, | |
| "learning_rate": 0.0002, | |
| "loss": 1.2444, | |
| "step": 4 | |
| }, | |
| { | |
| "epoch": 0.56, | |
| "learning_rate": 0.0002, | |
| "loss": 1.1357, | |
| "step": 5 | |
| }, | |
| { | |
| "epoch": 0.67, | |
| "learning_rate": 0.0002, | |
| "loss": 1.0412, | |
| "step": 6 | |
| }, | |
| { | |
| "epoch": 0.78, | |
| "learning_rate": 0.0002, | |
| "loss": 0.9966, | |
| "step": 7 | |
| }, | |
| { | |
| "epoch": 0.89, | |
| "learning_rate": 0.0002, | |
| "loss": 0.8682, | |
| "step": 8 | |
| }, | |
| { | |
| "epoch": 1.0, | |
| "learning_rate": 0.0002, | |
| "loss": 0.9019, | |
| "step": 9 | |
| }, | |
| { | |
| "epoch": 1.0, | |
| "eval_loss": 1.0177268981933594, | |
| "eval_runtime": 11.4956, | |
| "eval_samples_per_second": 4.349, | |
| "eval_steps_per_second": 0.174, | |
| "step": 9 | |
| }, | |
| { | |
| "epoch": 1.11, | |
| "learning_rate": 0.0002, | |
| "loss": 0.6961, | |
| "step": 10 | |
| }, | |
| { | |
| "epoch": 1.22, | |
| "learning_rate": 0.0002, | |
| "loss": 0.5139, | |
| "step": 11 | |
| }, | |
| { | |
| "epoch": 1.33, | |
| "learning_rate": 0.0002, | |
| "loss": 0.4883, | |
| "step": 12 | |
| }, | |
| { | |
| "epoch": 1.44, | |
| "learning_rate": 0.0002, | |
| "loss": 0.5516, | |
| "step": 13 | |
| }, | |
| { | |
| "epoch": 1.56, | |
| "learning_rate": 0.0002, | |
| "loss": 0.4948, | |
| "step": 14 | |
| }, | |
| { | |
| "epoch": 1.67, | |
| "learning_rate": 0.0002, | |
| "loss": 0.4729, | |
| "step": 15 | |
| }, | |
| { | |
| "epoch": 1.78, | |
| "learning_rate": 0.0002, | |
| "loss": 0.3157, | |
| "step": 16 | |
| }, | |
| { | |
| "epoch": 1.89, | |
| "learning_rate": 0.0002, | |
| "loss": 0.2823, | |
| "step": 17 | |
| }, | |
| { | |
| "epoch": 2.0, | |
| "learning_rate": 0.0002, | |
| "loss": 0.2356, | |
| "step": 18 | |
| }, | |
| { | |
| "epoch": 2.0, | |
| "eval_loss": 1.1176843643188477, | |
| "eval_runtime": 6.3692, | |
| "eval_samples_per_second": 7.85, | |
| "eval_steps_per_second": 0.314, | |
| "step": 18 | |
| }, | |
| { | |
| "epoch": 2.11, | |
| "learning_rate": 0.0002, | |
| "loss": 0.1731, | |
| "step": 19 | |
| }, | |
| { | |
| "epoch": 2.22, | |
| "learning_rate": 0.0002, | |
| "loss": 0.1148, | |
| "step": 20 | |
| }, | |
| { | |
| "epoch": 2.33, | |
| "learning_rate": 0.0002, | |
| "loss": 0.1148, | |
| "step": 21 | |
| }, | |
| { | |
| "epoch": 2.44, | |
| "learning_rate": 0.0002, | |
| "loss": 0.133, | |
| "step": 22 | |
| }, | |
| { | |
| "epoch": 2.56, | |
| "learning_rate": 0.0002, | |
| "loss": 0.0924, | |
| "step": 23 | |
| }, | |
| { | |
| "epoch": 2.67, | |
| "learning_rate": 0.0002, | |
| "loss": 0.0582, | |
| "step": 24 | |
| }, | |
| { | |
| "epoch": 2.78, | |
| "learning_rate": 0.0002, | |
| "loss": 0.0998, | |
| "step": 25 | |
| }, | |
| { | |
| "epoch": 2.89, | |
| "learning_rate": 0.0002, | |
| "loss": 0.099, | |
| "step": 26 | |
| }, | |
| { | |
| "epoch": 3.0, | |
| "learning_rate": 0.0002, | |
| "loss": 0.0761, | |
| "step": 27 | |
| }, | |
| { | |
| "epoch": 3.0, | |
| "eval_loss": 1.1807804107666016, | |
| "eval_runtime": 6.295, | |
| "eval_samples_per_second": 7.943, | |
| "eval_steps_per_second": 0.318, | |
| "step": 27 | |
| }, | |
| { | |
| "epoch": 3.11, | |
| "learning_rate": 0.0002, | |
| "loss": 0.0424, | |
| "step": 28 | |
| }, | |
| { | |
| "epoch": 3.22, | |
| "learning_rate": 0.0002, | |
| "loss": 0.0367, | |
| "step": 29 | |
| }, | |
| { | |
| "epoch": 3.33, | |
| "learning_rate": 0.0002, | |
| "loss": 0.0423, | |
| "step": 30 | |
| }, | |
| { | |
| "epoch": 3.44, | |
| "learning_rate": 0.0002, | |
| "loss": 0.0236, | |
| "step": 31 | |
| }, | |
| { | |
| "epoch": 3.56, | |
| "learning_rate": 0.0002, | |
| "loss": 0.0347, | |
| "step": 32 | |
| }, | |
| { | |
| "epoch": 3.67, | |
| "learning_rate": 0.0002, | |
| "loss": 0.0352, | |
| "step": 33 | |
| }, | |
| { | |
| "epoch": 3.78, | |
| "learning_rate": 0.0002, | |
| "loss": 0.0333, | |
| "step": 34 | |
| }, | |
| { | |
| "epoch": 3.89, | |
| "learning_rate": 0.0002, | |
| "loss": 0.0491, | |
| "step": 35 | |
| }, | |
| { | |
| "epoch": 4.0, | |
| "learning_rate": 0.0002, | |
| "loss": 0.0304, | |
| "step": 36 | |
| }, | |
| { | |
| "epoch": 4.0, | |
| "eval_loss": 1.3986564874649048, | |
| "eval_runtime": 6.3879, | |
| "eval_samples_per_second": 7.827, | |
| "eval_steps_per_second": 0.313, | |
| "step": 36 | |
| }, | |
| { | |
| "epoch": 4.11, | |
| "learning_rate": 0.0002, | |
| "loss": 0.0099, | |
| "step": 37 | |
| }, | |
| { | |
| "epoch": 4.22, | |
| "learning_rate": 0.0002, | |
| "loss": 0.016, | |
| "step": 38 | |
| }, | |
| { | |
| "epoch": 4.33, | |
| "learning_rate": 0.0002, | |
| "loss": 0.0133, | |
| "step": 39 | |
| }, | |
| { | |
| "epoch": 4.44, | |
| "learning_rate": 0.0002, | |
| "loss": 0.0128, | |
| "step": 40 | |
| }, | |
| { | |
| "epoch": 4.56, | |
| "learning_rate": 0.0002, | |
| "loss": 0.0101, | |
| "step": 41 | |
| }, | |
| { | |
| "epoch": 4.67, | |
| "learning_rate": 0.0002, | |
| "loss": 0.0204, | |
| "step": 42 | |
| }, | |
| { | |
| "epoch": 4.78, | |
| "learning_rate": 0.0002, | |
| "loss": 0.0263, | |
| "step": 43 | |
| }, | |
| { | |
| "epoch": 4.89, | |
| "learning_rate": 0.0002, | |
| "loss": 0.0111, | |
| "step": 44 | |
| }, | |
| { | |
| "epoch": 5.0, | |
| "learning_rate": 0.0002, | |
| "loss": 0.0092, | |
| "step": 45 | |
| }, | |
| { | |
| "epoch": 5.0, | |
| "eval_loss": 1.4470524787902832, | |
| "eval_runtime": 6.2792, | |
| "eval_samples_per_second": 7.963, | |
| "eval_steps_per_second": 0.319, | |
| "step": 45 | |
| }, | |
| { | |
| "epoch": 5.11, | |
| "learning_rate": 0.0002, | |
| "loss": 0.0043, | |
| "step": 46 | |
| }, | |
| { | |
| "epoch": 5.22, | |
| "learning_rate": 0.0002, | |
| "loss": 0.0111, | |
| "step": 47 | |
| }, | |
| { | |
| "epoch": 5.33, | |
| "learning_rate": 0.0002, | |
| "loss": 0.0084, | |
| "step": 48 | |
| }, | |
| { | |
| "epoch": 5.44, | |
| "learning_rate": 0.0002, | |
| "loss": 0.0075, | |
| "step": 49 | |
| }, | |
| { | |
| "epoch": 5.56, | |
| "learning_rate": 0.0002, | |
| "loss": 0.0048, | |
| "step": 50 | |
| }, | |
| { | |
| "epoch": 5.67, | |
| "learning_rate": 0.0002, | |
| "loss": 0.0125, | |
| "step": 51 | |
| }, | |
| { | |
| "epoch": 5.78, | |
| "learning_rate": 0.0002, | |
| "loss": 0.0069, | |
| "step": 52 | |
| }, | |
| { | |
| "epoch": 5.89, | |
| "learning_rate": 0.0002, | |
| "loss": 0.0058, | |
| "step": 53 | |
| }, | |
| { | |
| "epoch": 6.0, | |
| "learning_rate": 0.0002, | |
| "loss": 0.009, | |
| "step": 54 | |
| }, | |
| { | |
| "epoch": 6.0, | |
| "eval_loss": 1.4909963607788086, | |
| "eval_runtime": 6.398, | |
| "eval_samples_per_second": 7.815, | |
| "eval_steps_per_second": 0.313, | |
| "step": 54 | |
| }, | |
| { | |
| "epoch": 6.11, | |
| "learning_rate": 0.0002, | |
| "loss": 0.0041, | |
| "step": 55 | |
| }, | |
| { | |
| "epoch": 6.22, | |
| "learning_rate": 0.0002, | |
| "loss": 0.0048, | |
| "step": 56 | |
| }, | |
| { | |
| "epoch": 6.33, | |
| "learning_rate": 0.0002, | |
| "loss": 0.0053, | |
| "step": 57 | |
| }, | |
| { | |
| "epoch": 6.44, | |
| "learning_rate": 0.0002, | |
| "loss": 0.0045, | |
| "step": 58 | |
| }, | |
| { | |
| "epoch": 6.56, | |
| "learning_rate": 0.0002, | |
| "loss": 0.0109, | |
| "step": 59 | |
| }, | |
| { | |
| "epoch": 6.67, | |
| "learning_rate": 0.0002, | |
| "loss": 0.0064, | |
| "step": 60 | |
| }, | |
| { | |
| "epoch": 6.78, | |
| "learning_rate": 0.0002, | |
| "loss": 0.0056, | |
| "step": 61 | |
| }, | |
| { | |
| "epoch": 6.89, | |
| "learning_rate": 0.0002, | |
| "loss": 0.0038, | |
| "step": 62 | |
| }, | |
| { | |
| "epoch": 7.0, | |
| "learning_rate": 0.0002, | |
| "loss": 0.0055, | |
| "step": 63 | |
| }, | |
| { | |
| "epoch": 7.0, | |
| "eval_loss": 1.5328795909881592, | |
| "eval_runtime": 6.3671, | |
| "eval_samples_per_second": 7.853, | |
| "eval_steps_per_second": 0.314, | |
| "step": 63 | |
| }, | |
| { | |
| "epoch": 7.11, | |
| "learning_rate": 0.0002, | |
| "loss": 0.0022, | |
| "step": 64 | |
| }, | |
| { | |
| "epoch": 7.22, | |
| "learning_rate": 0.0002, | |
| "loss": 0.0022, | |
| "step": 65 | |
| }, | |
| { | |
| "epoch": 7.33, | |
| "learning_rate": 0.0002, | |
| "loss": 0.0089, | |
| "step": 66 | |
| }, | |
| { | |
| "epoch": 7.44, | |
| "learning_rate": 0.0002, | |
| "loss": 0.0057, | |
| "step": 67 | |
| }, | |
| { | |
| "epoch": 7.56, | |
| "learning_rate": 0.0002, | |
| "loss": 0.0073, | |
| "step": 68 | |
| }, | |
| { | |
| "epoch": 7.67, | |
| "learning_rate": 0.0002, | |
| "loss": 0.0038, | |
| "step": 69 | |
| }, | |
| { | |
| "epoch": 7.78, | |
| "learning_rate": 0.0002, | |
| "loss": 0.0137, | |
| "step": 70 | |
| }, | |
| { | |
| "epoch": 7.89, | |
| "learning_rate": 0.0002, | |
| "loss": 0.0055, | |
| "step": 71 | |
| }, | |
| { | |
| "epoch": 8.0, | |
| "learning_rate": 0.0002, | |
| "loss": 0.0022, | |
| "step": 72 | |
| }, | |
| { | |
| "epoch": 8.0, | |
| "eval_loss": 1.534345269203186, | |
| "eval_runtime": 6.3998, | |
| "eval_samples_per_second": 7.813, | |
| "eval_steps_per_second": 0.313, | |
| "step": 72 | |
| }, | |
| { | |
| "epoch": 8.11, | |
| "learning_rate": 0.0002, | |
| "loss": 0.0016, | |
| "step": 73 | |
| }, | |
| { | |
| "epoch": 8.22, | |
| "learning_rate": 0.0002, | |
| "loss": 0.0041, | |
| "step": 74 | |
| }, | |
| { | |
| "epoch": 8.33, | |
| "learning_rate": 0.0002, | |
| "loss": 0.0069, | |
| "step": 75 | |
| }, | |
| { | |
| "epoch": 8.44, | |
| "learning_rate": 0.0002, | |
| "loss": 0.0036, | |
| "step": 76 | |
| }, | |
| { | |
| "epoch": 8.56, | |
| "learning_rate": 0.0002, | |
| "loss": 0.0043, | |
| "step": 77 | |
| }, | |
| { | |
| "epoch": 8.67, | |
| "learning_rate": 0.0002, | |
| "loss": 0.0019, | |
| "step": 78 | |
| }, | |
| { | |
| "epoch": 8.78, | |
| "learning_rate": 0.0002, | |
| "loss": 0.0027, | |
| "step": 79 | |
| }, | |
| { | |
| "epoch": 8.89, | |
| "learning_rate": 0.0002, | |
| "loss": 0.0081, | |
| "step": 80 | |
| }, | |
| { | |
| "epoch": 9.0, | |
| "learning_rate": 0.0002, | |
| "loss": 0.002, | |
| "step": 81 | |
| }, | |
| { | |
| "epoch": 9.0, | |
| "eval_loss": 1.5295281410217285, | |
| "eval_runtime": 6.3871, | |
| "eval_samples_per_second": 7.828, | |
| "eval_steps_per_second": 0.313, | |
| "step": 81 | |
| }, | |
| { | |
| "epoch": 9.11, | |
| "learning_rate": 0.0002, | |
| "loss": 0.003, | |
| "step": 82 | |
| }, | |
| { | |
| "epoch": 9.22, | |
| "learning_rate": 0.0002, | |
| "loss": 0.0029, | |
| "step": 83 | |
| }, | |
| { | |
| "epoch": 9.33, | |
| "learning_rate": 0.0002, | |
| "loss": 0.0024, | |
| "step": 84 | |
| }, | |
| { | |
| "epoch": 9.44, | |
| "learning_rate": 0.0002, | |
| "loss": 0.002, | |
| "step": 85 | |
| }, | |
| { | |
| "epoch": 9.56, | |
| "learning_rate": 0.0002, | |
| "loss": 0.0028, | |
| "step": 86 | |
| }, | |
| { | |
| "epoch": 9.67, | |
| "learning_rate": 0.0002, | |
| "loss": 0.0036, | |
| "step": 87 | |
| }, | |
| { | |
| "epoch": 9.78, | |
| "learning_rate": 0.0002, | |
| "loss": 0.0021, | |
| "step": 88 | |
| }, | |
| { | |
| "epoch": 9.89, | |
| "learning_rate": 0.0002, | |
| "loss": 0.0033, | |
| "step": 89 | |
| }, | |
| { | |
| "epoch": 10.0, | |
| "learning_rate": 0.0002, | |
| "loss": 0.0011, | |
| "step": 90 | |
| }, | |
| { | |
| "epoch": 10.0, | |
| "eval_loss": 1.6014364957809448, | |
| "eval_runtime": 6.3819, | |
| "eval_samples_per_second": 7.835, | |
| "eval_steps_per_second": 0.313, | |
| "step": 90 | |
| }, | |
| { | |
| "epoch": 10.0, | |
| "step": 90, | |
| "total_flos": 1.6878235281679974e+17, | |
| "train_loss": 0.18350568334701367, | |
| "train_runtime": 641.931, | |
| "train_samples_per_second": 4.409, | |
| "train_steps_per_second": 0.14 | |
| } | |
| ], | |
| "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 | |
| } | |