| { | |
| "best_metric": 1.2429375648498535, | |
| "best_model_checkpoint": "./outputs/checkpoint-4100", | |
| "epoch": 2.9879781420765026, | |
| "eval_steps": 100, | |
| "global_step": 4100, | |
| "is_hyper_param_search": false, | |
| "is_local_process_zero": true, | |
| "is_world_process_zero": true, | |
| "log_history": [ | |
| { | |
| "epoch": 0.07, | |
| "learning_rate": 0.0002, | |
| "loss": 2.1831, | |
| "step": 100 | |
| }, | |
| { | |
| "epoch": 0.07, | |
| "eval_loss": 2.0129995346069336, | |
| "eval_runtime": 143.7126, | |
| "eval_samples_per_second": 43.657, | |
| "eval_steps_per_second": 5.462, | |
| "step": 100 | |
| }, | |
| { | |
| "epoch": 0.15, | |
| "learning_rate": 0.0002, | |
| "loss": 1.9629, | |
| "step": 200 | |
| }, | |
| { | |
| "epoch": 0.15, | |
| "eval_loss": 1.9196887016296387, | |
| "eval_runtime": 143.7285, | |
| "eval_samples_per_second": 43.652, | |
| "eval_steps_per_second": 5.462, | |
| "step": 200 | |
| }, | |
| { | |
| "epoch": 0.22, | |
| "learning_rate": 0.0002, | |
| "loss": 1.8881, | |
| "step": 300 | |
| }, | |
| { | |
| "epoch": 0.22, | |
| "eval_loss": 1.8589884042739868, | |
| "eval_runtime": 143.706, | |
| "eval_samples_per_second": 43.659, | |
| "eval_steps_per_second": 5.463, | |
| "step": 300 | |
| }, | |
| { | |
| "epoch": 0.29, | |
| "learning_rate": 0.0002, | |
| "loss": 1.8368, | |
| "step": 400 | |
| }, | |
| { | |
| "epoch": 0.29, | |
| "eval_loss": 1.813078761100769, | |
| "eval_runtime": 143.7443, | |
| "eval_samples_per_second": 43.647, | |
| "eval_steps_per_second": 5.461, | |
| "step": 400 | |
| }, | |
| { | |
| "epoch": 0.36, | |
| "learning_rate": 0.0002, | |
| "loss": 1.7849, | |
| "step": 500 | |
| }, | |
| { | |
| "epoch": 0.36, | |
| "eval_loss": 1.7718570232391357, | |
| "eval_runtime": 143.7117, | |
| "eval_samples_per_second": 43.657, | |
| "eval_steps_per_second": 5.462, | |
| "step": 500 | |
| }, | |
| { | |
| "epoch": 0.44, | |
| "learning_rate": 0.0002, | |
| "loss": 1.7567, | |
| "step": 600 | |
| }, | |
| { | |
| "epoch": 0.44, | |
| "eval_loss": 1.742559790611267, | |
| "eval_runtime": 143.6984, | |
| "eval_samples_per_second": 43.661, | |
| "eval_steps_per_second": 5.463, | |
| "step": 600 | |
| }, | |
| { | |
| "epoch": 0.51, | |
| "learning_rate": 0.0002, | |
| "loss": 1.718, | |
| "step": 700 | |
| }, | |
| { | |
| "epoch": 0.51, | |
| "eval_loss": 1.7121007442474365, | |
| "eval_runtime": 143.6682, | |
| "eval_samples_per_second": 43.67, | |
| "eval_steps_per_second": 5.464, | |
| "step": 700 | |
| }, | |
| { | |
| "epoch": 0.58, | |
| "learning_rate": 0.0002, | |
| "loss": 1.6976, | |
| "step": 800 | |
| }, | |
| { | |
| "epoch": 0.58, | |
| "eval_loss": 1.6834267377853394, | |
| "eval_runtime": 143.6959, | |
| "eval_samples_per_second": 43.662, | |
| "eval_steps_per_second": 5.463, | |
| "step": 800 | |
| }, | |
| { | |
| "epoch": 0.66, | |
| "learning_rate": 0.0002, | |
| "loss": 1.6624, | |
| "step": 900 | |
| }, | |
| { | |
| "epoch": 0.66, | |
| "eval_loss": 1.6576156616210938, | |
| "eval_runtime": 143.6985, | |
| "eval_samples_per_second": 43.661, | |
| "eval_steps_per_second": 5.463, | |
| "step": 900 | |
| }, | |
| { | |
| "epoch": 0.73, | |
| "learning_rate": 0.0002, | |
| "loss": 1.6475, | |
| "step": 1000 | |
| }, | |
| { | |
| "epoch": 0.73, | |
| "eval_loss": 1.6344480514526367, | |
| "eval_runtime": 143.8656, | |
| "eval_samples_per_second": 43.61, | |
| "eval_steps_per_second": 5.456, | |
| "step": 1000 | |
| }, | |
| { | |
| "epoch": 0.8, | |
| "learning_rate": 0.0002, | |
| "loss": 1.6277, | |
| "step": 1100 | |
| }, | |
| { | |
| "epoch": 0.8, | |
| "eval_loss": 1.613603115081787, | |
| "eval_runtime": 143.8266, | |
| "eval_samples_per_second": 43.622, | |
| "eval_steps_per_second": 5.458, | |
| "step": 1100 | |
| }, | |
| { | |
| "epoch": 0.87, | |
| "learning_rate": 0.0002, | |
| "loss": 1.5919, | |
| "step": 1200 | |
| }, | |
| { | |
| "epoch": 0.87, | |
| "eval_loss": 1.5947637557983398, | |
| "eval_runtime": 143.7429, | |
| "eval_samples_per_second": 43.647, | |
| "eval_steps_per_second": 5.461, | |
| "step": 1200 | |
| }, | |
| { | |
| "epoch": 0.95, | |
| "learning_rate": 0.0002, | |
| "loss": 1.5999, | |
| "step": 1300 | |
| }, | |
| { | |
| "epoch": 0.95, | |
| "eval_loss": 1.5728731155395508, | |
| "eval_runtime": 143.7499, | |
| "eval_samples_per_second": 43.645, | |
| "eval_steps_per_second": 5.461, | |
| "step": 1300 | |
| }, | |
| { | |
| "epoch": 1.02, | |
| "learning_rate": 0.0002, | |
| "loss": 1.5528, | |
| "step": 1400 | |
| }, | |
| { | |
| "epoch": 1.02, | |
| "eval_loss": 1.55287504196167, | |
| "eval_runtime": 143.8673, | |
| "eval_samples_per_second": 43.61, | |
| "eval_steps_per_second": 5.456, | |
| "step": 1400 | |
| }, | |
| { | |
| "epoch": 1.09, | |
| "learning_rate": 0.0002, | |
| "loss": 1.525, | |
| "step": 1500 | |
| }, | |
| { | |
| "epoch": 1.09, | |
| "eval_loss": 1.5351721048355103, | |
| "eval_runtime": 143.7854, | |
| "eval_samples_per_second": 43.634, | |
| "eval_steps_per_second": 5.46, | |
| "step": 1500 | |
| }, | |
| { | |
| "epoch": 1.17, | |
| "learning_rate": 0.0002, | |
| "loss": 1.5063, | |
| "step": 1600 | |
| }, | |
| { | |
| "epoch": 1.17, | |
| "eval_loss": 1.5172873735427856, | |
| "eval_runtime": 143.6475, | |
| "eval_samples_per_second": 43.676, | |
| "eval_steps_per_second": 5.465, | |
| "step": 1600 | |
| }, | |
| { | |
| "epoch": 1.24, | |
| "learning_rate": 0.0002, | |
| "loss": 1.5042, | |
| "step": 1700 | |
| }, | |
| { | |
| "epoch": 1.24, | |
| "eval_loss": 1.501122236251831, | |
| "eval_runtime": 145.0647, | |
| "eval_samples_per_second": 43.25, | |
| "eval_steps_per_second": 5.411, | |
| "step": 1700 | |
| }, | |
| { | |
| "epoch": 1.31, | |
| "learning_rate": 0.0002, | |
| "loss": 1.4777, | |
| "step": 1800 | |
| }, | |
| { | |
| "epoch": 1.31, | |
| "eval_loss": 1.4863215684890747, | |
| "eval_runtime": 143.8068, | |
| "eval_samples_per_second": 43.628, | |
| "eval_steps_per_second": 5.459, | |
| "step": 1800 | |
| }, | |
| { | |
| "epoch": 1.38, | |
| "learning_rate": 0.0002, | |
| "loss": 1.4677, | |
| "step": 1900 | |
| }, | |
| { | |
| "epoch": 1.38, | |
| "eval_loss": 1.4718947410583496, | |
| "eval_runtime": 143.7145, | |
| "eval_samples_per_second": 43.656, | |
| "eval_steps_per_second": 5.462, | |
| "step": 1900 | |
| }, | |
| { | |
| "epoch": 1.46, | |
| "learning_rate": 0.0002, | |
| "loss": 1.4515, | |
| "step": 2000 | |
| }, | |
| { | |
| "epoch": 1.46, | |
| "eval_loss": 1.4579730033874512, | |
| "eval_runtime": 143.8575, | |
| "eval_samples_per_second": 43.613, | |
| "eval_steps_per_second": 5.457, | |
| "step": 2000 | |
| }, | |
| { | |
| "epoch": 1.53, | |
| "learning_rate": 0.0002, | |
| "loss": 1.437, | |
| "step": 2100 | |
| }, | |
| { | |
| "epoch": 1.53, | |
| "eval_loss": 1.4419173002243042, | |
| "eval_runtime": 143.8022, | |
| "eval_samples_per_second": 43.629, | |
| "eval_steps_per_second": 5.459, | |
| "step": 2100 | |
| }, | |
| { | |
| "epoch": 1.6, | |
| "learning_rate": 0.0002, | |
| "loss": 1.4105, | |
| "step": 2200 | |
| }, | |
| { | |
| "epoch": 1.6, | |
| "eval_loss": 1.429054617881775, | |
| "eval_runtime": 143.9668, | |
| "eval_samples_per_second": 43.579, | |
| "eval_steps_per_second": 5.453, | |
| "step": 2200 | |
| }, | |
| { | |
| "epoch": 1.68, | |
| "learning_rate": 0.0002, | |
| "loss": 1.4103, | |
| "step": 2300 | |
| }, | |
| { | |
| "epoch": 1.68, | |
| "eval_loss": 1.4158506393432617, | |
| "eval_runtime": 143.8198, | |
| "eval_samples_per_second": 43.624, | |
| "eval_steps_per_second": 5.458, | |
| "step": 2300 | |
| }, | |
| { | |
| "epoch": 1.75, | |
| "learning_rate": 0.0002, | |
| "loss": 1.4092, | |
| "step": 2400 | |
| }, | |
| { | |
| "epoch": 1.75, | |
| "eval_loss": 1.4019700288772583, | |
| "eval_runtime": 143.831, | |
| "eval_samples_per_second": 43.621, | |
| "eval_steps_per_second": 5.458, | |
| "step": 2400 | |
| }, | |
| { | |
| "epoch": 1.82, | |
| "learning_rate": 0.0002, | |
| "loss": 1.3815, | |
| "step": 2500 | |
| }, | |
| { | |
| "epoch": 1.82, | |
| "eval_loss": 1.3896080255508423, | |
| "eval_runtime": 143.8457, | |
| "eval_samples_per_second": 43.616, | |
| "eval_steps_per_second": 5.457, | |
| "step": 2500 | |
| }, | |
| { | |
| "epoch": 1.89, | |
| "learning_rate": 0.0002, | |
| "loss": 1.3709, | |
| "step": 2600 | |
| }, | |
| { | |
| "epoch": 1.89, | |
| "eval_loss": 1.3768706321716309, | |
| "eval_runtime": 143.8343, | |
| "eval_samples_per_second": 43.62, | |
| "eval_steps_per_second": 5.458, | |
| "step": 2600 | |
| }, | |
| { | |
| "epoch": 1.97, | |
| "learning_rate": 0.0002, | |
| "loss": 1.3603, | |
| "step": 2700 | |
| }, | |
| { | |
| "epoch": 1.97, | |
| "eval_loss": 1.3619952201843262, | |
| "eval_runtime": 143.7532, | |
| "eval_samples_per_second": 43.644, | |
| "eval_steps_per_second": 5.461, | |
| "step": 2700 | |
| }, | |
| { | |
| "epoch": 2.04, | |
| "learning_rate": 0.0002, | |
| "loss": 1.331, | |
| "step": 2800 | |
| }, | |
| { | |
| "epoch": 2.04, | |
| "eval_loss": 1.3511581420898438, | |
| "eval_runtime": 143.762, | |
| "eval_samples_per_second": 43.642, | |
| "eval_steps_per_second": 5.46, | |
| "step": 2800 | |
| }, | |
| { | |
| "epoch": 2.11, | |
| "learning_rate": 0.0002, | |
| "loss": 1.3051, | |
| "step": 2900 | |
| }, | |
| { | |
| "epoch": 2.11, | |
| "eval_loss": 1.3384917974472046, | |
| "eval_runtime": 143.684, | |
| "eval_samples_per_second": 43.665, | |
| "eval_steps_per_second": 5.463, | |
| "step": 2900 | |
| }, | |
| { | |
| "epoch": 2.19, | |
| "learning_rate": 0.0002, | |
| "loss": 1.3058, | |
| "step": 3000 | |
| }, | |
| { | |
| "epoch": 2.19, | |
| "eval_loss": 1.327839732170105, | |
| "eval_runtime": 143.7326, | |
| "eval_samples_per_second": 43.65, | |
| "eval_steps_per_second": 5.462, | |
| "step": 3000 | |
| }, | |
| { | |
| "epoch": 2.26, | |
| "learning_rate": 0.0002, | |
| "loss": 1.2863, | |
| "step": 3100 | |
| }, | |
| { | |
| "epoch": 2.26, | |
| "eval_loss": 1.3162676095962524, | |
| "eval_runtime": 143.6371, | |
| "eval_samples_per_second": 43.68, | |
| "eval_steps_per_second": 5.465, | |
| "step": 3100 | |
| }, | |
| { | |
| "epoch": 2.33, | |
| "learning_rate": 0.0002, | |
| "loss": 1.2907, | |
| "step": 3200 | |
| }, | |
| { | |
| "epoch": 2.33, | |
| "eval_loss": 1.3046828508377075, | |
| "eval_runtime": 143.8556, | |
| "eval_samples_per_second": 43.613, | |
| "eval_steps_per_second": 5.457, | |
| "step": 3200 | |
| }, | |
| { | |
| "epoch": 2.4, | |
| "learning_rate": 0.0002, | |
| "loss": 1.2715, | |
| "step": 3300 | |
| }, | |
| { | |
| "epoch": 2.4, | |
| "eval_loss": 1.2920022010803223, | |
| "eval_runtime": 143.7686, | |
| "eval_samples_per_second": 43.64, | |
| "eval_steps_per_second": 5.46, | |
| "step": 3300 | |
| }, | |
| { | |
| "epoch": 2.48, | |
| "learning_rate": 0.0002, | |
| "loss": 1.2533, | |
| "step": 3400 | |
| }, | |
| { | |
| "epoch": 2.48, | |
| "eval_loss": 1.2834123373031616, | |
| "eval_runtime": 143.8354, | |
| "eval_samples_per_second": 43.619, | |
| "eval_steps_per_second": 5.458, | |
| "step": 3400 | |
| }, | |
| { | |
| "epoch": 2.55, | |
| "learning_rate": 0.0002, | |
| "loss": 1.2554, | |
| "step": 3500 | |
| }, | |
| { | |
| "epoch": 2.55, | |
| "eval_loss": 1.2722229957580566, | |
| "eval_runtime": 143.7943, | |
| "eval_samples_per_second": 43.632, | |
| "eval_steps_per_second": 5.459, | |
| "step": 3500 | |
| }, | |
| { | |
| "epoch": 2.62, | |
| "learning_rate": 0.0002, | |
| "loss": 1.3031, | |
| "step": 3600 | |
| }, | |
| { | |
| "epoch": 2.62, | |
| "eval_loss": 1.3205279111862183, | |
| "eval_runtime": 138.0704, | |
| "eval_samples_per_second": 45.441, | |
| "eval_steps_per_second": 5.686, | |
| "step": 3600 | |
| }, | |
| { | |
| "epoch": 2.7, | |
| "learning_rate": 0.0002, | |
| "loss": 1.2798, | |
| "step": 3700 | |
| }, | |
| { | |
| "epoch": 2.7, | |
| "eval_loss": 1.3004039525985718, | |
| "eval_runtime": 134.3049, | |
| "eval_samples_per_second": 46.715, | |
| "eval_steps_per_second": 5.845, | |
| "step": 3700 | |
| }, | |
| { | |
| "epoch": 2.77, | |
| "learning_rate": 0.0002, | |
| "loss": 1.2528, | |
| "step": 3800 | |
| }, | |
| { | |
| "epoch": 2.77, | |
| "eval_loss": 1.2842642068862915, | |
| "eval_runtime": 133.1291, | |
| "eval_samples_per_second": 47.127, | |
| "eval_steps_per_second": 5.897, | |
| "step": 3800 | |
| }, | |
| { | |
| "epoch": 2.84, | |
| "learning_rate": 0.0002, | |
| "loss": 1.2453, | |
| "step": 3900 | |
| }, | |
| { | |
| "epoch": 2.84, | |
| "eval_loss": 1.268664002418518, | |
| "eval_runtime": 133.1158, | |
| "eval_samples_per_second": 47.132, | |
| "eval_steps_per_second": 5.897, | |
| "step": 3900 | |
| }, | |
| { | |
| "epoch": 2.92, | |
| "learning_rate": 0.0002, | |
| "loss": 1.2321, | |
| "step": 4000 | |
| }, | |
| { | |
| "epoch": 2.92, | |
| "eval_loss": 1.2541390657424927, | |
| "eval_runtime": 133.1492, | |
| "eval_samples_per_second": 47.12, | |
| "eval_steps_per_second": 5.896, | |
| "step": 4000 | |
| }, | |
| { | |
| "epoch": 2.99, | |
| "learning_rate": 0.0002, | |
| "loss": 1.2188, | |
| "step": 4100 | |
| }, | |
| { | |
| "epoch": 2.99, | |
| "eval_loss": 1.2429375648498535, | |
| "eval_runtime": 133.6701, | |
| "eval_samples_per_second": 46.936, | |
| "eval_steps_per_second": 5.873, | |
| "step": 4100 | |
| } | |
| ], | |
| "logging_steps": 100, | |
| "max_steps": 4116, | |
| "num_train_epochs": 3, | |
| "save_steps": 100, | |
| "total_flos": 2.4439923911887258e+17, | |
| "trial_name": null, | |
| "trial_params": null | |
| } | |