| { | |
| "best_metric": 0.8720703125, | |
| "best_model_checkpoint": "./alma-7b-parallel-ft-lora-canto/checkpoint-589", | |
| "epoch": 1.99673735725938, | |
| "global_step": 612, | |
| "is_hyper_param_search": false, | |
| "is_local_process_zero": true, | |
| "is_world_process_zero": true, | |
| "log_history": [ | |
| { | |
| "epoch": 0.1, | |
| "learning_rate": 0.0006831300510639731, | |
| "loss": 1.8545, | |
| "step": 31 | |
| }, | |
| { | |
| "epoch": 0.1, | |
| "eval_loss": 1.275390625, | |
| "eval_runtime": 161.389, | |
| "eval_samples_per_second": 26.049, | |
| "eval_steps_per_second": 1.63, | |
| "step": 31 | |
| }, | |
| { | |
| "epoch": 0.2, | |
| "learning_rate": 0.0003944053188733077, | |
| "loss": 1.2078, | |
| "step": 62 | |
| }, | |
| { | |
| "epoch": 0.2, | |
| "eval_loss": 1.1513671875, | |
| "eval_runtime": 161.4937, | |
| "eval_samples_per_second": 26.032, | |
| "eval_steps_per_second": 1.629, | |
| "step": 62 | |
| }, | |
| { | |
| "epoch": 0.3, | |
| "learning_rate": 0.00030348848933344196, | |
| "loss": 1.1077, | |
| "step": 93 | |
| }, | |
| { | |
| "epoch": 0.3, | |
| "eval_loss": 1.0869140625, | |
| "eval_runtime": 161.573, | |
| "eval_samples_per_second": 26.019, | |
| "eval_steps_per_second": 1.628, | |
| "step": 93 | |
| }, | |
| { | |
| "epoch": 0.4, | |
| "learning_rate": 0.00025577443333465435, | |
| "loss": 1.067, | |
| "step": 124 | |
| }, | |
| { | |
| "epoch": 0.4, | |
| "eval_loss": 1.0439453125, | |
| "eval_runtime": 161.4591, | |
| "eval_samples_per_second": 26.038, | |
| "eval_steps_per_second": 1.629, | |
| "step": 124 | |
| }, | |
| { | |
| "epoch": 0.51, | |
| "learning_rate": 0.0002252213082307254, | |
| "loss": 1.0179, | |
| "step": 155 | |
| }, | |
| { | |
| "epoch": 0.51, | |
| "eval_loss": 1.015625, | |
| "eval_runtime": 161.6031, | |
| "eval_samples_per_second": 26.014, | |
| "eval_steps_per_second": 1.627, | |
| "step": 155 | |
| }, | |
| { | |
| "epoch": 0.61, | |
| "learning_rate": 0.00020351933162035313, | |
| "loss": 0.9852, | |
| "step": 186 | |
| }, | |
| { | |
| "epoch": 0.61, | |
| "eval_loss": 0.98876953125, | |
| "eval_runtime": 161.723, | |
| "eval_samples_per_second": 25.995, | |
| "eval_steps_per_second": 1.626, | |
| "step": 186 | |
| }, | |
| { | |
| "epoch": 0.71, | |
| "learning_rate": 0.0001870828693386971, | |
| "loss": 0.9682, | |
| "step": 217 | |
| }, | |
| { | |
| "epoch": 0.71, | |
| "eval_loss": 0.97314453125, | |
| "eval_runtime": 161.5091, | |
| "eval_samples_per_second": 26.029, | |
| "eval_steps_per_second": 1.628, | |
| "step": 217 | |
| }, | |
| { | |
| "epoch": 0.81, | |
| "learning_rate": 0.00017407765595569785, | |
| "loss": 0.9565, | |
| "step": 248 | |
| }, | |
| { | |
| "epoch": 0.81, | |
| "eval_loss": 0.95751953125, | |
| "eval_runtime": 161.4684, | |
| "eval_samples_per_second": 26.036, | |
| "eval_steps_per_second": 1.629, | |
| "step": 248 | |
| }, | |
| { | |
| "epoch": 0.91, | |
| "learning_rate": 0.00016345506187300654, | |
| "loss": 0.9322, | |
| "step": 279 | |
| }, | |
| { | |
| "epoch": 0.91, | |
| "eval_loss": 0.94091796875, | |
| "eval_runtime": 161.6846, | |
| "eval_samples_per_second": 26.001, | |
| "eval_steps_per_second": 1.627, | |
| "step": 279 | |
| }, | |
| { | |
| "epoch": 1.01, | |
| "learning_rate": 0.0001545664419689318, | |
| "loss": 0.9079, | |
| "step": 310 | |
| }, | |
| { | |
| "epoch": 1.01, | |
| "eval_loss": 0.93212890625, | |
| "eval_runtime": 159.6077, | |
| "eval_samples_per_second": 26.34, | |
| "eval_steps_per_second": 1.648, | |
| "step": 310 | |
| }, | |
| { | |
| "epoch": 1.11, | |
| "learning_rate": 0.00014698618394803282, | |
| "loss": 0.8272, | |
| "step": 341 | |
| }, | |
| { | |
| "epoch": 1.11, | |
| "eval_loss": 0.92333984375, | |
| "eval_runtime": 159.6433, | |
| "eval_samples_per_second": 26.334, | |
| "eval_steps_per_second": 1.647, | |
| "step": 341 | |
| }, | |
| { | |
| "epoch": 1.21, | |
| "learning_rate": 0.0001404218994998819, | |
| "loss": 0.7966, | |
| "step": 372 | |
| }, | |
| { | |
| "epoch": 1.21, | |
| "eval_loss": 0.92041015625, | |
| "eval_runtime": 159.6168, | |
| "eval_samples_per_second": 26.338, | |
| "eval_steps_per_second": 1.648, | |
| "step": 372 | |
| }, | |
| { | |
| "epoch": 1.31, | |
| "learning_rate": 0.00013466519604525415, | |
| "loss": 0.7947, | |
| "step": 403 | |
| }, | |
| { | |
| "epoch": 1.31, | |
| "eval_loss": 0.9150390625, | |
| "eval_runtime": 159.659, | |
| "eval_samples_per_second": 26.331, | |
| "eval_steps_per_second": 1.647, | |
| "step": 403 | |
| }, | |
| { | |
| "epoch": 1.42, | |
| "learning_rate": 0.00012956299912940142, | |
| "loss": 0.8089, | |
| "step": 434 | |
| }, | |
| { | |
| "epoch": 1.42, | |
| "eval_loss": 0.9052734375, | |
| "eval_runtime": 159.599, | |
| "eval_samples_per_second": 26.341, | |
| "eval_steps_per_second": 1.648, | |
| "step": 434 | |
| }, | |
| { | |
| "epoch": 1.52, | |
| "learning_rate": 0.000125, | |
| "loss": 0.8113, | |
| "step": 465 | |
| }, | |
| { | |
| "epoch": 1.52, | |
| "eval_loss": 0.89990234375, | |
| "eval_runtime": 159.5797, | |
| "eval_samples_per_second": 26.344, | |
| "eval_steps_per_second": 1.648, | |
| "step": 465 | |
| }, | |
| { | |
| "epoch": 1.62, | |
| "learning_rate": 0.00012088746297956931, | |
| "loss": 0.7849, | |
| "step": 496 | |
| }, | |
| { | |
| "epoch": 1.62, | |
| "eval_loss": 0.89111328125, | |
| "eval_runtime": 159.5898, | |
| "eval_samples_per_second": 26.343, | |
| "eval_steps_per_second": 1.648, | |
| "step": 496 | |
| }, | |
| { | |
| "epoch": 1.72, | |
| "learning_rate": 0.00011715583722580123, | |
| "loss": 0.7796, | |
| "step": 527 | |
| }, | |
| { | |
| "epoch": 1.72, | |
| "eval_loss": 0.8828125, | |
| "eval_runtime": 159.6567, | |
| "eval_samples_per_second": 26.331, | |
| "eval_steps_per_second": 1.647, | |
| "step": 527 | |
| }, | |
| { | |
| "epoch": 1.82, | |
| "learning_rate": 0.00011374973339937476, | |
| "loss": 0.7768, | |
| "step": 558 | |
| }, | |
| { | |
| "epoch": 1.82, | |
| "eval_loss": 0.87548828125, | |
| "eval_runtime": 159.5528, | |
| "eval_samples_per_second": 26.349, | |
| "eval_steps_per_second": 1.648, | |
| "step": 558 | |
| }, | |
| { | |
| "epoch": 1.92, | |
| "learning_rate": 0.00011062441971717747, | |
| "loss": 0.7784, | |
| "step": 589 | |
| }, | |
| { | |
| "epoch": 1.92, | |
| "eval_loss": 0.8720703125, | |
| "eval_runtime": 159.542, | |
| "eval_samples_per_second": 26.35, | |
| "eval_steps_per_second": 1.648, | |
| "step": 589 | |
| }, | |
| { | |
| "epoch": 2.0, | |
| "step": 612, | |
| "total_flos": 7.958718731863982e+17, | |
| "train_loss": 0.9485874550015319, | |
| "train_runtime": 6936.4877, | |
| "train_samples_per_second": 5.654, | |
| "train_steps_per_second": 0.088 | |
| } | |
| ], | |
| "max_steps": 612, | |
| "num_train_epochs": 2, | |
| "total_flos": 7.958718731863982e+17, | |
| "trial_name": null, | |
| "trial_params": null | |
| } | |