| { | |
| "best_metric": 2.08911395072937, | |
| "best_model_checkpoint": "./outputs/checkpoint-3600", | |
| "epoch": 2.6229508196721314, | |
| "eval_steps": 100, | |
| "global_step": 3600, | |
| "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": 2.8176, | |
| "step": 100 | |
| }, | |
| { | |
| "epoch": 0.07, | |
| "eval_loss": 2.7248005867004395, | |
| "eval_runtime": 204.526, | |
| "eval_samples_per_second": 30.676, | |
| "eval_steps_per_second": 3.838, | |
| "step": 100 | |
| }, | |
| { | |
| "epoch": 0.15, | |
| "learning_rate": 0.0002, | |
| "loss": 2.6959, | |
| "step": 200 | |
| }, | |
| { | |
| "epoch": 0.15, | |
| "eval_loss": 2.6743462085723877, | |
| "eval_runtime": 205.6888, | |
| "eval_samples_per_second": 30.502, | |
| "eval_steps_per_second": 3.816, | |
| "step": 200 | |
| }, | |
| { | |
| "epoch": 0.22, | |
| "learning_rate": 0.0002, | |
| "loss": 2.6512, | |
| "step": 300 | |
| }, | |
| { | |
| "epoch": 0.22, | |
| "eval_loss": 2.6400036811828613, | |
| "eval_runtime": 204.5763, | |
| "eval_samples_per_second": 30.668, | |
| "eval_steps_per_second": 3.837, | |
| "step": 300 | |
| }, | |
| { | |
| "epoch": 0.29, | |
| "learning_rate": 0.0002, | |
| "loss": 2.6241, | |
| "step": 400 | |
| }, | |
| { | |
| "epoch": 0.29, | |
| "eval_loss": 2.6055357456207275, | |
| "eval_runtime": 204.4295, | |
| "eval_samples_per_second": 30.69, | |
| "eval_steps_per_second": 3.84, | |
| "step": 400 | |
| }, | |
| { | |
| "epoch": 0.36, | |
| "learning_rate": 0.0002, | |
| "loss": 2.5812, | |
| "step": 500 | |
| }, | |
| { | |
| "epoch": 0.36, | |
| "eval_loss": 2.5787856578826904, | |
| "eval_runtime": 204.1968, | |
| "eval_samples_per_second": 30.725, | |
| "eval_steps_per_second": 3.844, | |
| "step": 500 | |
| }, | |
| { | |
| "epoch": 0.44, | |
| "learning_rate": 0.0002, | |
| "loss": 2.553, | |
| "step": 600 | |
| }, | |
| { | |
| "epoch": 0.44, | |
| "eval_loss": 2.552586555480957, | |
| "eval_runtime": 204.5367, | |
| "eval_samples_per_second": 30.674, | |
| "eval_steps_per_second": 3.838, | |
| "step": 600 | |
| }, | |
| { | |
| "epoch": 0.51, | |
| "learning_rate": 0.0002, | |
| "loss": 2.5368, | |
| "step": 700 | |
| }, | |
| { | |
| "epoch": 0.51, | |
| "eval_loss": 2.527393341064453, | |
| "eval_runtime": 204.8117, | |
| "eval_samples_per_second": 30.633, | |
| "eval_steps_per_second": 3.833, | |
| "step": 700 | |
| }, | |
| { | |
| "epoch": 0.58, | |
| "learning_rate": 0.0002, | |
| "loss": 2.5124, | |
| "step": 800 | |
| }, | |
| { | |
| "epoch": 0.58, | |
| "eval_loss": 2.506849527359009, | |
| "eval_runtime": 204.0454, | |
| "eval_samples_per_second": 30.748, | |
| "eval_steps_per_second": 3.847, | |
| "step": 800 | |
| }, | |
| { | |
| "epoch": 0.66, | |
| "learning_rate": 0.0002, | |
| "loss": 2.4756, | |
| "step": 900 | |
| }, | |
| { | |
| "epoch": 0.66, | |
| "eval_loss": 2.4826207160949707, | |
| "eval_runtime": 204.4447, | |
| "eval_samples_per_second": 30.688, | |
| "eval_steps_per_second": 3.84, | |
| "step": 900 | |
| }, | |
| { | |
| "epoch": 0.73, | |
| "learning_rate": 0.0002, | |
| "loss": 2.4724, | |
| "step": 1000 | |
| }, | |
| { | |
| "epoch": 0.73, | |
| "eval_loss": 2.463895559310913, | |
| "eval_runtime": 204.6362, | |
| "eval_samples_per_second": 30.659, | |
| "eval_steps_per_second": 3.836, | |
| "step": 1000 | |
| }, | |
| { | |
| "epoch": 0.8, | |
| "learning_rate": 0.0002, | |
| "loss": 2.4666, | |
| "step": 1100 | |
| }, | |
| { | |
| "epoch": 0.8, | |
| "eval_loss": 2.439748525619507, | |
| "eval_runtime": 204.5892, | |
| "eval_samples_per_second": 30.666, | |
| "eval_steps_per_second": 3.837, | |
| "step": 1100 | |
| }, | |
| { | |
| "epoch": 0.87, | |
| "learning_rate": 0.0002, | |
| "loss": 2.4189, | |
| "step": 1200 | |
| }, | |
| { | |
| "epoch": 0.87, | |
| "eval_loss": 2.421973943710327, | |
| "eval_runtime": 204.4873, | |
| "eval_samples_per_second": 30.682, | |
| "eval_steps_per_second": 3.839, | |
| "step": 1200 | |
| }, | |
| { | |
| "epoch": 0.95, | |
| "learning_rate": 0.0002, | |
| "loss": 2.4317, | |
| "step": 1300 | |
| }, | |
| { | |
| "epoch": 0.95, | |
| "eval_loss": 2.4035205841064453, | |
| "eval_runtime": 204.5174, | |
| "eval_samples_per_second": 30.677, | |
| "eval_steps_per_second": 3.838, | |
| "step": 1300 | |
| }, | |
| { | |
| "epoch": 1.02, | |
| "learning_rate": 0.0002, | |
| "loss": 2.3762, | |
| "step": 1400 | |
| }, | |
| { | |
| "epoch": 1.02, | |
| "eval_loss": 2.379911422729492, | |
| "eval_runtime": 204.9943, | |
| "eval_samples_per_second": 30.606, | |
| "eval_steps_per_second": 3.829, | |
| "step": 1400 | |
| }, | |
| { | |
| "epoch": 1.09, | |
| "learning_rate": 0.0002, | |
| "loss": 2.337, | |
| "step": 1500 | |
| }, | |
| { | |
| "epoch": 1.09, | |
| "eval_loss": 2.3603107929229736, | |
| "eval_runtime": 204.9808, | |
| "eval_samples_per_second": 30.608, | |
| "eval_steps_per_second": 3.83, | |
| "step": 1500 | |
| }, | |
| { | |
| "epoch": 1.17, | |
| "learning_rate": 0.0002, | |
| "loss": 2.3297, | |
| "step": 1600 | |
| }, | |
| { | |
| "epoch": 1.17, | |
| "eval_loss": 2.3435933589935303, | |
| "eval_runtime": 204.6503, | |
| "eval_samples_per_second": 30.657, | |
| "eval_steps_per_second": 3.836, | |
| "step": 1600 | |
| }, | |
| { | |
| "epoch": 1.24, | |
| "learning_rate": 0.0002, | |
| "loss": 2.3291, | |
| "step": 1700 | |
| }, | |
| { | |
| "epoch": 1.24, | |
| "eval_loss": 2.325985908508301, | |
| "eval_runtime": 204.8064, | |
| "eval_samples_per_second": 30.634, | |
| "eval_steps_per_second": 3.833, | |
| "step": 1700 | |
| }, | |
| { | |
| "epoch": 1.31, | |
| "learning_rate": 0.0002, | |
| "loss": 2.2916, | |
| "step": 1800 | |
| }, | |
| { | |
| "epoch": 1.31, | |
| "eval_loss": 2.3098464012145996, | |
| "eval_runtime": 204.5629, | |
| "eval_samples_per_second": 30.67, | |
| "eval_steps_per_second": 3.837, | |
| "step": 1800 | |
| }, | |
| { | |
| "epoch": 1.38, | |
| "learning_rate": 0.0002, | |
| "loss": 2.2727, | |
| "step": 1900 | |
| }, | |
| { | |
| "epoch": 1.38, | |
| "eval_loss": 2.2953364849090576, | |
| "eval_runtime": 204.8297, | |
| "eval_samples_per_second": 30.63, | |
| "eval_steps_per_second": 3.832, | |
| "step": 1900 | |
| }, | |
| { | |
| "epoch": 1.46, | |
| "learning_rate": 0.0002, | |
| "loss": 2.2697, | |
| "step": 2000 | |
| }, | |
| { | |
| "epoch": 1.46, | |
| "eval_loss": 2.279343843460083, | |
| "eval_runtime": 205.0674, | |
| "eval_samples_per_second": 30.595, | |
| "eval_steps_per_second": 3.828, | |
| "step": 2000 | |
| }, | |
| { | |
| "epoch": 1.53, | |
| "learning_rate": 0.0002, | |
| "loss": 2.2494, | |
| "step": 2100 | |
| }, | |
| { | |
| "epoch": 1.53, | |
| "eval_loss": 2.266407012939453, | |
| "eval_runtime": 204.6679, | |
| "eval_samples_per_second": 30.655, | |
| "eval_steps_per_second": 3.835, | |
| "step": 2100 | |
| }, | |
| { | |
| "epoch": 1.6, | |
| "learning_rate": 0.0002, | |
| "loss": 2.232, | |
| "step": 2200 | |
| }, | |
| { | |
| "epoch": 1.6, | |
| "eval_loss": 2.24958872795105, | |
| "eval_runtime": 204.8091, | |
| "eval_samples_per_second": 30.633, | |
| "eval_steps_per_second": 3.833, | |
| "step": 2200 | |
| }, | |
| { | |
| "epoch": 1.68, | |
| "learning_rate": 0.0002, | |
| "loss": 2.2233, | |
| "step": 2300 | |
| }, | |
| { | |
| "epoch": 1.68, | |
| "eval_loss": 2.2403359413146973, | |
| "eval_runtime": 204.8335, | |
| "eval_samples_per_second": 30.63, | |
| "eval_steps_per_second": 3.832, | |
| "step": 2300 | |
| }, | |
| { | |
| "epoch": 1.75, | |
| "learning_rate": 0.0002, | |
| "loss": 2.2295, | |
| "step": 2400 | |
| }, | |
| { | |
| "epoch": 1.75, | |
| "eval_loss": 2.2235193252563477, | |
| "eval_runtime": 204.6216, | |
| "eval_samples_per_second": 30.661, | |
| "eval_steps_per_second": 3.836, | |
| "step": 2400 | |
| }, | |
| { | |
| "epoch": 1.82, | |
| "learning_rate": 0.0002, | |
| "loss": 2.1957, | |
| "step": 2500 | |
| }, | |
| { | |
| "epoch": 1.82, | |
| "eval_loss": 2.2098121643066406, | |
| "eval_runtime": 204.6979, | |
| "eval_samples_per_second": 30.65, | |
| "eval_steps_per_second": 3.835, | |
| "step": 2500 | |
| }, | |
| { | |
| "epoch": 1.89, | |
| "learning_rate": 0.0002, | |
| "loss": 2.1964, | |
| "step": 2600 | |
| }, | |
| { | |
| "epoch": 1.89, | |
| "eval_loss": 2.197718620300293, | |
| "eval_runtime": 204.9105, | |
| "eval_samples_per_second": 30.618, | |
| "eval_steps_per_second": 3.831, | |
| "step": 2600 | |
| }, | |
| { | |
| "epoch": 1.97, | |
| "learning_rate": 0.0002, | |
| "loss": 2.1798, | |
| "step": 2700 | |
| }, | |
| { | |
| "epoch": 1.97, | |
| "eval_loss": 2.1822831630706787, | |
| "eval_runtime": 204.5843, | |
| "eval_samples_per_second": 30.667, | |
| "eval_steps_per_second": 3.837, | |
| "step": 2700 | |
| }, | |
| { | |
| "epoch": 2.04, | |
| "learning_rate": 0.0002, | |
| "loss": 2.136, | |
| "step": 2800 | |
| }, | |
| { | |
| "epoch": 2.04, | |
| "eval_loss": 2.1763219833374023, | |
| "eval_runtime": 204.7219, | |
| "eval_samples_per_second": 30.646, | |
| "eval_steps_per_second": 3.834, | |
| "step": 2800 | |
| }, | |
| { | |
| "epoch": 2.11, | |
| "learning_rate": 0.0002, | |
| "loss": 2.1076, | |
| "step": 2900 | |
| }, | |
| { | |
| "epoch": 2.11, | |
| "eval_loss": 2.1633405685424805, | |
| "eval_runtime": 205.0465, | |
| "eval_samples_per_second": 30.598, | |
| "eval_steps_per_second": 3.828, | |
| "step": 2900 | |
| }, | |
| { | |
| "epoch": 2.19, | |
| "learning_rate": 0.0002, | |
| "loss": 2.1237, | |
| "step": 3000 | |
| }, | |
| { | |
| "epoch": 2.19, | |
| "eval_loss": 2.152331590652466, | |
| "eval_runtime": 204.9684, | |
| "eval_samples_per_second": 30.61, | |
| "eval_steps_per_second": 3.83, | |
| "step": 3000 | |
| }, | |
| { | |
| "epoch": 2.26, | |
| "learning_rate": 0.0002, | |
| "loss": 2.0927, | |
| "step": 3100 | |
| }, | |
| { | |
| "epoch": 2.26, | |
| "eval_loss": 2.139225959777832, | |
| "eval_runtime": 205.1929, | |
| "eval_samples_per_second": 30.576, | |
| "eval_steps_per_second": 3.826, | |
| "step": 3100 | |
| }, | |
| { | |
| "epoch": 2.33, | |
| "learning_rate": 0.0002, | |
| "loss": 2.1127, | |
| "step": 3200 | |
| }, | |
| { | |
| "epoch": 2.33, | |
| "eval_loss": 2.1339216232299805, | |
| "eval_runtime": 205.3153, | |
| "eval_samples_per_second": 30.558, | |
| "eval_steps_per_second": 3.823, | |
| "step": 3200 | |
| }, | |
| { | |
| "epoch": 2.4, | |
| "learning_rate": 0.0002, | |
| "loss": 2.0736, | |
| "step": 3300 | |
| }, | |
| { | |
| "epoch": 2.4, | |
| "eval_loss": 2.1192281246185303, | |
| "eval_runtime": 208.2692, | |
| "eval_samples_per_second": 30.124, | |
| "eval_steps_per_second": 3.769, | |
| "step": 3300 | |
| }, | |
| { | |
| "epoch": 2.48, | |
| "learning_rate": 0.0002, | |
| "loss": 2.0687, | |
| "step": 3400 | |
| }, | |
| { | |
| "epoch": 2.48, | |
| "eval_loss": 2.114696979522705, | |
| "eval_runtime": 205.5149, | |
| "eval_samples_per_second": 30.528, | |
| "eval_steps_per_second": 3.82, | |
| "step": 3400 | |
| }, | |
| { | |
| "epoch": 2.55, | |
| "learning_rate": 0.0002, | |
| "loss": 2.0674, | |
| "step": 3500 | |
| }, | |
| { | |
| "epoch": 2.55, | |
| "eval_loss": 2.0993354320526123, | |
| "eval_runtime": 205.1654, | |
| "eval_samples_per_second": 30.58, | |
| "eval_steps_per_second": 3.826, | |
| "step": 3500 | |
| }, | |
| { | |
| "epoch": 2.62, | |
| "learning_rate": 0.0002, | |
| "loss": 2.0638, | |
| "step": 3600 | |
| }, | |
| { | |
| "epoch": 2.62, | |
| "eval_loss": 2.08911395072937, | |
| "eval_runtime": 206.3208, | |
| "eval_samples_per_second": 30.409, | |
| "eval_steps_per_second": 3.805, | |
| "step": 3600 | |
| } | |
| ], | |
| "logging_steps": 100, | |
| "max_steps": 4116, | |
| "num_train_epochs": 3, | |
| "save_steps": 100, | |
| "total_flos": 1.0251223767558144e+17, | |
| "trial_name": null, | |
| "trial_params": null | |
| } | |