checkpoint-400/trainer_state.json

{
 "best_metric": 0.43381235003471375,
 "best_model_checkpoint": "./output_v2/7b_cluster00_Nous-Hermes-llama-2-7b_partitioned_v3_standardized_00/checkpoint-400",
 "epoch": 0.4241781548250265,
 "global_step": 400,
 "is_hyper_param_search": false,
 "is_local_process_zero": true,
 "is_world_process_zero": true,
 "log_history": [
 {
 "epoch": 0.01,
 "learning_rate": 0.0002,
 "loss": 0.5637,
 "step": 10
 },
 {
 "epoch": 0.02,
 "learning_rate": 0.0002,
 "loss": 0.4775,
 "step": 20
 },
 {
 "epoch": 0.03,
 "learning_rate": 0.0002,
 "loss": 0.5079,
 "step": 30
 },
 {
 "epoch": 0.04,
 "learning_rate": 0.0002,
 "loss": 0.4849,
 "step": 40
 },
 {
 "epoch": 0.05,
 "learning_rate": 0.0002,
 "loss": 0.4587,
 "step": 50
 },
 {
 "epoch": 0.06,
 "learning_rate": 0.0002,
 "loss": 0.4867,
 "step": 60
 },
 {
 "epoch": 0.07,
 "learning_rate": 0.0002,
 "loss": 0.4578,
 "step": 70
 },
 {
 "epoch": 0.08,
 "learning_rate": 0.0002,
 "loss": 0.4665,
 "step": 80
 },
 {
 "epoch": 0.1,
 "learning_rate": 0.0002,
 "loss": 0.5044,
 "step": 90
 },
 {
 "epoch": 0.11,
 "learning_rate": 0.0002,
 "loss": 0.4979,
 "step": 100
 },
 {
 "epoch": 0.12,
 "learning_rate": 0.0002,
 "loss": 0.4715,
 "step": 110
 },
 {
 "epoch": 0.13,
 "learning_rate": 0.0002,
 "loss": 0.4717,
 "step": 120
 },
 {
 "epoch": 0.14,
 "learning_rate": 0.0002,
 "loss": 0.4419,
 "step": 130
 },
 {
 "epoch": 0.15,
 "learning_rate": 0.0002,
 "loss": 0.4671,
 "step": 140
 },
 {
 "epoch": 0.16,
 "learning_rate": 0.0002,
 "loss": 0.4503,
 "step": 150
 },
 {
 "epoch": 0.17,
 "learning_rate": 0.0002,
 "loss": 0.4878,
 "step": 160
 },
 {
 "epoch": 0.18,
 "learning_rate": 0.0002,
 "loss": 0.4648,
 "step": 170
 },
 {
 "epoch": 0.19,
 "learning_rate": 0.0002,
 "loss": 0.4625,
 "step": 180
 },
 {
 "epoch": 0.2,
 "learning_rate": 0.0002,
 "loss": 0.4473,
 "step": 190
 },
 {
 "epoch": 0.21,
 "learning_rate": 0.0002,
 "loss": 0.4585,
 "step": 200
 },
 {
 "epoch": 0.21,
 "eval_loss": 0.4498298466205597,
 "eval_runtime": 253.5551,
 "eval_samples_per_second": 3.944,
 "eval_steps_per_second": 0.986,
 "step": 200
 },
 {
 "epoch": 0.22,
 "learning_rate": 0.0002,
 "loss": 0.4396,
 "step": 210
 },
 {
 "epoch": 0.23,
 "learning_rate": 0.0002,
 "loss": 0.4545,
 "step": 220
 },
 {
 "epoch": 0.24,
 "learning_rate": 0.0002,
 "loss": 0.4632,
 "step": 230
 },
 {
 "epoch": 0.25,
 "learning_rate": 0.0002,
 "loss": 0.4458,
 "step": 240
 },
 {
 "epoch": 0.27,
 "learning_rate": 0.0002,
 "loss": 0.4612,
 "step": 250
 },
 {
 "epoch": 0.28,
 "learning_rate": 0.0002,
 "loss": 0.4493,
 "step": 260
 },
 {
 "epoch": 0.29,
 "learning_rate": 0.0002,
 "loss": 0.4652,
 "step": 270
 },
 {
 "epoch": 0.3,
 "learning_rate": 0.0002,
 "loss": 0.4596,
 "step": 280
 },
 {
 "epoch": 0.31,
 "learning_rate": 0.0002,
 "loss": 0.4483,
 "step": 290
 },
 {
 "epoch": 0.32,
 "learning_rate": 0.0002,
 "loss": 0.4386,
 "step": 300
 },
 {
 "epoch": 0.33,
 "learning_rate": 0.0002,
 "loss": 0.4746,
 "step": 310
 },
 {
 "epoch": 0.34,
 "learning_rate": 0.0002,
 "loss": 0.4299,
 "step": 320
 },
 {
 "epoch": 0.35,
 "learning_rate": 0.0002,
 "loss": 0.4462,
 "step": 330
 },
 {
 "epoch": 0.36,
 "learning_rate": 0.0002,
 "loss": 0.4346,
 "step": 340
 },
 {
 "epoch": 0.37,
 "learning_rate": 0.0002,
 "loss": 0.4425,
 "step": 350
 },
 {
 "epoch": 0.38,
 "learning_rate": 0.0002,
 "loss": 0.4356,
 "step": 360
 },
 {
 "epoch": 0.39,
 "learning_rate": 0.0002,
 "loss": 0.4647,
 "step": 370
 },
 {
 "epoch": 0.4,
 "learning_rate": 0.0002,
 "loss": 0.4448,
 "step": 380
 },
 {
 "epoch": 0.41,
 "learning_rate": 0.0002,
 "loss": 0.4208,
 "step": 390
 },
 {
 "epoch": 0.42,
 "learning_rate": 0.0002,
 "loss": 0.4095,
 "step": 400
 },
 {
 "epoch": 0.42,
 "eval_loss": 0.43381235003471375,
 "eval_runtime": 253.5754,
 "eval_samples_per_second": 3.944,
 "eval_steps_per_second": 0.986,
 "step": 400
 }
 ],
 "max_steps": 5000,
 "num_train_epochs": 6,
 "total_flos": 1.0997720781284966e+17,
 "trial_name": null,
 "trial_params": null
}