add llama-13b model

with_input/llama-13b/best_model/adapter_config.json

@@ -0,0 +1,21 @@
+{
+  "base_model_name_or_path": "huggyllama/llama-13b",
+  "bias": "none",
+  "enable_lora": null,
+  "fan_in_fan_out": false,
+  "inference_mode": true,
+  "init_lora_weights": true,
+  "lora_alpha": 16,
+  "lora_dropout": 0.05,
+  "merge_weights": false,
+  "modules_to_save": null,
+  "peft_type": "LORA",
+  "r": 16,
+  "target_modules": [
+    "q_proj",
+    "k_proj",
+    "v_proj",
+    "o_proj"
+  ],
+  "task_type": "CAUSAL_LM"
+}

with_input/llama-13b/best_model/adapter_model.bin

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:8e1f97abb4adcf7b9a532f4138d6a6faf01490b35c448f75bf326b153296bc0e
+size 104973389

with_input/llama-13b/best_model/optimizer.pt

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:0893afbc1a4c63a79a1887926395c822c2ce227cbf8ed7e380ad03f0a7f1b41b
+size 209984517

with_input/llama-13b/best_model/rng_state.pth

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:c1c4eb763e21ceebb380f75de90c4d2c30f5cd927cb45e4977b62bc55bcef362
+size 14575

with_input/llama-13b/best_model/scheduler.pt

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:aaf3b274bb9cc118ee254a99daf710e7721c99ca8673122b1053fda825177be7
+size 627

with_input/llama-13b/best_model/trainer_state.json

@@ -0,0 +1,2963 @@
+{
+  "best_metric": 0.2995336055755615,
+  "best_model_checkpoint": "checkpoints/instrucode/decomp_inst_code_output/llama-13b/checkpoint-4600",
+  "epoch": 2.958199356913183,
+  "eval_steps": 200,
+  "global_step": 4600,
+  "is_hyper_param_search": false,
+  "is_local_process_zero": true,
+  "is_world_process_zero": true,
+  "log_history": [
+    {
+      "epoch": 0.01,
+      "learning_rate": 2.9999999999999997e-05,
+      "loss": 1.0025,
+      "step": 10
+    },
+    {
+      "epoch": 0.01,
+      "learning_rate": 5.9999999999999995e-05,
+      "loss": 0.9559,
+      "step": 20
+    },
+    {
+      "epoch": 0.02,
+      "learning_rate": 8.999999999999999e-05,
+      "loss": 0.8381,
+      "step": 30
+    },
+    {
+      "epoch": 0.03,
+      "learning_rate": 0.00011999999999999999,
+      "loss": 0.5633,
+      "step": 40
+    },
+    {
+      "epoch": 0.03,
+      "learning_rate": 0.00015,
+      "loss": 0.4615,
+      "step": 50
+    },
+    {
+      "epoch": 0.04,
+      "learning_rate": 0.00017999999999999998,
+      "loss": 0.4224,
+      "step": 60
+    },
+    {
+      "epoch": 0.05,
+      "learning_rate": 0.00020999999999999998,
+      "loss": 0.4029,
+      "step": 70
+    },
+    {
+      "epoch": 0.05,
+      "learning_rate": 0.00023999999999999998,
+      "loss": 0.3932,
+      "step": 80
+    },
+    {
+      "epoch": 0.06,
+      "learning_rate": 0.00027,
+      "loss": 0.3894,
+      "step": 90
+    },
+    {
+      "epoch": 0.06,
+      "learning_rate": 0.0003,
+      "loss": 0.3798,
+      "step": 100
+    },
+    {
+      "epoch": 0.07,
+      "learning_rate": 0.00029960912052117263,
+      "loss": 0.3762,
+      "step": 110
+    },
+    {
+      "epoch": 0.08,
+      "learning_rate": 0.0002992182410423453,
+      "loss": 0.3678,
+      "step": 120
+    },
+    {
+      "epoch": 0.08,
+      "learning_rate": 0.0002988273615635179,
+      "loss": 0.3574,
+      "step": 130
+    },
+    {
+      "epoch": 0.09,
+      "learning_rate": 0.00029843648208469054,
+      "loss": 0.3576,
+      "step": 140
+    },
+    {
+      "epoch": 0.1,
+      "learning_rate": 0.00029804560260586314,
+      "loss": 0.3585,
+      "step": 150
+    },
+    {
+      "epoch": 0.1,
+      "learning_rate": 0.0002976547231270358,
+      "loss": 0.365,
+      "step": 160
+    },
+    {
+      "epoch": 0.11,
+      "learning_rate": 0.00029726384364820845,
+      "loss": 0.3606,
+      "step": 170
+    },
+    {
+      "epoch": 0.12,
+      "learning_rate": 0.00029687296416938105,
+      "loss": 0.352,
+      "step": 180
+    },
+    {
+      "epoch": 0.12,
+      "learning_rate": 0.0002964820846905537,
+      "loss": 0.3528,
+      "step": 190
+    },
+    {
+      "epoch": 0.13,
+      "learning_rate": 0.00029609120521172636,
+      "loss": 0.3573,
+      "step": 200
+    },
+    {
+      "epoch": 0.13,
+      "eval_loss": 0.35078638792037964,
+      "eval_runtime": 652.6775,
+      "eval_samples_per_second": 3.064,
+      "eval_steps_per_second": 0.383,
+      "step": 200
+    },
+    {
+      "epoch": 0.14,
+      "learning_rate": 0.000295700325732899,
+      "loss": 0.3546,
+      "step": 210
+    },
+    {
+      "epoch": 0.14,
+      "learning_rate": 0.0002953094462540716,
+      "loss": 0.3587,
+      "step": 220
+    },
+    {
+      "epoch": 0.15,
+      "learning_rate": 0.0002949185667752443,
+      "loss": 0.3523,
+      "step": 230
+    },
+    {
+      "epoch": 0.15,
+      "learning_rate": 0.00029452768729641693,
+      "loss": 0.3547,
+      "step": 240
+    },
+    {
+      "epoch": 0.16,
+      "learning_rate": 0.00029413680781758953,
+      "loss": 0.3617,
+      "step": 250
+    },
+    {
+      "epoch": 0.17,
+      "learning_rate": 0.0002937459283387622,
+      "loss": 0.3423,
+      "step": 260
+    },
+    {
+      "epoch": 0.17,
+      "learning_rate": 0.00029335504885993484,
+      "loss": 0.3585,
+      "step": 270
+    },
+    {
+      "epoch": 0.18,
+      "learning_rate": 0.00029296416938110744,
+      "loss": 0.3504,
+      "step": 280
+    },
+    {
+      "epoch": 0.19,
+      "learning_rate": 0.0002925732899022801,
+      "loss": 0.3493,
+      "step": 290
+    },
+    {
+      "epoch": 0.19,
+      "learning_rate": 0.00029218241042345275,
+      "loss": 0.3439,
+      "step": 300
+    },
+    {
+      "epoch": 0.2,
+      "learning_rate": 0.0002917915309446254,
+      "loss": 0.3436,
+      "step": 310
+    },
+    {
+      "epoch": 0.21,
+      "learning_rate": 0.000291400651465798,
+      "loss": 0.3364,
+      "step": 320
+    },
+    {
+      "epoch": 0.21,
+      "learning_rate": 0.00029100977198697066,
+      "loss": 0.3533,
+      "step": 330
+    },
+    {
+      "epoch": 0.22,
+      "learning_rate": 0.00029061889250814327,
+      "loss": 0.3377,
+      "step": 340
+    },
+    {
+      "epoch": 0.23,
+      "learning_rate": 0.0002902280130293159,
+      "loss": 0.3526,
+      "step": 350
+    },
+    {
+      "epoch": 0.23,
+      "learning_rate": 0.0002898371335504886,
+      "loss": 0.3488,
+      "step": 360
+    },
+    {
+      "epoch": 0.24,
+      "learning_rate": 0.00028944625407166123,
+      "loss": 0.3406,
+      "step": 370
+    },
+    {
+      "epoch": 0.24,
+      "learning_rate": 0.0002890553745928339,
+      "loss": 0.3334,
+      "step": 380
+    },
+    {
+      "epoch": 0.25,
+      "learning_rate": 0.0002886644951140065,
+      "loss": 0.3364,
+      "step": 390
+    },
+    {
+      "epoch": 0.26,
+      "learning_rate": 0.00028827361563517914,
+      "loss": 0.3391,
+      "step": 400
+    },
+    {
+      "epoch": 0.26,
+      "eval_loss": 0.3346547484397888,
+      "eval_runtime": 652.5321,
+      "eval_samples_per_second": 3.065,
+      "eval_steps_per_second": 0.383,
+      "step": 400
+    },
+    {
+      "epoch": 0.26,
+      "learning_rate": 0.00028788273615635174,
+      "loss": 0.3266,
+      "step": 410
+    },
+    {
+      "epoch": 0.27,
+      "learning_rate": 0.0002874918566775244,
+      "loss": 0.3484,
+      "step": 420
+    },
+    {
+      "epoch": 0.28,
+      "learning_rate": 0.00028710097719869705,
+      "loss": 0.3382,
+      "step": 430
+    },
+    {
+      "epoch": 0.28,
+      "learning_rate": 0.00028671009771986966,
+      "loss": 0.3467,
+      "step": 440
+    },
+    {
+      "epoch": 0.29,
+      "learning_rate": 0.0002863192182410423,
+      "loss": 0.3454,
+      "step": 450
+    },
+    {
+      "epoch": 0.3,
+      "learning_rate": 0.00028592833876221497,
+      "loss": 0.3327,
+      "step": 460
+    },
+    {
+      "epoch": 0.3,
+      "learning_rate": 0.0002855374592833876,
+      "loss": 0.3391,
+      "step": 470
+    },
+    {
+      "epoch": 0.31,
+      "learning_rate": 0.0002851465798045602,
+      "loss": 0.3365,
+      "step": 480
+    },
+    {
+      "epoch": 0.32,
+      "learning_rate": 0.0002847557003257329,
+      "loss": 0.3372,
+      "step": 490
+    },
+    {
+      "epoch": 0.32,
+      "learning_rate": 0.0002843648208469055,
+      "loss": 0.3414,
+      "step": 500
+    },
+    {
+      "epoch": 0.33,
+      "learning_rate": 0.00028397394136807813,
+      "loss": 0.3289,
+      "step": 510
+    },
+    {
+      "epoch": 0.33,
+      "learning_rate": 0.0002835830618892508,
+      "loss": 0.3287,
+      "step": 520
+    },
+    {
+      "epoch": 0.34,
+      "learning_rate": 0.00028319218241042344,
+      "loss": 0.3244,
+      "step": 530
+    },
+    {
+      "epoch": 0.35,
+      "learning_rate": 0.00028280130293159605,
+      "loss": 0.3416,
+      "step": 540
+    },
+    {
+      "epoch": 0.35,
+      "learning_rate": 0.0002824104234527687,
+      "loss": 0.3307,
+      "step": 550
+    },
+    {
+      "epoch": 0.36,
+      "learning_rate": 0.00028201954397394136,
+      "loss": 0.3299,
+      "step": 560
+    },
+    {
+      "epoch": 0.37,
+      "learning_rate": 0.000281628664495114,
+      "loss": 0.3252,
+      "step": 570
+    },
+    {
+      "epoch": 0.37,
+      "learning_rate": 0.0002812377850162866,
+      "loss": 0.332,
+      "step": 580
+    },
+    {
+      "epoch": 0.38,
+      "learning_rate": 0.00028084690553745927,
+      "loss": 0.3405,
+      "step": 590
+    },
+    {
+      "epoch": 0.39,
+      "learning_rate": 0.00028045602605863187,
+      "loss": 0.3358,
+      "step": 600
+    },
+    {
+      "epoch": 0.39,
+      "eval_loss": 0.32762062549591064,
+      "eval_runtime": 652.4973,
+      "eval_samples_per_second": 3.065,
+      "eval_steps_per_second": 0.383,
+      "step": 600
+    },
+    {
+      "epoch": 0.39,
+      "learning_rate": 0.0002800651465798045,
+      "loss": 0.3281,
+      "step": 610
+    },
+    {
+      "epoch": 0.4,
+      "learning_rate": 0.0002796742671009772,
+      "loss": 0.3271,
+      "step": 620
+    },
+    {
+      "epoch": 0.41,
+      "learning_rate": 0.00027928338762214983,
+      "loss": 0.34,
+      "step": 630
+    },
+    {
+      "epoch": 0.41,
+      "learning_rate": 0.00027889250814332244,
+      "loss": 0.3342,
+      "step": 640
+    },
+    {
+      "epoch": 0.42,
+      "learning_rate": 0.0002785016286644951,
+      "loss": 0.3219,
+      "step": 650
+    },
+    {
+      "epoch": 0.42,
+      "learning_rate": 0.00027811074918566775,
+      "loss": 0.3335,
+      "step": 660
+    },
+    {
+      "epoch": 0.43,
+      "learning_rate": 0.00027771986970684035,
+      "loss": 0.3354,
+      "step": 670
+    },
+    {
+      "epoch": 0.44,
+      "learning_rate": 0.000277328990228013,
+      "loss": 0.3197,
+      "step": 680
+    },
+    {
+      "epoch": 0.44,
+      "learning_rate": 0.00027693811074918566,
+      "loss": 0.3287,
+      "step": 690
+    },
+    {
+      "epoch": 0.45,
+      "learning_rate": 0.00027654723127035826,
+      "loss": 0.3237,
+      "step": 700
+    },
+    {
+      "epoch": 0.46,
+      "learning_rate": 0.0002761563517915309,
+      "loss": 0.315,
+      "step": 710
+    },
+    {
+      "epoch": 0.46,
+      "learning_rate": 0.00027576547231270357,
+      "loss": 0.3304,
+      "step": 720
+    },
+    {
+      "epoch": 0.47,
+      "learning_rate": 0.0002753745928338762,
+      "loss": 0.3237,
+      "step": 730
+    },
+    {
+      "epoch": 0.48,
+      "learning_rate": 0.0002749837133550488,
+      "loss": 0.34,
+      "step": 740
+    },
+    {
+      "epoch": 0.48,
+      "learning_rate": 0.0002745928338762215,
+      "loss": 0.328,
+      "step": 750
+    },
+    {
+      "epoch": 0.49,
+      "learning_rate": 0.0002742019543973941,
+      "loss": 0.324,
+      "step": 760
+    },
+    {
+      "epoch": 0.5,
+      "learning_rate": 0.00027381107491856674,
+      "loss": 0.3266,
+      "step": 770
+    },
+    {
+      "epoch": 0.5,
+      "learning_rate": 0.0002734201954397394,
+      "loss": 0.321,
+      "step": 780
+    },
+    {
+      "epoch": 0.51,
+      "learning_rate": 0.00027302931596091205,
+      "loss": 0.3193,
+      "step": 790
+    },
+    {
+      "epoch": 0.51,
+      "learning_rate": 0.00027263843648208465,
+      "loss": 0.3252,
+      "step": 800
+    },
+    {
+      "epoch": 0.51,
+      "eval_loss": 0.3219829201698303,
+      "eval_runtime": 652.1833,
+      "eval_samples_per_second": 3.067,
+      "eval_steps_per_second": 0.383,
+      "step": 800
+    },
+    {
+      "epoch": 0.52,
+      "learning_rate": 0.0002722475570032573,
+      "loss": 0.3195,
+      "step": 810
+    },
+    {
+      "epoch": 0.53,
+      "learning_rate": 0.00027185667752442996,
+      "loss": 0.3203,
+      "step": 820
+    },
+    {
+      "epoch": 0.53,
+      "learning_rate": 0.0002714657980456026,
+      "loss": 0.3228,
+      "step": 830
+    },
+    {
+      "epoch": 0.54,
+      "learning_rate": 0.0002710749185667752,
+      "loss": 0.3269,
+      "step": 840
+    },
+    {
+      "epoch": 0.55,
+      "learning_rate": 0.00027068403908794787,
+      "loss": 0.3274,
+      "step": 850
+    },
+    {
+      "epoch": 0.55,
+      "learning_rate": 0.00027029315960912047,
+      "loss": 0.3261,
+      "step": 860
+    },
+    {
+      "epoch": 0.56,
+      "learning_rate": 0.0002699022801302931,
+      "loss": 0.3274,
+      "step": 870
+    },
+    {
+      "epoch": 0.57,
+      "learning_rate": 0.0002695114006514658,
+      "loss": 0.3364,
+      "step": 880
+    },
+    {
+      "epoch": 0.57,
+      "learning_rate": 0.00026912052117263844,
+      "loss": 0.3273,
+      "step": 890
+    },
+    {
+      "epoch": 0.58,
+      "learning_rate": 0.00026872964169381104,
+      "loss": 0.3269,
+      "step": 900
+    },
+    {
+      "epoch": 0.59,
+      "learning_rate": 0.0002683387622149837,
+      "loss": 0.319,
+      "step": 910
+    },
+    {
+      "epoch": 0.59,
+      "learning_rate": 0.00026794788273615635,
+      "loss": 0.3246,
+      "step": 920
+    },
+    {
+      "epoch": 0.6,
+      "learning_rate": 0.00026755700325732895,
+      "loss": 0.322,
+      "step": 930
+    },
+    {
+      "epoch": 0.6,
+      "learning_rate": 0.0002671661237785016,
+      "loss": 0.3138,
+      "step": 940
+    },
+    {
+      "epoch": 0.61,
+      "learning_rate": 0.0002667752442996742,
+      "loss": 0.3218,
+      "step": 950
+    },
+    {
+      "epoch": 0.62,
+      "learning_rate": 0.00026638436482084686,
+      "loss": 0.3328,
+      "step": 960
+    },
+    {
+      "epoch": 0.62,
+      "learning_rate": 0.0002659934853420195,
+      "loss": 0.3228,
+      "step": 970
+    },
+    {
+      "epoch": 0.63,
+      "learning_rate": 0.00026560260586319217,
+      "loss": 0.3374,
+      "step": 980
+    },
+    {
+      "epoch": 0.64,
+      "learning_rate": 0.0002652117263843648,
+      "loss": 0.324,
+      "step": 990
+    },
+    {
+      "epoch": 0.64,
+      "learning_rate": 0.00026482084690553743,
+      "loss": 0.3221,
+      "step": 1000
+    },
+    {
+      "epoch": 0.64,
+      "eval_loss": 0.31776905059814453,
+      "eval_runtime": 652.1008,
+      "eval_samples_per_second": 3.067,
+      "eval_steps_per_second": 0.383,
+      "step": 1000
+    },
+    {
+      "epoch": 0.65,
+      "learning_rate": 0.0002644299674267101,
+      "loss": 0.3237,
+      "step": 1010
+    },
+    {
+      "epoch": 0.66,
+      "learning_rate": 0.0002640390879478827,
+      "loss": 0.3153,
+      "step": 1020
+    },
+    {
+      "epoch": 0.66,
+      "learning_rate": 0.00026364820846905534,
+      "loss": 0.326,
+      "step": 1030
+    },
+    {
+      "epoch": 0.67,
+      "learning_rate": 0.000263257328990228,
+      "loss": 0.3261,
+      "step": 1040
+    },
+    {
+      "epoch": 0.68,
+      "learning_rate": 0.0002628664495114006,
+      "loss": 0.3269,
+      "step": 1050
+    },
+    {
+      "epoch": 0.68,
+      "learning_rate": 0.00026247557003257325,
+      "loss": 0.3173,
+      "step": 1060
+    },
+    {
+      "epoch": 0.69,
+      "learning_rate": 0.0002620846905537459,
+      "loss": 0.3178,
+      "step": 1070
+    },
+    {
+      "epoch": 0.69,
+      "learning_rate": 0.00026169381107491856,
+      "loss": 0.3119,
+      "step": 1080
+    },
+    {
+      "epoch": 0.7,
+      "learning_rate": 0.0002613029315960912,
+      "loss": 0.3281,
+      "step": 1090
+    },
+    {
+      "epoch": 0.71,
+      "learning_rate": 0.0002609120521172638,
+      "loss": 0.3223,
+      "step": 1100
+    },
+    {
+      "epoch": 0.71,
+      "learning_rate": 0.0002605211726384365,
+      "loss": 0.3335,
+      "step": 1110
+    },
+    {
+      "epoch": 0.72,
+      "learning_rate": 0.0002601302931596091,
+      "loss": 0.3281,
+      "step": 1120
+    },
+    {
+      "epoch": 0.73,
+      "learning_rate": 0.00025973941368078173,
+      "loss": 0.3138,
+      "step": 1130
+    },
+    {
+      "epoch": 0.73,
+      "learning_rate": 0.0002593485342019544,
+      "loss": 0.3214,
+      "step": 1140
+    },
+    {
+      "epoch": 0.74,
+      "learning_rate": 0.00025895765472312704,
+      "loss": 0.3101,
+      "step": 1150
+    },
+    {
+      "epoch": 0.75,
+      "learning_rate": 0.00025856677524429964,
+      "loss": 0.3134,
+      "step": 1160
+    },
+    {
+      "epoch": 0.75,
+      "learning_rate": 0.0002581758957654723,
+      "loss": 0.3255,
+      "step": 1170
+    },
+    {
+      "epoch": 0.76,
+      "learning_rate": 0.00025778501628664495,
+      "loss": 0.3131,
+      "step": 1180
+    },
+    {
+      "epoch": 0.77,
+      "learning_rate": 0.00025739413680781755,
+      "loss": 0.3189,
+      "step": 1190
+    },
+    {
+      "epoch": 0.77,
+      "learning_rate": 0.0002570032573289902,
+      "loss": 0.3242,
+      "step": 1200
+    },
+    {
+      "epoch": 0.77,
+      "eval_loss": 0.31515294313430786,
+      "eval_runtime": 652.6339,
+      "eval_samples_per_second": 3.065,
+      "eval_steps_per_second": 0.383,
+      "step": 1200
+    },
+    {
+      "epoch": 0.78,
+      "learning_rate": 0.0002566123778501628,
+      "loss": 0.321,
+      "step": 1210
+    },
+    {
+      "epoch": 0.78,
+      "learning_rate": 0.00025622149837133546,
+      "loss": 0.3149,
+      "step": 1220
+    },
+    {
+      "epoch": 0.79,
+      "learning_rate": 0.0002558306188925081,
+      "loss": 0.3213,
+      "step": 1230
+    },
+    {
+      "epoch": 0.8,
+      "learning_rate": 0.0002554397394136808,
+      "loss": 0.3168,
+      "step": 1240
+    },
+    {
+      "epoch": 0.8,
+      "learning_rate": 0.00025504885993485343,
+      "loss": 0.3128,
+      "step": 1250
+    },
+    {
+      "epoch": 0.81,
+      "learning_rate": 0.00025465798045602603,
+      "loss": 0.3212,
+      "step": 1260
+    },
+    {
+      "epoch": 0.82,
+      "learning_rate": 0.0002542671009771987,
+      "loss": 0.317,
+      "step": 1270
+    },
+    {
+      "epoch": 0.82,
+      "learning_rate": 0.0002538762214983713,
+      "loss": 0.3322,
+      "step": 1280
+    },
+    {
+      "epoch": 0.83,
+      "learning_rate": 0.00025348534201954394,
+      "loss": 0.3238,
+      "step": 1290
+    },
+    {
+      "epoch": 0.84,
+      "learning_rate": 0.0002530944625407166,
+      "loss": 0.3229,
+      "step": 1300
+    },
+    {
+      "epoch": 0.84,
+      "learning_rate": 0.0002527035830618892,
+      "loss": 0.3159,
+      "step": 1310
+    },
+    {
+      "epoch": 0.85,
+      "learning_rate": 0.00025231270358306185,
+      "loss": 0.3081,
+      "step": 1320
+    },
+    {
+      "epoch": 0.86,
+      "learning_rate": 0.0002519218241042345,
+      "loss": 0.3184,
+      "step": 1330
+    },
+    {
+      "epoch": 0.86,
+      "learning_rate": 0.00025153094462540716,
+      "loss": 0.3088,
+      "step": 1340
+    },
+    {
+      "epoch": 0.87,
+      "learning_rate": 0.0002511400651465798,
+      "loss": 0.3176,
+      "step": 1350
+    },
+    {
+      "epoch": 0.87,
+      "learning_rate": 0.0002507491856677524,
+      "loss": 0.3228,
+      "step": 1360
+    },
+    {
+      "epoch": 0.88,
+      "learning_rate": 0.0002503583061889251,
+      "loss": 0.3082,
+      "step": 1370
+    },
+    {
+      "epoch": 0.89,
+      "learning_rate": 0.0002499674267100977,
+      "loss": 0.3217,
+      "step": 1380
+    },
+    {
+      "epoch": 0.89,
+      "learning_rate": 0.00024957654723127033,
+      "loss": 0.321,
+      "step": 1390
+    },
+    {
+      "epoch": 0.9,
+      "learning_rate": 0.000249185667752443,
+      "loss": 0.3245,
+      "step": 1400
+    },
+    {
+      "epoch": 0.9,
+      "eval_loss": 0.312686949968338,
+      "eval_runtime": 652.3104,
+      "eval_samples_per_second": 3.066,
+      "eval_steps_per_second": 0.383,
+      "step": 1400
+    },
+    {
+      "epoch": 0.91,
+      "learning_rate": 0.0002487947882736156,
+      "loss": 0.3123,
+      "step": 1410
+    },
+    {
+      "epoch": 0.91,
+      "learning_rate": 0.00024840390879478824,
+      "loss": 0.3151,
+      "step": 1420
+    },
+    {
+      "epoch": 0.92,
+      "learning_rate": 0.0002480130293159609,
+      "loss": 0.3161,
+      "step": 1430
+    },
+    {
+      "epoch": 0.93,
+      "learning_rate": 0.00024762214983713355,
+      "loss": 0.3213,
+      "step": 1440
+    },
+    {
+      "epoch": 0.93,
+      "learning_rate": 0.00024723127035830616,
+      "loss": 0.3257,
+      "step": 1450
+    },
+    {
+      "epoch": 0.94,
+      "learning_rate": 0.0002468403908794788,
+      "loss": 0.3156,
+      "step": 1460
+    },
+    {
+      "epoch": 0.95,
+      "learning_rate": 0.0002464495114006514,
+      "loss": 0.3317,
+      "step": 1470
+    },
+    {
+      "epoch": 0.95,
+      "learning_rate": 0.00024605863192182407,
+      "loss": 0.3281,
+      "step": 1480
+    },
+    {
+      "epoch": 0.96,
+      "learning_rate": 0.0002456677524429967,
+      "loss": 0.3064,
+      "step": 1490
+    },
+    {
+      "epoch": 0.96,
+      "learning_rate": 0.0002452768729641694,
+      "loss": 0.3161,
+      "step": 1500
+    },
+    {
+      "epoch": 0.97,
+      "learning_rate": 0.00024488599348534203,
+      "loss": 0.3222,
+      "step": 1510
+    },
+    {
+      "epoch": 0.98,
+      "learning_rate": 0.00024449511400651463,
+      "loss": 0.3028,
+      "step": 1520
+    },
+    {
+      "epoch": 0.98,
+      "learning_rate": 0.00024410423452768726,
+      "loss": 0.3153,
+      "step": 1530
+    },
+    {
+      "epoch": 0.99,
+      "learning_rate": 0.00024371335504885992,
+      "loss": 0.3186,
+      "step": 1540
+    },
+    {
+      "epoch": 1.0,
+      "learning_rate": 0.00024332247557003257,
+      "loss": 0.3096,
+      "step": 1550
+    },
+    {
+      "epoch": 1.0,
+      "learning_rate": 0.0002429315960912052,
+      "loss": 0.3169,
+      "step": 1560
+    },
+    {
+      "epoch": 1.01,
+      "learning_rate": 0.00024254071661237783,
+      "loss": 0.3068,
+      "step": 1570
+    },
+    {
+      "epoch": 1.02,
+      "learning_rate": 0.00024214983713355046,
+      "loss": 0.2979,
+      "step": 1580
+    },
+    {
+      "epoch": 1.02,
+      "learning_rate": 0.0002417589576547231,
+      "loss": 0.2879,
+      "step": 1590
+    },
+    {
+      "epoch": 1.03,
+      "learning_rate": 0.00024136807817589574,
+      "loss": 0.3003,
+      "step": 1600
+    },
+    {
+      "epoch": 1.03,
+      "eval_loss": 0.3113372325897217,
+      "eval_runtime": 653.0377,
+      "eval_samples_per_second": 3.063,
+      "eval_steps_per_second": 0.383,
+      "step": 1600
+    },
+    {
+      "epoch": 1.04,
+      "learning_rate": 0.0002409771986970684,
+      "loss": 0.303,
+      "step": 1610
+    },
+    {
+      "epoch": 1.04,
+      "learning_rate": 0.000240586319218241,
+      "loss": 0.3087,
+      "step": 1620
+    },
+    {
+      "epoch": 1.05,
+      "learning_rate": 0.00024019543973941365,
+      "loss": 0.2985,
+      "step": 1630
+    },
+    {
+      "epoch": 1.05,
+      "learning_rate": 0.0002398045602605863,
+      "loss": 0.2957,
+      "step": 1640
+    },
+    {
+      "epoch": 1.06,
+      "learning_rate": 0.00023941368078175893,
+      "loss": 0.2972,
+      "step": 1650
+    },
+    {
+      "epoch": 1.07,
+      "learning_rate": 0.0002390228013029316,
+      "loss": 0.3098,
+      "step": 1660
+    },
+    {
+      "epoch": 1.07,
+      "learning_rate": 0.0002386319218241042,
+      "loss": 0.2927,
+      "step": 1670
+    },
+    {
+      "epoch": 1.08,
+      "learning_rate": 0.00023824104234527685,
+      "loss": 0.309,
+      "step": 1680
+    },
+    {
+      "epoch": 1.09,
+      "learning_rate": 0.0002378501628664495,
+      "loss": 0.2963,
+      "step": 1690
+    },
+    {
+      "epoch": 1.09,
+      "learning_rate": 0.00023745928338762213,
+      "loss": 0.2942,
+      "step": 1700
+    },
+    {
+      "epoch": 1.1,
+      "learning_rate": 0.00023706840390879478,
+      "loss": 0.3026,
+      "step": 1710
+    },
+    {
+      "epoch": 1.11,
+      "learning_rate": 0.00023667752442996739,
+      "loss": 0.2983,
+      "step": 1720
+    },
+    {
+      "epoch": 1.11,
+      "learning_rate": 0.00023628664495114004,
+      "loss": 0.3076,
+      "step": 1730
+    },
+    {
+      "epoch": 1.12,
+      "learning_rate": 0.00023589576547231267,
+      "loss": 0.2844,
+      "step": 1740
+    },
+    {
+      "epoch": 1.13,
+      "learning_rate": 0.00023550488599348532,
+      "loss": 0.3025,
+      "step": 1750
+    },
+    {
+      "epoch": 1.13,
+      "learning_rate": 0.00023511400651465798,
+      "loss": 0.3109,
+      "step": 1760
+    },
+    {
+      "epoch": 1.14,
+      "learning_rate": 0.00023472312703583058,
+      "loss": 0.3024,
+      "step": 1770
+    },
+    {
+      "epoch": 1.14,
+      "learning_rate": 0.00023433224755700324,
+      "loss": 0.3048,
+      "step": 1780
+    },
+    {
+      "epoch": 1.15,
+      "learning_rate": 0.00023394136807817586,
+      "loss": 0.3046,
+      "step": 1790
+    },
+    {
+      "epoch": 1.16,
+      "learning_rate": 0.00023355048859934852,
+      "loss": 0.3042,
+      "step": 1800
+    },
+    {
+      "epoch": 1.16,
+      "eval_loss": 0.30957579612731934,
+      "eval_runtime": 652.4621,
+      "eval_samples_per_second": 3.065,
+      "eval_steps_per_second": 0.383,
+      "step": 1800
+    },
+    {
+      "epoch": 1.16,
+      "learning_rate": 0.00023315960912052117,
+      "loss": 0.3001,
+      "step": 1810
+    },
+    {
+      "epoch": 1.17,
+      "learning_rate": 0.00023276872964169378,
+      "loss": 0.3057,
+      "step": 1820
+    },
+    {
+      "epoch": 1.18,
+      "learning_rate": 0.00023237785016286643,
+      "loss": 0.3024,
+      "step": 1830
+    },
+    {
+      "epoch": 1.18,
+      "learning_rate": 0.00023198697068403906,
+      "loss": 0.2999,
+      "step": 1840
+    },
+    {
+      "epoch": 1.19,
+      "learning_rate": 0.00023159609120521171,
+      "loss": 0.3065,
+      "step": 1850
+    },
+    {
+      "epoch": 1.2,
+      "learning_rate": 0.00023120521172638434,
+      "loss": 0.3082,
+      "step": 1860
+    },
+    {
+      "epoch": 1.2,
+      "learning_rate": 0.000230814332247557,
+      "loss": 0.3017,
+      "step": 1870
+    },
+    {
+      "epoch": 1.21,
+      "learning_rate": 0.0002304234527687296,
+      "loss": 0.2988,
+      "step": 1880
+    },
+    {
+      "epoch": 1.22,
+      "learning_rate": 0.00023003257328990225,
+      "loss": 0.2953,
+      "step": 1890
+    },
+    {
+      "epoch": 1.22,
+      "learning_rate": 0.0002296416938110749,
+      "loss": 0.304,
+      "step": 1900
+    },
+    {
+      "epoch": 1.23,
+      "learning_rate": 0.00022925081433224754,
+      "loss": 0.2932,
+      "step": 1910
+    },
+    {
+      "epoch": 1.23,
+      "learning_rate": 0.0002288599348534202,
+      "loss": 0.3024,
+      "step": 1920
+    },
+    {
+      "epoch": 1.24,
+      "learning_rate": 0.0002284690553745928,
+      "loss": 0.3101,
+      "step": 1930
+    },
+    {
+      "epoch": 1.25,
+      "learning_rate": 0.00022807817589576545,
+      "loss": 0.3137,
+      "step": 1940
+    },
+    {
+      "epoch": 1.25,
+      "learning_rate": 0.0002276872964169381,
+      "loss": 0.3107,
+      "step": 1950
+    },
+    {
+      "epoch": 1.26,
+      "learning_rate": 0.00022729641693811073,
+      "loss": 0.2929,
+      "step": 1960
+    },
+    {
+      "epoch": 1.27,
+      "learning_rate": 0.0002269055374592834,
+      "loss": 0.3111,
+      "step": 1970
+    },
+    {
+      "epoch": 1.27,
+      "learning_rate": 0.000226514657980456,
+      "loss": 0.3,
+      "step": 1980
+    },
+    {
+      "epoch": 1.28,
+      "learning_rate": 0.00022612377850162864,
+      "loss": 0.3153,
+      "step": 1990
+    },
+    {
+      "epoch": 1.29,
+      "learning_rate": 0.00022573289902280127,
+      "loss": 0.2988,
+      "step": 2000
+    },
+    {
+      "epoch": 1.29,
+      "eval_loss": 0.3079931437969208,
+      "eval_runtime": 652.3691,
+      "eval_samples_per_second": 3.066,
+      "eval_steps_per_second": 0.383,
+      "step": 2000
+    },
+    {
+      "epoch": 1.29,
+      "learning_rate": 0.00022534201954397393,
+      "loss": 0.2983,
+      "step": 2010
+    },
+    {
+      "epoch": 1.3,
+      "learning_rate": 0.00022495114006514658,
+      "loss": 0.2936,
+      "step": 2020
+    },
+    {
+      "epoch": 1.31,
+      "learning_rate": 0.00022456026058631918,
+      "loss": 0.301,
+      "step": 2030
+    },
+    {
+      "epoch": 1.31,
+      "learning_rate": 0.00022416938110749184,
+      "loss": 0.3107,
+      "step": 2040
+    },
+    {
+      "epoch": 1.32,
+      "learning_rate": 0.00022377850162866447,
+      "loss": 0.3012,
+      "step": 2050
+    },
+    {
+      "epoch": 1.32,
+      "learning_rate": 0.00022338762214983712,
+      "loss": 0.3013,
+      "step": 2060
+    },
+    {
+      "epoch": 1.33,
+      "learning_rate": 0.00022299674267100978,
+      "loss": 0.2957,
+      "step": 2070
+    },
+    {
+      "epoch": 1.34,
+      "learning_rate": 0.00022260586319218238,
+      "loss": 0.2945,
+      "step": 2080
+    },
+    {
+      "epoch": 1.34,
+      "learning_rate": 0.00022221498371335503,
+      "loss": 0.2982,
+      "step": 2090
+    },
+    {
+      "epoch": 1.35,
+      "learning_rate": 0.00022182410423452766,
+      "loss": 0.3027,
+      "step": 2100
+    },
+    {
+      "epoch": 1.36,
+      "learning_rate": 0.00022143322475570032,
+      "loss": 0.301,
+      "step": 2110
+    },
+    {
+      "epoch": 1.36,
+      "learning_rate": 0.00022104234527687295,
+      "loss": 0.2915,
+      "step": 2120
+    },
+    {
+      "epoch": 1.37,
+      "learning_rate": 0.00022065146579804557,
+      "loss": 0.3011,
+      "step": 2130
+    },
+    {
+      "epoch": 1.38,
+      "learning_rate": 0.0002202605863192182,
+      "loss": 0.3054,
+      "step": 2140
+    },
+    {
+      "epoch": 1.38,
+      "learning_rate": 0.00021986970684039086,
+      "loss": 0.2921,
+      "step": 2150
+    },
+    {
+      "epoch": 1.39,
+      "learning_rate": 0.0002194788273615635,
+      "loss": 0.3061,
+      "step": 2160
+    },
+    {
+      "epoch": 1.4,
+      "learning_rate": 0.00021908794788273614,
+      "loss": 0.3046,
+      "step": 2170
+    },
+    {
+      "epoch": 1.4,
+      "learning_rate": 0.00021869706840390877,
+      "loss": 0.2986,
+      "step": 2180
+    },
+    {
+      "epoch": 1.41,
+      "learning_rate": 0.0002183061889250814,
+      "loss": 0.2946,
+      "step": 2190
+    },
+    {
+      "epoch": 1.41,
+      "learning_rate": 0.00021791530944625405,
+      "loss": 0.3072,
+      "step": 2200
+    },
+    {
+      "epoch": 1.41,
+      "eval_loss": 0.3071443438529968,
+      "eval_runtime": 652.3269,
+      "eval_samples_per_second": 3.066,
+      "eval_steps_per_second": 0.383,
+      "step": 2200
+    },
+    {
+      "epoch": 1.42,
+      "learning_rate": 0.0002175244299674267,
+      "loss": 0.2987,
+      "step": 2210
+    },
+    {
+      "epoch": 1.43,
+      "learning_rate": 0.00021713355048859934,
+      "loss": 0.3064,
+      "step": 2220
+    },
+    {
+      "epoch": 1.43,
+      "learning_rate": 0.000216742671009772,
+      "loss": 0.3116,
+      "step": 2230
+    },
+    {
+      "epoch": 1.44,
+      "learning_rate": 0.0002163517915309446,
+      "loss": 0.3075,
+      "step": 2240
+    },
+    {
+      "epoch": 1.45,
+      "learning_rate": 0.00021596091205211725,
+      "loss": 0.3067,
+      "step": 2250
+    },
+    {
+      "epoch": 1.45,
+      "learning_rate": 0.00021557003257328987,
+      "loss": 0.2994,
+      "step": 2260
+    },
+    {
+      "epoch": 1.46,
+      "learning_rate": 0.00021517915309446253,
+      "loss": 0.3098,
+      "step": 2270
+    },
+    {
+      "epoch": 1.47,
+      "learning_rate": 0.00021478827361563519,
+      "loss": 0.3059,
+      "step": 2280
+    },
+    {
+      "epoch": 1.47,
+      "learning_rate": 0.00021439739413680779,
+      "loss": 0.2999,
+      "step": 2290
+    },
+    {
+      "epoch": 1.48,
+      "learning_rate": 0.00021400651465798044,
+      "loss": 0.2987,
+      "step": 2300
+    },
+    {
+      "epoch": 1.49,
+      "learning_rate": 0.00021361563517915307,
+      "loss": 0.3023,
+      "step": 2310
+    },
+    {
+      "epoch": 1.49,
+      "learning_rate": 0.00021322475570032572,
+      "loss": 0.3032,
+      "step": 2320
+    },
+    {
+      "epoch": 1.5,
+      "learning_rate": 0.00021283387622149838,
+      "loss": 0.2963,
+      "step": 2330
+    },
+    {
+      "epoch": 1.5,
+      "learning_rate": 0.00021244299674267098,
+      "loss": 0.3026,
+      "step": 2340
+    },
+    {
+      "epoch": 1.51,
+      "learning_rate": 0.00021205211726384364,
+      "loss": 0.2954,
+      "step": 2350
+    },
+    {
+      "epoch": 1.52,
+      "learning_rate": 0.00021166123778501626,
+      "loss": 0.2953,
+      "step": 2360
+    },
+    {
+      "epoch": 1.52,
+      "learning_rate": 0.00021127035830618892,
+      "loss": 0.2856,
+      "step": 2370
+    },
+    {
+      "epoch": 1.53,
+      "learning_rate": 0.00021087947882736155,
+      "loss": 0.3047,
+      "step": 2380
+    },
+    {
+      "epoch": 1.54,
+      "learning_rate": 0.00021048859934853418,
+      "loss": 0.3155,
+      "step": 2390
+    },
+    {
+      "epoch": 1.54,
+      "learning_rate": 0.0002100977198697068,
+      "loss": 0.2948,
+      "step": 2400
+    },
+    {
+      "epoch": 1.54,
+      "eval_loss": 0.30597493052482605,
+      "eval_runtime": 652.522,
+      "eval_samples_per_second": 3.065,
+      "eval_steps_per_second": 0.383,
+      "step": 2400
+    },
+    {
+      "epoch": 1.55,
+      "learning_rate": 0.00020970684039087946,
+      "loss": 0.302,
+      "step": 2410
+    },
+    {
+      "epoch": 1.56,
+      "learning_rate": 0.00020931596091205211,
+      "loss": 0.313,
+      "step": 2420
+    },
+    {
+      "epoch": 1.56,
+      "learning_rate": 0.00020892508143322474,
+      "loss": 0.2996,
+      "step": 2430
+    },
+    {
+      "epoch": 1.57,
+      "learning_rate": 0.00020853420195439737,
+      "loss": 0.2997,
+      "step": 2440
+    },
+    {
+      "epoch": 1.58,
+      "learning_rate": 0.00020814332247557,
+      "loss": 0.2958,
+      "step": 2450
+    },
+    {
+      "epoch": 1.58,
+      "learning_rate": 0.00020775244299674265,
+      "loss": 0.2885,
+      "step": 2460
+    },
+    {
+      "epoch": 1.59,
+      "learning_rate": 0.0002073615635179153,
+      "loss": 0.2875,
+      "step": 2470
+    },
+    {
+      "epoch": 1.59,
+      "learning_rate": 0.00020697068403908794,
+      "loss": 0.3041,
+      "step": 2480
+    },
+    {
+      "epoch": 1.6,
+      "learning_rate": 0.00020657980456026057,
+      "loss": 0.3042,
+      "step": 2490
+    },
+    {
+      "epoch": 1.61,
+      "learning_rate": 0.0002061889250814332,
+      "loss": 0.3024,
+      "step": 2500
+    },
+    {
+      "epoch": 1.61,
+      "learning_rate": 0.00020579804560260585,
+      "loss": 0.295,
+      "step": 2510
+    },
+    {
+      "epoch": 1.62,
+      "learning_rate": 0.00020540716612377848,
+      "loss": 0.2995,
+      "step": 2520
+    },
+    {
+      "epoch": 1.63,
+      "learning_rate": 0.00020501628664495113,
+      "loss": 0.3031,
+      "step": 2530
+    },
+    {
+      "epoch": 1.63,
+      "learning_rate": 0.00020462540716612373,
+      "loss": 0.2957,
+      "step": 2540
+    },
+    {
+      "epoch": 1.64,
+      "learning_rate": 0.0002042345276872964,
+      "loss": 0.3092,
+      "step": 2550
+    },
+    {
+      "epoch": 1.65,
+      "learning_rate": 0.00020384364820846904,
+      "loss": 0.2886,
+      "step": 2560
+    },
+    {
+      "epoch": 1.65,
+      "learning_rate": 0.00020345276872964167,
+      "loss": 0.295,
+      "step": 2570
+    },
+    {
+      "epoch": 1.66,
+      "learning_rate": 0.00020306188925081433,
+      "loss": 0.3001,
+      "step": 2580
+    },
+    {
+      "epoch": 1.67,
+      "learning_rate": 0.00020267100977198693,
+      "loss": 0.2987,
+      "step": 2590
+    },
+    {
+      "epoch": 1.67,
+      "learning_rate": 0.00020228013029315958,
+      "loss": 0.2969,
+      "step": 2600
+    },
+    {
+      "epoch": 1.67,
+      "eval_loss": 0.30438002943992615,
+      "eval_runtime": 652.754,
+      "eval_samples_per_second": 3.064,
+      "eval_steps_per_second": 0.383,
+      "step": 2600
+    },
+    {
+      "epoch": 1.68,
+      "learning_rate": 0.00020188925081433224,
+      "loss": 0.3108,
+      "step": 2610
+    },
+    {
+      "epoch": 1.68,
+      "learning_rate": 0.00020149837133550487,
+      "loss": 0.3136,
+      "step": 2620
+    },
+    {
+      "epoch": 1.69,
+      "learning_rate": 0.00020110749185667752,
+      "loss": 0.2966,
+      "step": 2630
+    },
+    {
+      "epoch": 1.7,
+      "learning_rate": 0.00020071661237785015,
+      "loss": 0.2933,
+      "step": 2640
+    },
+    {
+      "epoch": 1.7,
+      "learning_rate": 0.00020032573289902278,
+      "loss": 0.2982,
+      "step": 2650
+    },
+    {
+      "epoch": 1.71,
+      "learning_rate": 0.0001999348534201954,
+      "loss": 0.2979,
+      "step": 2660
+    },
+    {
+      "epoch": 1.72,
+      "learning_rate": 0.00019954397394136806,
+      "loss": 0.294,
+      "step": 2670
+    },
+    {
+      "epoch": 1.72,
+      "learning_rate": 0.00019915309446254072,
+      "loss": 0.3042,
+      "step": 2680
+    },
+    {
+      "epoch": 1.73,
+      "learning_rate": 0.00019876221498371335,
+      "loss": 0.3027,
+      "step": 2690
+    },
+    {
+      "epoch": 1.74,
+      "learning_rate": 0.00019837133550488597,
+      "loss": 0.2985,
+      "step": 2700
+    },
+    {
+      "epoch": 1.74,
+      "learning_rate": 0.0001979804560260586,
+      "loss": 0.2956,
+      "step": 2710
+    },
+    {
+      "epoch": 1.75,
+      "learning_rate": 0.00019758957654723126,
+      "loss": 0.2918,
+      "step": 2720
+    },
+    {
+      "epoch": 1.76,
+      "learning_rate": 0.0001971986970684039,
+      "loss": 0.2973,
+      "step": 2730
+    },
+    {
+      "epoch": 1.76,
+      "learning_rate": 0.00019680781758957654,
+      "loss": 0.2976,
+      "step": 2740
+    },
+    {
+      "epoch": 1.77,
+      "learning_rate": 0.00019641693811074917,
+      "loss": 0.3139,
+      "step": 2750
+    },
+    {
+      "epoch": 1.77,
+      "learning_rate": 0.0001960260586319218,
+      "loss": 0.3028,
+      "step": 2760
+    },
+    {
+      "epoch": 1.78,
+      "learning_rate": 0.00019563517915309445,
+      "loss": 0.2972,
+      "step": 2770
+    },
+    {
+      "epoch": 1.79,
+      "learning_rate": 0.00019524429967426708,
+      "loss": 0.3028,
+      "step": 2780
+    },
+    {
+      "epoch": 1.79,
+      "learning_rate": 0.00019485342019543974,
+      "loss": 0.2972,
+      "step": 2790
+    },
+    {
+      "epoch": 1.8,
+      "learning_rate": 0.00019446254071661234,
+      "loss": 0.2964,
+      "step": 2800
+    },
+    {
+      "epoch": 1.8,
+      "eval_loss": 0.3036755323410034,
+      "eval_runtime": 652.7658,
+      "eval_samples_per_second": 3.064,
+      "eval_steps_per_second": 0.383,
+      "step": 2800
+    },
+    {
+      "epoch": 1.81,
+      "learning_rate": 0.000194071661237785,
+      "loss": 0.296,
+      "step": 2810
+    },
+    {
+      "epoch": 1.81,
+      "learning_rate": 0.00019368078175895765,
+      "loss": 0.2949,
+      "step": 2820
+    },
+    {
+      "epoch": 1.82,
+      "learning_rate": 0.00019328990228013028,
+      "loss": 0.2817,
+      "step": 2830
+    },
+    {
+      "epoch": 1.83,
+      "learning_rate": 0.00019289902280130293,
+      "loss": 0.3067,
+      "step": 2840
+    },
+    {
+      "epoch": 1.83,
+      "learning_rate": 0.00019250814332247553,
+      "loss": 0.3074,
+      "step": 2850
+    },
+    {
+      "epoch": 1.84,
+      "learning_rate": 0.0001921172638436482,
+      "loss": 0.3094,
+      "step": 2860
+    },
+    {
+      "epoch": 1.85,
+      "learning_rate": 0.00019172638436482084,
+      "loss": 0.2883,
+      "step": 2870
+    },
+    {
+      "epoch": 1.85,
+      "learning_rate": 0.00019133550488599347,
+      "loss": 0.2962,
+      "step": 2880
+    },
+    {
+      "epoch": 1.86,
+      "learning_rate": 0.00019094462540716613,
+      "loss": 0.3033,
+      "step": 2890
+    },
+    {
+      "epoch": 1.86,
+      "learning_rate": 0.00019055374592833873,
+      "loss": 0.2962,
+      "step": 2900
+    },
+    {
+      "epoch": 1.87,
+      "learning_rate": 0.00019016286644951138,
+      "loss": 0.2978,
+      "step": 2910
+    },
+    {
+      "epoch": 1.88,
+      "learning_rate": 0.000189771986970684,
+      "loss": 0.293,
+      "step": 2920
+    },
+    {
+      "epoch": 1.88,
+      "learning_rate": 0.00018938110749185666,
+      "loss": 0.3042,
+      "step": 2930
+    },
+    {
+      "epoch": 1.89,
+      "learning_rate": 0.00018899022801302932,
+      "loss": 0.3068,
+      "step": 2940
+    },
+    {
+      "epoch": 1.9,
+      "learning_rate": 0.00018859934853420192,
+      "loss": 0.3076,
+      "step": 2950
+    },
+    {
+      "epoch": 1.9,
+      "learning_rate": 0.00018820846905537458,
+      "loss": 0.2969,
+      "step": 2960
+    },
+    {
+      "epoch": 1.91,
+      "learning_rate": 0.0001878175895765472,
+      "loss": 0.3023,
+      "step": 2970
+    },
+    {
+      "epoch": 1.92,
+      "learning_rate": 0.00018742671009771986,
+      "loss": 0.3004,
+      "step": 2980
+    },
+    {
+      "epoch": 1.92,
+      "learning_rate": 0.00018703583061889252,
+      "loss": 0.2872,
+      "step": 2990
+    },
+    {
+      "epoch": 1.93,
+      "learning_rate": 0.00018664495114006514,
+      "loss": 0.3013,
+      "step": 3000
+    },
+    {
+      "epoch": 1.93,
+      "eval_loss": 0.3025144636631012,
+      "eval_runtime": 652.6847,
+      "eval_samples_per_second": 3.064,
+      "eval_steps_per_second": 0.383,
+      "step": 3000
+    },
+    {
+      "epoch": 1.94,
+      "learning_rate": 0.00018625407166123777,
+      "loss": 0.3009,
+      "step": 3010
+    },
+    {
+      "epoch": 1.94,
+      "learning_rate": 0.0001858631921824104,
+      "loss": 0.2969,
+      "step": 3020
+    },
+    {
+      "epoch": 1.95,
+      "learning_rate": 0.00018547231270358305,
+      "loss": 0.3166,
+      "step": 3030
+    },
+    {
+      "epoch": 1.95,
+      "learning_rate": 0.00018508143322475568,
+      "loss": 0.3045,
+      "step": 3040
+    },
+    {
+      "epoch": 1.96,
+      "learning_rate": 0.00018469055374592834,
+      "loss": 0.2901,
+      "step": 3050
+    },
+    {
+      "epoch": 1.97,
+      "learning_rate": 0.00018429967426710094,
+      "loss": 0.3019,
+      "step": 3060
+    },
+    {
+      "epoch": 1.97,
+      "learning_rate": 0.0001839087947882736,
+      "loss": 0.2987,
+      "step": 3070
+    },
+    {
+      "epoch": 1.98,
+      "learning_rate": 0.00018351791530944625,
+      "loss": 0.2898,
+      "step": 3080
+    },
+    {
+      "epoch": 1.99,
+      "learning_rate": 0.00018312703583061888,
+      "loss": 0.2902,
+      "step": 3090
+    },
+    {
+      "epoch": 1.99,
+      "learning_rate": 0.00018273615635179153,
+      "loss": 0.2925,
+      "step": 3100
+    },
+    {
+      "epoch": 2.0,
+      "learning_rate": 0.00018234527687296413,
+      "loss": 0.2999,
+      "step": 3110
+    },
+    {
+      "epoch": 2.01,
+      "learning_rate": 0.0001819543973941368,
+      "loss": 0.27,
+      "step": 3120
+    },
+    {
+      "epoch": 2.01,
+      "learning_rate": 0.00018156351791530944,
+      "loss": 0.2845,
+      "step": 3130
+    },
+    {
+      "epoch": 2.02,
+      "learning_rate": 0.00018117263843648207,
+      "loss": 0.2868,
+      "step": 3140
+    },
+    {
+      "epoch": 2.03,
+      "learning_rate": 0.00018078175895765473,
+      "loss": 0.2779,
+      "step": 3150
+    },
+    {
+      "epoch": 2.03,
+      "learning_rate": 0.00018039087947882733,
+      "loss": 0.2733,
+      "step": 3160
+    },
+    {
+      "epoch": 2.04,
+      "learning_rate": 0.00017999999999999998,
+      "loss": 0.2715,
+      "step": 3170
+    },
+    {
+      "epoch": 2.05,
+      "learning_rate": 0.0001796091205211726,
+      "loss": 0.2934,
+      "step": 3180
+    },
+    {
+      "epoch": 2.05,
+      "learning_rate": 0.00017921824104234527,
+      "loss": 0.2839,
+      "step": 3190
+    },
+    {
+      "epoch": 2.06,
+      "learning_rate": 0.00017882736156351792,
+      "loss": 0.2896,
+      "step": 3200
+    },
+    {
+      "epoch": 2.06,
+      "eval_loss": 0.30334094166755676,
+      "eval_runtime": 652.4252,
+      "eval_samples_per_second": 3.065,
+      "eval_steps_per_second": 0.383,
+      "step": 3200
+    },
+    {
+      "epoch": 2.06,
+      "learning_rate": 0.00017843648208469052,
+      "loss": 0.2782,
+      "step": 3210
+    },
+    {
+      "epoch": 2.07,
+      "learning_rate": 0.00017804560260586318,
+      "loss": 0.2808,
+      "step": 3220
+    },
+    {
+      "epoch": 2.08,
+      "learning_rate": 0.0001776547231270358,
+      "loss": 0.2698,
+      "step": 3230
+    },
+    {
+      "epoch": 2.08,
+      "learning_rate": 0.00017726384364820846,
+      "loss": 0.2767,
+      "step": 3240
+    },
+    {
+      "epoch": 2.09,
+      "learning_rate": 0.0001768729641693811,
+      "loss": 0.2906,
+      "step": 3250
+    },
+    {
+      "epoch": 2.1,
+      "learning_rate": 0.00017648208469055372,
+      "loss": 0.2666,
+      "step": 3260
+    },
+    {
+      "epoch": 2.1,
+      "learning_rate": 0.00017609120521172637,
+      "loss": 0.2865,
+      "step": 3270
+    },
+    {
+      "epoch": 2.11,
+      "learning_rate": 0.000175700325732899,
+      "loss": 0.2783,
+      "step": 3280
+    },
+    {
+      "epoch": 2.12,
+      "learning_rate": 0.00017530944625407166,
+      "loss": 0.277,
+      "step": 3290
+    },
+    {
+      "epoch": 2.12,
+      "learning_rate": 0.00017491856677524429,
+      "loss": 0.2755,
+      "step": 3300
+    },
+    {
+      "epoch": 2.13,
+      "learning_rate": 0.00017452768729641691,
+      "loss": 0.2833,
+      "step": 3310
+    },
+    {
+      "epoch": 2.14,
+      "learning_rate": 0.00017413680781758954,
+      "loss": 0.2801,
+      "step": 3320
+    },
+    {
+      "epoch": 2.14,
+      "learning_rate": 0.0001737459283387622,
+      "loss": 0.2841,
+      "step": 3330
+    },
+    {
+      "epoch": 2.15,
+      "learning_rate": 0.00017335504885993485,
+      "loss": 0.2934,
+      "step": 3340
+    },
+    {
+      "epoch": 2.15,
+      "learning_rate": 0.00017296416938110748,
+      "loss": 0.2761,
+      "step": 3350
+    },
+    {
+      "epoch": 2.16,
+      "learning_rate": 0.0001725732899022801,
+      "loss": 0.2791,
+      "step": 3360
+    },
+    {
+      "epoch": 2.17,
+      "learning_rate": 0.00017218241042345274,
+      "loss": 0.289,
+      "step": 3370
+    },
+    {
+      "epoch": 2.17,
+      "learning_rate": 0.0001717915309446254,
+      "loss": 0.2777,
+      "step": 3380
+    },
+    {
+      "epoch": 2.18,
+      "learning_rate": 0.00017140065146579802,
+      "loss": 0.2928,
+      "step": 3390
+    },
+    {
+      "epoch": 2.19,
+      "learning_rate": 0.00017100977198697068,
+      "loss": 0.2783,
+      "step": 3400
+    },
+    {
+      "epoch": 2.19,
+      "eval_loss": 0.30402716994285583,
+      "eval_runtime": 652.056,
+      "eval_samples_per_second": 3.067,
+      "eval_steps_per_second": 0.383,
+      "step": 3400
+    },
+    {
+      "epoch": 2.19,
+      "learning_rate": 0.00017061889250814333,
+      "loss": 0.2855,
+      "step": 3410
+    },
+    {
+      "epoch": 2.2,
+      "learning_rate": 0.00017022801302931593,
+      "loss": 0.2792,
+      "step": 3420
+    },
+    {
+      "epoch": 2.21,
+      "learning_rate": 0.0001698371335504886,
+      "loss": 0.2837,
+      "step": 3430
+    },
+    {
+      "epoch": 2.21,
+      "learning_rate": 0.00016944625407166122,
+      "loss": 0.2783,
+      "step": 3440
+    },
+    {
+      "epoch": 2.22,
+      "learning_rate": 0.00016905537459283387,
+      "loss": 0.2872,
+      "step": 3450
+    },
+    {
+      "epoch": 2.23,
+      "learning_rate": 0.00016866449511400653,
+      "loss": 0.2774,
+      "step": 3460
+    },
+    {
+      "epoch": 2.23,
+      "learning_rate": 0.00016827361563517913,
+      "loss": 0.2842,
+      "step": 3470
+    },
+    {
+      "epoch": 2.24,
+      "learning_rate": 0.00016788273615635178,
+      "loss": 0.2858,
+      "step": 3480
+    },
+    {
+      "epoch": 2.24,
+      "learning_rate": 0.0001674918566775244,
+      "loss": 0.2772,
+      "step": 3490
+    },
+    {
+      "epoch": 2.25,
+      "learning_rate": 0.00016710097719869707,
+      "loss": 0.2903,
+      "step": 3500
+    },
+    {
+      "epoch": 2.26,
+      "learning_rate": 0.0001667100977198697,
+      "loss": 0.2875,
+      "step": 3510
+    },
+    {
+      "epoch": 2.26,
+      "learning_rate": 0.00016631921824104232,
+      "loss": 0.2684,
+      "step": 3520
+    },
+    {
+      "epoch": 2.27,
+      "learning_rate": 0.00016592833876221495,
+      "loss": 0.2844,
+      "step": 3530
+    },
+    {
+      "epoch": 2.28,
+      "learning_rate": 0.0001655374592833876,
+      "loss": 0.2815,
+      "step": 3540
+    },
+    {
+      "epoch": 2.28,
+      "learning_rate": 0.00016514657980456026,
+      "loss": 0.2793,
+      "step": 3550
+    },
+    {
+      "epoch": 2.29,
+      "learning_rate": 0.0001647557003257329,
+      "loss": 0.2882,
+      "step": 3560
+    },
+    {
+      "epoch": 2.3,
+      "learning_rate": 0.00016436482084690552,
+      "loss": 0.2879,
+      "step": 3570
+    },
+    {
+      "epoch": 2.3,
+      "learning_rate": 0.00016397394136807814,
+      "loss": 0.2715,
+      "step": 3580
+    },
+    {
+      "epoch": 2.31,
+      "learning_rate": 0.0001635830618892508,
+      "loss": 0.2779,
+      "step": 3590
+    },
+    {
+      "epoch": 2.32,
+      "learning_rate": 0.00016319218241042346,
+      "loss": 0.2783,
+      "step": 3600
+    },
+    {
+      "epoch": 2.32,
+      "eval_loss": 0.3026120662689209,
+      "eval_runtime": 652.1724,
+      "eval_samples_per_second": 3.067,
+      "eval_steps_per_second": 0.383,
+      "step": 3600
+    },
+    {
+      "epoch": 2.32,
+      "learning_rate": 0.00016280130293159608,
+      "loss": 0.2773,
+      "step": 3610
+    },
+    {
+      "epoch": 2.33,
+      "learning_rate": 0.0001624104234527687,
+      "loss": 0.28,
+      "step": 3620
+    },
+    {
+      "epoch": 2.33,
+      "learning_rate": 0.00016201954397394134,
+      "loss": 0.2759,
+      "step": 3630
+    },
+    {
+      "epoch": 2.34,
+      "learning_rate": 0.000161628664495114,
+      "loss": 0.2833,
+      "step": 3640
+    },
+    {
+      "epoch": 2.35,
+      "learning_rate": 0.00016123778501628662,
+      "loss": 0.2795,
+      "step": 3650
+    },
+    {
+      "epoch": 2.35,
+      "learning_rate": 0.00016084690553745928,
+      "loss": 0.2764,
+      "step": 3660
+    },
+    {
+      "epoch": 2.36,
+      "learning_rate": 0.00016045602605863188,
+      "loss": 0.2821,
+      "step": 3670
+    },
+    {
+      "epoch": 2.37,
+      "learning_rate": 0.00016006514657980453,
+      "loss": 0.2946,
+      "step": 3680
+    },
+    {
+      "epoch": 2.37,
+      "learning_rate": 0.0001596742671009772,
+      "loss": 0.285,
+      "step": 3690
+    },
+    {
+      "epoch": 2.38,
+      "learning_rate": 0.00015928338762214982,
+      "loss": 0.2809,
+      "step": 3700
+    },
+    {
+      "epoch": 2.39,
+      "learning_rate": 0.00015889250814332247,
+      "loss": 0.2814,
+      "step": 3710
+    },
+    {
+      "epoch": 2.39,
+      "learning_rate": 0.00015850162866449507,
+      "loss": 0.2822,
+      "step": 3720
+    },
+    {
+      "epoch": 2.4,
+      "learning_rate": 0.00015811074918566773,
+      "loss": 0.2856,
+      "step": 3730
+    },
+    {
+      "epoch": 2.41,
+      "learning_rate": 0.00015771986970684038,
+      "loss": 0.2796,
+      "step": 3740
+    },
+    {
+      "epoch": 2.41,
+      "learning_rate": 0.000157328990228013,
+      "loss": 0.2884,
+      "step": 3750
+    },
+    {
+      "epoch": 2.42,
+      "learning_rate": 0.00015693811074918567,
+      "loss": 0.2869,
+      "step": 3760
+    },
+    {
+      "epoch": 2.42,
+      "learning_rate": 0.0001565472312703583,
+      "loss": 0.2941,
+      "step": 3770
+    },
+    {
+      "epoch": 2.43,
+      "learning_rate": 0.00015615635179153092,
+      "loss": 0.2771,
+      "step": 3780
+    },
+    {
+      "epoch": 2.44,
+      "learning_rate": 0.00015576547231270355,
+      "loss": 0.2844,
+      "step": 3790
+    },
+    {
+      "epoch": 2.44,
+      "learning_rate": 0.0001553745928338762,
+      "loss": 0.2859,
+      "step": 3800
+    },
+    {
+      "epoch": 2.44,
+      "eval_loss": 0.3023357689380646,
+      "eval_runtime": 652.1026,
+      "eval_samples_per_second": 3.067,
+      "eval_steps_per_second": 0.383,
+      "step": 3800
+    },
+    {
+      "epoch": 2.45,
+      "learning_rate": 0.00015498371335504886,
+      "loss": 0.3061,
+      "step": 3810
+    },
+    {
+      "epoch": 2.46,
+      "learning_rate": 0.0001545928338762215,
+      "loss": 0.28,
+      "step": 3820
+    },
+    {
+      "epoch": 2.46,
+      "learning_rate": 0.00015420195439739412,
+      "loss": 0.2785,
+      "step": 3830
+    },
+    {
+      "epoch": 2.47,
+      "learning_rate": 0.00015381107491856675,
+      "loss": 0.2974,
+      "step": 3840
+    },
+    {
+      "epoch": 2.48,
+      "learning_rate": 0.0001534201954397394,
+      "loss": 0.2767,
+      "step": 3850
+    },
+    {
+      "epoch": 2.48,
+      "learning_rate": 0.00015302931596091206,
+      "loss": 0.2913,
+      "step": 3860
+    },
+    {
+      "epoch": 2.49,
+      "learning_rate": 0.00015263843648208469,
+      "loss": 0.2902,
+      "step": 3870
+    },
+    {
+      "epoch": 2.5,
+      "learning_rate": 0.00015224755700325731,
+      "loss": 0.288,
+      "step": 3880
+    },
+    {
+      "epoch": 2.5,
+      "learning_rate": 0.00015185667752442994,
+      "loss": 0.2686,
+      "step": 3890
+    },
+    {
+      "epoch": 2.51,
+      "learning_rate": 0.0001514657980456026,
+      "loss": 0.2869,
+      "step": 3900
+    },
+    {
+      "epoch": 2.51,
+      "learning_rate": 0.00015107491856677523,
+      "loss": 0.2765,
+      "step": 3910
+    },
+    {
+      "epoch": 2.52,
+      "learning_rate": 0.00015068403908794788,
+      "loss": 0.285,
+      "step": 3920
+    },
+    {
+      "epoch": 2.53,
+      "learning_rate": 0.00015029315960912048,
+      "loss": 0.2763,
+      "step": 3930
+    },
+    {
+      "epoch": 2.53,
+      "learning_rate": 0.00014990228013029314,
+      "loss": 0.2889,
+      "step": 3940
+    },
+    {
+      "epoch": 2.54,
+      "learning_rate": 0.0001495114006514658,
+      "loss": 0.2811,
+      "step": 3950
+    },
+    {
+      "epoch": 2.55,
+      "learning_rate": 0.00014912052117263842,
+      "loss": 0.2848,
+      "step": 3960
+    },
+    {
+      "epoch": 2.55,
+      "learning_rate": 0.00014872964169381105,
+      "loss": 0.2737,
+      "step": 3970
+    },
+    {
+      "epoch": 2.56,
+      "learning_rate": 0.0001483387622149837,
+      "loss": 0.2889,
+      "step": 3980
+    },
+    {
+      "epoch": 2.57,
+      "learning_rate": 0.00014794788273615633,
+      "loss": 0.2785,
+      "step": 3990
+    },
+    {
+      "epoch": 2.57,
+      "learning_rate": 0.000147557003257329,
+      "loss": 0.2795,
+      "step": 4000
+    },
+    {
+      "epoch": 2.57,
+      "eval_loss": 0.3021774888038635,
+      "eval_runtime": 652.172,
+      "eval_samples_per_second": 3.067,
+      "eval_steps_per_second": 0.383,
+      "step": 4000
+    },
+    {
+      "epoch": 2.58,
+      "learning_rate": 0.00014716612377850162,
+      "loss": 0.2744,
+      "step": 4010
+    },
+    {
+      "epoch": 2.59,
+      "learning_rate": 0.00014677524429967424,
+      "loss": 0.2801,
+      "step": 4020
+    },
+    {
+      "epoch": 2.59,
+      "learning_rate": 0.0001463843648208469,
+      "loss": 0.2917,
+      "step": 4030
+    },
+    {
+      "epoch": 2.6,
+      "learning_rate": 0.00014599348534201953,
+      "loss": 0.2828,
+      "step": 4040
+    },
+    {
+      "epoch": 2.6,
+      "learning_rate": 0.00014560260586319216,
+      "loss": 0.2822,
+      "step": 4050
+    },
+    {
+      "epoch": 2.61,
+      "learning_rate": 0.0001452117263843648,
+      "loss": 0.2828,
+      "step": 4060
+    },
+    {
+      "epoch": 2.62,
+      "learning_rate": 0.00014482084690553744,
+      "loss": 0.2777,
+      "step": 4070
+    },
+    {
+      "epoch": 2.62,
+      "learning_rate": 0.0001444299674267101,
+      "loss": 0.2819,
+      "step": 4080
+    },
+    {
+      "epoch": 2.63,
+      "learning_rate": 0.00014403908794788272,
+      "loss": 0.2823,
+      "step": 4090
+    },
+    {
+      "epoch": 2.64,
+      "learning_rate": 0.00014364820846905535,
+      "loss": 0.2741,
+      "step": 4100
+    },
+    {
+      "epoch": 2.64,
+      "learning_rate": 0.000143257328990228,
+      "loss": 0.2755,
+      "step": 4110
+    },
+    {
+      "epoch": 2.65,
+      "learning_rate": 0.00014286644951140063,
+      "loss": 0.2947,
+      "step": 4120
+    },
+    {
+      "epoch": 2.66,
+      "learning_rate": 0.0001424755700325733,
+      "loss": 0.2788,
+      "step": 4130
+    },
+    {
+      "epoch": 2.66,
+      "learning_rate": 0.00014208469055374592,
+      "loss": 0.2737,
+      "step": 4140
+    },
+    {
+      "epoch": 2.67,
+      "learning_rate": 0.00014169381107491854,
+      "loss": 0.2779,
+      "step": 4150
+    },
+    {
+      "epoch": 2.68,
+      "learning_rate": 0.0001413029315960912,
+      "loss": 0.2764,
+      "step": 4160
+    },
+    {
+      "epoch": 2.68,
+      "learning_rate": 0.00014091205211726383,
+      "loss": 0.2803,
+      "step": 4170
+    },
+    {
+      "epoch": 2.69,
+      "learning_rate": 0.00014052117263843646,
+      "loss": 0.2788,
+      "step": 4180
+    },
+    {
+      "epoch": 2.69,
+      "learning_rate": 0.0001401302931596091,
+      "loss": 0.2808,
+      "step": 4190
+    },
+    {
+      "epoch": 2.7,
+      "learning_rate": 0.00013973941368078174,
+      "loss": 0.2821,
+      "step": 4200
+    },
+    {
+      "epoch": 2.7,
+      "eval_loss": 0.3011840879917145,
+      "eval_runtime": 652.1651,
+      "eval_samples_per_second": 3.067,
+      "eval_steps_per_second": 0.383,
+      "step": 4200
+    },
+    {
+      "epoch": 2.71,
+      "learning_rate": 0.0001393485342019544,
+      "loss": 0.2733,
+      "step": 4210
+    },
+    {
+      "epoch": 2.71,
+      "learning_rate": 0.00013895765472312702,
+      "loss": 0.28,
+      "step": 4220
+    },
+    {
+      "epoch": 2.72,
+      "learning_rate": 0.00013856677524429965,
+      "loss": 0.2738,
+      "step": 4230
+    },
+    {
+      "epoch": 2.73,
+      "learning_rate": 0.0001381758957654723,
+      "loss": 0.2875,
+      "step": 4240
+    },
+    {
+      "epoch": 2.73,
+      "learning_rate": 0.00013778501628664493,
+      "loss": 0.2764,
+      "step": 4250
+    },
+    {
+      "epoch": 2.74,
+      "learning_rate": 0.0001373941368078176,
+      "loss": 0.2822,
+      "step": 4260
+    },
+    {
+      "epoch": 2.75,
+      "learning_rate": 0.00013700325732899022,
+      "loss": 0.2752,
+      "step": 4270
+    },
+    {
+      "epoch": 2.75,
+      "learning_rate": 0.00013661237785016285,
+      "loss": 0.2901,
+      "step": 4280
+    },
+    {
+      "epoch": 2.76,
+      "learning_rate": 0.0001362214983713355,
+      "loss": 0.2829,
+      "step": 4290
+    },
+    {
+      "epoch": 2.77,
+      "learning_rate": 0.00013583061889250813,
+      "loss": 0.2769,
+      "step": 4300
+    },
+    {
+      "epoch": 2.77,
+      "learning_rate": 0.00013543973941368076,
+      "loss": 0.3043,
+      "step": 4310
+    },
+    {
+      "epoch": 2.78,
+      "learning_rate": 0.0001350488599348534,
+      "loss": 0.2837,
+      "step": 4320
+    },
+    {
+      "epoch": 2.78,
+      "learning_rate": 0.00013465798045602604,
+      "loss": 0.2855,
+      "step": 4330
+    },
+    {
+      "epoch": 2.79,
+      "learning_rate": 0.0001342671009771987,
+      "loss": 0.2882,
+      "step": 4340
+    },
+    {
+      "epoch": 2.8,
+      "learning_rate": 0.00013387622149837132,
+      "loss": 0.2805,
+      "step": 4350
+    },
+    {
+      "epoch": 2.8,
+      "learning_rate": 0.00013348534201954395,
+      "loss": 0.2683,
+      "step": 4360
+    },
+    {
+      "epoch": 2.81,
+      "learning_rate": 0.0001330944625407166,
+      "loss": 0.2843,
+      "step": 4370
+    },
+    {
+      "epoch": 2.82,
+      "learning_rate": 0.00013270358306188924,
+      "loss": 0.2766,
+      "step": 4380
+    },
+    {
+      "epoch": 2.82,
+      "learning_rate": 0.0001323127035830619,
+      "loss": 0.2892,
+      "step": 4390
+    },
+    {
+      "epoch": 2.83,
+      "learning_rate": 0.00013192182410423452,
+      "loss": 0.2826,
+      "step": 4400
+    },
+    {
+      "epoch": 2.83,
+      "eval_loss": 0.3005044758319855,
+      "eval_runtime": 652.1777,
+      "eval_samples_per_second": 3.067,
+      "eval_steps_per_second": 0.383,
+      "step": 4400
+    },
+    {
+      "epoch": 2.84,
+      "learning_rate": 0.00013153094462540715,
+      "loss": 0.2816,
+      "step": 4410
+    },
+    {
+      "epoch": 2.84,
+      "learning_rate": 0.0001311400651465798,
+      "loss": 0.2956,
+      "step": 4420
+    },
+    {
+      "epoch": 2.85,
+      "learning_rate": 0.00013074918566775243,
+      "loss": 0.274,
+      "step": 4430
+    },
+    {
+      "epoch": 2.86,
+      "learning_rate": 0.00013035830618892506,
+      "loss": 0.2791,
+      "step": 4440
+    },
+    {
+      "epoch": 2.86,
+      "learning_rate": 0.00012996742671009771,
+      "loss": 0.2792,
+      "step": 4450
+    },
+    {
+      "epoch": 2.87,
+      "learning_rate": 0.00012957654723127034,
+      "loss": 0.292,
+      "step": 4460
+    },
+    {
+      "epoch": 2.87,
+      "learning_rate": 0.000129185667752443,
+      "loss": 0.2788,
+      "step": 4470
+    },
+    {
+      "epoch": 2.88,
+      "learning_rate": 0.00012879478827361563,
+      "loss": 0.2915,
+      "step": 4480
+    },
+    {
+      "epoch": 2.89,
+      "learning_rate": 0.00012840390879478825,
+      "loss": 0.2781,
+      "step": 4490
+    },
+    {
+      "epoch": 2.89,
+      "learning_rate": 0.0001280130293159609,
+      "loss": 0.2794,
+      "step": 4500
+    },
+    {
+      "epoch": 2.9,
+      "learning_rate": 0.00012762214983713354,
+      "loss": 0.2819,
+      "step": 4510
+    },
+    {
+      "epoch": 2.91,
+      "learning_rate": 0.0001272312703583062,
+      "loss": 0.284,
+      "step": 4520
+    },
+    {
+      "epoch": 2.91,
+      "learning_rate": 0.00012684039087947882,
+      "loss": 0.2737,
+      "step": 4530
+    },
+    {
+      "epoch": 2.92,
+      "learning_rate": 0.00012644951140065145,
+      "loss": 0.2655,
+      "step": 4540
+    },
+    {
+      "epoch": 2.93,
+      "learning_rate": 0.0001260586319218241,
+      "loss": 0.2701,
+      "step": 4550
+    },
+    {
+      "epoch": 2.93,
+      "learning_rate": 0.00012566775244299673,
+      "loss": 0.2848,
+      "step": 4560
+    },
+    {
+      "epoch": 2.94,
+      "learning_rate": 0.00012527687296416936,
+      "loss": 0.2682,
+      "step": 4570
+    },
+    {
+      "epoch": 2.95,
+      "learning_rate": 0.000124885993485342,
+      "loss": 0.2874,
+      "step": 4580
+    },
+    {
+      "epoch": 2.95,
+      "learning_rate": 0.00012449511400651464,
+      "loss": 0.2798,
+      "step": 4590
+    },
+    {
+      "epoch": 2.96,
+      "learning_rate": 0.0001241042345276873,
+      "loss": 0.284,
+      "step": 4600
+    },
+    {
+      "epoch": 2.96,
+      "eval_loss": 0.2995336055755615,
+      "eval_runtime": 652.509,
+      "eval_samples_per_second": 3.065,
+      "eval_steps_per_second": 0.383,
+      "step": 4600
+    }
+  ],
+  "logging_steps": 10,
+  "max_steps": 7775,
+  "num_train_epochs": 5,
+  "save_steps": 200,
+  "total_flos": 8.108270971276493e+18,
+  "trial_name": null,
+  "trial_params": null
+}

with_input/llama-13b/best_model/training_args.bin

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:0a20f1d68bcc87daa45f39faa69f88f7c45cebdaeb0631117cb121da43123712
+size 4091