add plan model 7b

with_input/decomp_plan/llama-2-7b/best_model/adapter_config.json

@@ -0,0 +1,21 @@
+{
+  "base_model_name_or_path": "NousResearch/Llama-2-7b-hf",
+  "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/decomp_plan/llama-2-7b/best_model/adapter_model.bin

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:30c63c970df76e6f61f548be677be5e0931caed7546a72a3b37075eb893f225c
+size 67201357

with_input/decomp_plan/llama-2-7b/best_model/optimizer.pt

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:4a4359f4abeda6c9551feaaa79fc27b19d97dbce887d95affbfb74649f546029
+size 134433093

with_input/decomp_plan/llama-2-7b/best_model/rng_state.pth

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:1d173e283bca40272a2bd6130b3240649f4c15d436953c1a112038ddcbb7c7ec
+size 14575

with_input/decomp_plan/llama-2-7b/best_model/scheduler.pt

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:abfed23e6d6d347e99d70b8c08d200a84418499ce9da89b5fe35ff4e4e832571
+size 627

with_input/decomp_plan/llama-2-7b/best_model/trainer_state.json

@@ -0,0 +1,2963 @@
+{
+  "best_metric": 0.7489736080169678,
+  "best_model_checkpoint": "checkpoints/instrucode/with_input/decomp_plan/llama-2-7b/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": 1.616,
+      "step": 10
+    },
+    {
+      "epoch": 0.01,
+      "learning_rate": 5.9999999999999995e-05,
+      "loss": 1.5251,
+      "step": 20
+    },
+    {
+      "epoch": 0.02,
+      "learning_rate": 8.999999999999999e-05,
+      "loss": 1.3375,
+      "step": 30
+    },
+    {
+      "epoch": 0.03,
+      "learning_rate": 0.00011999999999999999,
+      "loss": 1.0578,
+      "step": 40
+    },
+    {
+      "epoch": 0.03,
+      "learning_rate": 0.00015,
+      "loss": 0.9937,
+      "step": 50
+    },
+    {
+      "epoch": 0.04,
+      "learning_rate": 0.00017999999999999998,
+      "loss": 0.9623,
+      "step": 60
+    },
+    {
+      "epoch": 0.05,
+      "learning_rate": 0.00020999999999999998,
+      "loss": 0.9619,
+      "step": 70
+    },
+    {
+      "epoch": 0.05,
+      "learning_rate": 0.00023999999999999998,
+      "loss": 0.8903,
+      "step": 80
+    },
+    {
+      "epoch": 0.06,
+      "learning_rate": 0.00027,
+      "loss": 0.872,
+      "step": 90
+    },
+    {
+      "epoch": 0.06,
+      "learning_rate": 0.0003,
+      "loss": 0.873,
+      "step": 100
+    },
+    {
+      "epoch": 0.07,
+      "learning_rate": 0.00029960912052117263,
+      "loss": 0.8698,
+      "step": 110
+    },
+    {
+      "epoch": 0.08,
+      "learning_rate": 0.0002992182410423453,
+      "loss": 0.8579,
+      "step": 120
+    },
+    {
+      "epoch": 0.08,
+      "learning_rate": 0.0002988273615635179,
+      "loss": 0.8908,
+      "step": 130
+    },
+    {
+      "epoch": 0.09,
+      "learning_rate": 0.00029843648208469054,
+      "loss": 0.8485,
+      "step": 140
+    },
+    {
+      "epoch": 0.1,
+      "learning_rate": 0.00029804560260586314,
+      "loss": 0.8484,
+      "step": 150
+    },
+    {
+      "epoch": 0.1,
+      "learning_rate": 0.0002976547231270358,
+      "loss": 0.8501,
+      "step": 160
+    },
+    {
+      "epoch": 0.11,
+      "learning_rate": 0.00029726384364820845,
+      "loss": 0.8702,
+      "step": 170
+    },
+    {
+      "epoch": 0.12,
+      "learning_rate": 0.00029687296416938105,
+      "loss": 0.8489,
+      "step": 180
+    },
+    {
+      "epoch": 0.12,
+      "learning_rate": 0.0002964820846905537,
+      "loss": 0.8231,
+      "step": 190
+    },
+    {
+      "epoch": 0.13,
+      "learning_rate": 0.00029609120521172636,
+      "loss": 0.8268,
+      "step": 200
+    },
+    {
+      "epoch": 0.13,
+      "eval_loss": 0.8351926207542419,
+      "eval_runtime": 184.9393,
+      "eval_samples_per_second": 10.814,
+      "eval_steps_per_second": 1.352,
+      "step": 200
+    },
+    {
+      "epoch": 0.14,
+      "learning_rate": 0.000295700325732899,
+      "loss": 0.831,
+      "step": 210
+    },
+    {
+      "epoch": 0.14,
+      "learning_rate": 0.0002953094462540716,
+      "loss": 0.8504,
+      "step": 220
+    },
+    {
+      "epoch": 0.15,
+      "learning_rate": 0.0002949185667752443,
+      "loss": 0.8159,
+      "step": 230
+    },
+    {
+      "epoch": 0.15,
+      "learning_rate": 0.00029452768729641693,
+      "loss": 0.8278,
+      "step": 240
+    },
+    {
+      "epoch": 0.16,
+      "learning_rate": 0.00029413680781758953,
+      "loss": 0.8202,
+      "step": 250
+    },
+    {
+      "epoch": 0.17,
+      "learning_rate": 0.0002937459283387622,
+      "loss": 0.8194,
+      "step": 260
+    },
+    {
+      "epoch": 0.17,
+      "learning_rate": 0.00029335504885993484,
+      "loss": 0.8035,
+      "step": 270
+    },
+    {
+      "epoch": 0.18,
+      "learning_rate": 0.00029296416938110744,
+      "loss": 0.8088,
+      "step": 280
+    },
+    {
+      "epoch": 0.19,
+      "learning_rate": 0.0002925732899022801,
+      "loss": 0.8261,
+      "step": 290
+    },
+    {
+      "epoch": 0.19,
+      "learning_rate": 0.00029218241042345275,
+      "loss": 0.8228,
+      "step": 300
+    },
+    {
+      "epoch": 0.2,
+      "learning_rate": 0.0002917915309446254,
+      "loss": 0.8105,
+      "step": 310
+    },
+    {
+      "epoch": 0.21,
+      "learning_rate": 0.000291400651465798,
+      "loss": 0.8173,
+      "step": 320
+    },
+    {
+      "epoch": 0.21,
+      "learning_rate": 0.00029100977198697066,
+      "loss": 0.8167,
+      "step": 330
+    },
+    {
+      "epoch": 0.22,
+      "learning_rate": 0.00029061889250814327,
+      "loss": 0.8196,
+      "step": 340
+    },
+    {
+      "epoch": 0.23,
+      "learning_rate": 0.0002902280130293159,
+      "loss": 0.8208,
+      "step": 350
+    },
+    {
+      "epoch": 0.23,
+      "learning_rate": 0.0002898371335504886,
+      "loss": 0.8112,
+      "step": 360
+    },
+    {
+      "epoch": 0.24,
+      "learning_rate": 0.00028944625407166123,
+      "loss": 0.7977,
+      "step": 370
+    },
+    {
+      "epoch": 0.24,
+      "learning_rate": 0.0002890553745928339,
+      "loss": 0.8114,
+      "step": 380
+    },
+    {
+      "epoch": 0.25,
+      "learning_rate": 0.0002886644951140065,
+      "loss": 0.8101,
+      "step": 390
+    },
+    {
+      "epoch": 0.26,
+      "learning_rate": 0.00028827361563517914,
+      "loss": 0.8098,
+      "step": 400
+    },
+    {
+      "epoch": 0.26,
+      "eval_loss": 0.8081860542297363,
+      "eval_runtime": 185.1672,
+      "eval_samples_per_second": 10.801,
+      "eval_steps_per_second": 1.35,
+      "step": 400
+    },
+    {
+      "epoch": 0.26,
+      "learning_rate": 0.00028788273615635174,
+      "loss": 0.8183,
+      "step": 410
+    },
+    {
+      "epoch": 0.27,
+      "learning_rate": 0.0002874918566775244,
+      "loss": 0.8198,
+      "step": 420
+    },
+    {
+      "epoch": 0.28,
+      "learning_rate": 0.00028710097719869705,
+      "loss": 0.8108,
+      "step": 430
+    },
+    {
+      "epoch": 0.28,
+      "learning_rate": 0.00028671009771986966,
+      "loss": 0.8164,
+      "step": 440
+    },
+    {
+      "epoch": 0.29,
+      "learning_rate": 0.0002863192182410423,
+      "loss": 0.8111,
+      "step": 450
+    },
+    {
+      "epoch": 0.3,
+      "learning_rate": 0.00028592833876221497,
+      "loss": 0.7942,
+      "step": 460
+    },
+    {
+      "epoch": 0.3,
+      "learning_rate": 0.0002855374592833876,
+      "loss": 0.8058,
+      "step": 470
+    },
+    {
+      "epoch": 0.31,
+      "learning_rate": 0.0002851465798045602,
+      "loss": 0.7998,
+      "step": 480
+    },
+    {
+      "epoch": 0.32,
+      "learning_rate": 0.0002847557003257329,
+      "loss": 0.8095,
+      "step": 490
+    },
+    {
+      "epoch": 0.32,
+      "learning_rate": 0.0002843648208469055,
+      "loss": 0.8147,
+      "step": 500
+    },
+    {
+      "epoch": 0.33,
+      "learning_rate": 0.00028397394136807813,
+      "loss": 0.8131,
+      "step": 510
+    },
+    {
+      "epoch": 0.33,
+      "learning_rate": 0.0002835830618892508,
+      "loss": 0.7979,
+      "step": 520
+    },
+    {
+      "epoch": 0.34,
+      "learning_rate": 0.00028319218241042344,
+      "loss": 0.8035,
+      "step": 530
+    },
+    {
+      "epoch": 0.35,
+      "learning_rate": 0.00028280130293159605,
+      "loss": 0.7957,
+      "step": 540
+    },
+    {
+      "epoch": 0.35,
+      "learning_rate": 0.0002824104234527687,
+      "loss": 0.8185,
+      "step": 550
+    },
+    {
+      "epoch": 0.36,
+      "learning_rate": 0.00028201954397394136,
+      "loss": 0.7836,
+      "step": 560
+    },
+    {
+      "epoch": 0.37,
+      "learning_rate": 0.000281628664495114,
+      "loss": 0.8297,
+      "step": 570
+    },
+    {
+      "epoch": 0.37,
+      "learning_rate": 0.0002812377850162866,
+      "loss": 0.7929,
+      "step": 580
+    },
+    {
+      "epoch": 0.38,
+      "learning_rate": 0.00028084690553745927,
+      "loss": 0.7956,
+      "step": 590
+    },
+    {
+      "epoch": 0.39,
+      "learning_rate": 0.00028045602605863187,
+      "loss": 0.7778,
+      "step": 600
+    },
+    {
+      "epoch": 0.39,
+      "eval_loss": 0.7940636873245239,
+      "eval_runtime": 185.1799,
+      "eval_samples_per_second": 10.8,
+      "eval_steps_per_second": 1.35,
+      "step": 600
+    },
+    {
+      "epoch": 0.39,
+      "learning_rate": 0.0002800651465798045,
+      "loss": 0.795,
+      "step": 610
+    },
+    {
+      "epoch": 0.4,
+      "learning_rate": 0.0002796742671009772,
+      "loss": 0.7999,
+      "step": 620
+    },
+    {
+      "epoch": 0.41,
+      "learning_rate": 0.00027928338762214983,
+      "loss": 0.7887,
+      "step": 630
+    },
+    {
+      "epoch": 0.41,
+      "learning_rate": 0.00027889250814332244,
+      "loss": 0.7765,
+      "step": 640
+    },
+    {
+      "epoch": 0.42,
+      "learning_rate": 0.0002785016286644951,
+      "loss": 0.7885,
+      "step": 650
+    },
+    {
+      "epoch": 0.42,
+      "learning_rate": 0.00027811074918566775,
+      "loss": 0.7847,
+      "step": 660
+    },
+    {
+      "epoch": 0.43,
+      "learning_rate": 0.00027771986970684035,
+      "loss": 0.7928,
+      "step": 670
+    },
+    {
+      "epoch": 0.44,
+      "learning_rate": 0.000277328990228013,
+      "loss": 0.768,
+      "step": 680
+    },
+    {
+      "epoch": 0.44,
+      "learning_rate": 0.00027693811074918566,
+      "loss": 0.7943,
+      "step": 690
+    },
+    {
+      "epoch": 0.45,
+      "learning_rate": 0.00027654723127035826,
+      "loss": 0.7936,
+      "step": 700
+    },
+    {
+      "epoch": 0.46,
+      "learning_rate": 0.0002761563517915309,
+      "loss": 0.7887,
+      "step": 710
+    },
+    {
+      "epoch": 0.46,
+      "learning_rate": 0.00027576547231270357,
+      "loss": 0.8029,
+      "step": 720
+    },
+    {
+      "epoch": 0.47,
+      "learning_rate": 0.0002753745928338762,
+      "loss": 0.7891,
+      "step": 730
+    },
+    {
+      "epoch": 0.48,
+      "learning_rate": 0.0002749837133550488,
+      "loss": 0.7803,
+      "step": 740
+    },
+    {
+      "epoch": 0.48,
+      "learning_rate": 0.0002745928338762215,
+      "loss": 0.7781,
+      "step": 750
+    },
+    {
+      "epoch": 0.49,
+      "learning_rate": 0.0002742019543973941,
+      "loss": 0.8151,
+      "step": 760
+    },
+    {
+      "epoch": 0.5,
+      "learning_rate": 0.00027381107491856674,
+      "loss": 0.7623,
+      "step": 770
+    },
+    {
+      "epoch": 0.5,
+      "learning_rate": 0.0002734201954397394,
+      "loss": 0.7673,
+      "step": 780
+    },
+    {
+      "epoch": 0.51,
+      "learning_rate": 0.00027302931596091205,
+      "loss": 0.7691,
+      "step": 790
+    },
+    {
+      "epoch": 0.51,
+      "learning_rate": 0.00027263843648208465,
+      "loss": 0.8055,
+      "step": 800
+    },
+    {
+      "epoch": 0.51,
+      "eval_loss": 0.7852708697319031,
+      "eval_runtime": 185.5285,
+      "eval_samples_per_second": 10.78,
+      "eval_steps_per_second": 1.348,
+      "step": 800
+    },
+    {
+      "epoch": 0.52,
+      "learning_rate": 0.0002722475570032573,
+      "loss": 0.7996,
+      "step": 810
+    },
+    {
+      "epoch": 0.53,
+      "learning_rate": 0.00027185667752442996,
+      "loss": 0.7933,
+      "step": 820
+    },
+    {
+      "epoch": 0.53,
+      "learning_rate": 0.0002714657980456026,
+      "loss": 0.7955,
+      "step": 830
+    },
+    {
+      "epoch": 0.54,
+      "learning_rate": 0.0002710749185667752,
+      "loss": 0.787,
+      "step": 840
+    },
+    {
+      "epoch": 0.55,
+      "learning_rate": 0.00027068403908794787,
+      "loss": 0.772,
+      "step": 850
+    },
+    {
+      "epoch": 0.55,
+      "learning_rate": 0.00027029315960912047,
+      "loss": 0.7873,
+      "step": 860
+    },
+    {
+      "epoch": 0.56,
+      "learning_rate": 0.0002699022801302931,
+      "loss": 0.7965,
+      "step": 870
+    },
+    {
+      "epoch": 0.57,
+      "learning_rate": 0.0002695114006514658,
+      "loss": 0.7966,
+      "step": 880
+    },
+    {
+      "epoch": 0.57,
+      "learning_rate": 0.00026912052117263844,
+      "loss": 0.7822,
+      "step": 890
+    },
+    {
+      "epoch": 0.58,
+      "learning_rate": 0.00026872964169381104,
+      "loss": 0.7949,
+      "step": 900
+    },
+    {
+      "epoch": 0.59,
+      "learning_rate": 0.0002683387622149837,
+      "loss": 0.7949,
+      "step": 910
+    },
+    {
+      "epoch": 0.59,
+      "learning_rate": 0.00026794788273615635,
+      "loss": 0.7648,
+      "step": 920
+    },
+    {
+      "epoch": 0.6,
+      "learning_rate": 0.00026755700325732895,
+      "loss": 0.7895,
+      "step": 930
+    },
+    {
+      "epoch": 0.6,
+      "learning_rate": 0.0002671661237785016,
+      "loss": 0.7815,
+      "step": 940
+    },
+    {
+      "epoch": 0.61,
+      "learning_rate": 0.0002667752442996742,
+      "loss": 0.7917,
+      "step": 950
+    },
+    {
+      "epoch": 0.62,
+      "learning_rate": 0.00026638436482084686,
+      "loss": 0.7845,
+      "step": 960
+    },
+    {
+      "epoch": 0.62,
+      "learning_rate": 0.0002659934853420195,
+      "loss": 0.7969,
+      "step": 970
+    },
+    {
+      "epoch": 0.63,
+      "learning_rate": 0.00026560260586319217,
+      "loss": 0.7837,
+      "step": 980
+    },
+    {
+      "epoch": 0.64,
+      "learning_rate": 0.0002652117263843648,
+      "loss": 0.7791,
+      "step": 990
+    },
+    {
+      "epoch": 0.64,
+      "learning_rate": 0.00026482084690553743,
+      "loss": 0.7949,
+      "step": 1000
+    },
+    {
+      "epoch": 0.64,
+      "eval_loss": 0.7794002890586853,
+      "eval_runtime": 185.3123,
+      "eval_samples_per_second": 10.793,
+      "eval_steps_per_second": 1.349,
+      "step": 1000
+    },
+    {
+      "epoch": 0.65,
+      "learning_rate": 0.0002644299674267101,
+      "loss": 0.8007,
+      "step": 1010
+    },
+    {
+      "epoch": 0.66,
+      "learning_rate": 0.0002640390879478827,
+      "loss": 0.7807,
+      "step": 1020
+    },
+    {
+      "epoch": 0.66,
+      "learning_rate": 0.00026364820846905534,
+      "loss": 0.798,
+      "step": 1030
+    },
+    {
+      "epoch": 0.67,
+      "learning_rate": 0.000263257328990228,
+      "loss": 0.7893,
+      "step": 1040
+    },
+    {
+      "epoch": 0.68,
+      "learning_rate": 0.0002628664495114006,
+      "loss": 0.7684,
+      "step": 1050
+    },
+    {
+      "epoch": 0.68,
+      "learning_rate": 0.00026247557003257325,
+      "loss": 0.7835,
+      "step": 1060
+    },
+    {
+      "epoch": 0.69,
+      "learning_rate": 0.0002620846905537459,
+      "loss": 0.7534,
+      "step": 1070
+    },
+    {
+      "epoch": 0.69,
+      "learning_rate": 0.00026169381107491856,
+      "loss": 0.7922,
+      "step": 1080
+    },
+    {
+      "epoch": 0.7,
+      "learning_rate": 0.0002613029315960912,
+      "loss": 0.7851,
+      "step": 1090
+    },
+    {
+      "epoch": 0.71,
+      "learning_rate": 0.0002609120521172638,
+      "loss": 0.7935,
+      "step": 1100
+    },
+    {
+      "epoch": 0.71,
+      "learning_rate": 0.0002605211726384365,
+      "loss": 0.8018,
+      "step": 1110
+    },
+    {
+      "epoch": 0.72,
+      "learning_rate": 0.0002601302931596091,
+      "loss": 0.7629,
+      "step": 1120
+    },
+    {
+      "epoch": 0.73,
+      "learning_rate": 0.00025973941368078173,
+      "loss": 0.7813,
+      "step": 1130
+    },
+    {
+      "epoch": 0.73,
+      "learning_rate": 0.0002593485342019544,
+      "loss": 0.7662,
+      "step": 1140
+    },
+    {
+      "epoch": 0.74,
+      "learning_rate": 0.00025895765472312704,
+      "loss": 0.761,
+      "step": 1150
+    },
+    {
+      "epoch": 0.75,
+      "learning_rate": 0.00025856677524429964,
+      "loss": 0.7757,
+      "step": 1160
+    },
+    {
+      "epoch": 0.75,
+      "learning_rate": 0.0002581758957654723,
+      "loss": 0.7695,
+      "step": 1170
+    },
+    {
+      "epoch": 0.76,
+      "learning_rate": 0.00025778501628664495,
+      "loss": 0.7699,
+      "step": 1180
+    },
+    {
+      "epoch": 0.77,
+      "learning_rate": 0.00025739413680781755,
+      "loss": 0.7843,
+      "step": 1190
+    },
+    {
+      "epoch": 0.77,
+      "learning_rate": 0.0002570032573289902,
+      "loss": 0.774,
+      "step": 1200
+    },
+    {
+      "epoch": 0.77,
+      "eval_loss": 0.7737606167793274,
+      "eval_runtime": 185.305,
+      "eval_samples_per_second": 10.793,
+      "eval_steps_per_second": 1.349,
+      "step": 1200
+    },
+    {
+      "epoch": 0.78,
+      "learning_rate": 0.0002566123778501628,
+      "loss": 0.7607,
+      "step": 1210
+    },
+    {
+      "epoch": 0.78,
+      "learning_rate": 0.00025622149837133546,
+      "loss": 0.7681,
+      "step": 1220
+    },
+    {
+      "epoch": 0.79,
+      "learning_rate": 0.0002558306188925081,
+      "loss": 0.7774,
+      "step": 1230
+    },
+    {
+      "epoch": 0.8,
+      "learning_rate": 0.0002554397394136808,
+      "loss": 0.7837,
+      "step": 1240
+    },
+    {
+      "epoch": 0.8,
+      "learning_rate": 0.00025504885993485343,
+      "loss": 0.7869,
+      "step": 1250
+    },
+    {
+      "epoch": 0.81,
+      "learning_rate": 0.00025465798045602603,
+      "loss": 0.7491,
+      "step": 1260
+    },
+    {
+      "epoch": 0.82,
+      "learning_rate": 0.0002542671009771987,
+      "loss": 0.7723,
+      "step": 1270
+    },
+    {
+      "epoch": 0.82,
+      "learning_rate": 0.0002538762214983713,
+      "loss": 0.7748,
+      "step": 1280
+    },
+    {
+      "epoch": 0.83,
+      "learning_rate": 0.00025348534201954394,
+      "loss": 0.7709,
+      "step": 1290
+    },
+    {
+      "epoch": 0.84,
+      "learning_rate": 0.0002530944625407166,
+      "loss": 0.7809,
+      "step": 1300
+    },
+    {
+      "epoch": 0.84,
+      "learning_rate": 0.0002527035830618892,
+      "loss": 0.7487,
+      "step": 1310
+    },
+    {
+      "epoch": 0.85,
+      "learning_rate": 0.00025231270358306185,
+      "loss": 0.7925,
+      "step": 1320
+    },
+    {
+      "epoch": 0.86,
+      "learning_rate": 0.0002519218241042345,
+      "loss": 0.7713,
+      "step": 1330
+    },
+    {
+      "epoch": 0.86,
+      "learning_rate": 0.00025153094462540716,
+      "loss": 0.7681,
+      "step": 1340
+    },
+    {
+      "epoch": 0.87,
+      "learning_rate": 0.0002511400651465798,
+      "loss": 0.7546,
+      "step": 1350
+    },
+    {
+      "epoch": 0.87,
+      "learning_rate": 0.0002507491856677524,
+      "loss": 0.7953,
+      "step": 1360
+    },
+    {
+      "epoch": 0.88,
+      "learning_rate": 0.0002503583061889251,
+      "loss": 0.7559,
+      "step": 1370
+    },
+    {
+      "epoch": 0.89,
+      "learning_rate": 0.0002499674267100977,
+      "loss": 0.7703,
+      "step": 1380
+    },
+    {
+      "epoch": 0.89,
+      "learning_rate": 0.00024957654723127033,
+      "loss": 0.7867,
+      "step": 1390
+    },
+    {
+      "epoch": 0.9,
+      "learning_rate": 0.000249185667752443,
+      "loss": 0.7526,
+      "step": 1400
+    },
+    {
+      "epoch": 0.9,
+      "eval_loss": 0.7698599696159363,
+      "eval_runtime": 186.0927,
+      "eval_samples_per_second": 10.747,
+      "eval_steps_per_second": 1.343,
+      "step": 1400
+    },
+    {
+      "epoch": 0.91,
+      "learning_rate": 0.0002487947882736156,
+      "loss": 0.78,
+      "step": 1410
+    },
+    {
+      "epoch": 0.91,
+      "learning_rate": 0.00024840390879478824,
+      "loss": 0.7691,
+      "step": 1420
+    },
+    {
+      "epoch": 0.92,
+      "learning_rate": 0.0002480130293159609,
+      "loss": 0.7588,
+      "step": 1430
+    },
+    {
+      "epoch": 0.93,
+      "learning_rate": 0.00024762214983713355,
+      "loss": 0.764,
+      "step": 1440
+    },
+    {
+      "epoch": 0.93,
+      "learning_rate": 0.00024723127035830616,
+      "loss": 0.7969,
+      "step": 1450
+    },
+    {
+      "epoch": 0.94,
+      "learning_rate": 0.0002468403908794788,
+      "loss": 0.7578,
+      "step": 1460
+    },
+    {
+      "epoch": 0.95,
+      "learning_rate": 0.0002464495114006514,
+      "loss": 0.7426,
+      "step": 1470
+    },
+    {
+      "epoch": 0.95,
+      "learning_rate": 0.00024605863192182407,
+      "loss": 0.7909,
+      "step": 1480
+    },
+    {
+      "epoch": 0.96,
+      "learning_rate": 0.0002456677524429967,
+      "loss": 0.7635,
+      "step": 1490
+    },
+    {
+      "epoch": 0.96,
+      "learning_rate": 0.0002452768729641694,
+      "loss": 0.7539,
+      "step": 1500
+    },
+    {
+      "epoch": 0.97,
+      "learning_rate": 0.00024488599348534203,
+      "loss": 0.7742,
+      "step": 1510
+    },
+    {
+      "epoch": 0.98,
+      "learning_rate": 0.00024449511400651463,
+      "loss": 0.7637,
+      "step": 1520
+    },
+    {
+      "epoch": 0.98,
+      "learning_rate": 0.00024410423452768726,
+      "loss": 0.7846,
+      "step": 1530
+    },
+    {
+      "epoch": 0.99,
+      "learning_rate": 0.00024371335504885992,
+      "loss": 0.7723,
+      "step": 1540
+    },
+    {
+      "epoch": 1.0,
+      "learning_rate": 0.00024332247557003257,
+      "loss": 0.7525,
+      "step": 1550
+    },
+    {
+      "epoch": 1.0,
+      "learning_rate": 0.0002429315960912052,
+      "loss": 0.7614,
+      "step": 1560
+    },
+    {
+      "epoch": 1.01,
+      "learning_rate": 0.00024254071661237783,
+      "loss": 0.758,
+      "step": 1570
+    },
+    {
+      "epoch": 1.02,
+      "learning_rate": 0.00024214983713355046,
+      "loss": 0.7427,
+      "step": 1580
+    },
+    {
+      "epoch": 1.02,
+      "learning_rate": 0.0002417589576547231,
+      "loss": 0.7309,
+      "step": 1590
+    },
+    {
+      "epoch": 1.03,
+      "learning_rate": 0.00024136807817589574,
+      "loss": 0.7298,
+      "step": 1600
+    },
+    {
+      "epoch": 1.03,
+      "eval_loss": 0.7665953040122986,
+      "eval_runtime": 185.4725,
+      "eval_samples_per_second": 10.783,
+      "eval_steps_per_second": 1.348,
+      "step": 1600
+    },
+    {
+      "epoch": 1.04,
+      "learning_rate": 0.0002409771986970684,
+      "loss": 0.7318,
+      "step": 1610
+    },
+    {
+      "epoch": 1.04,
+      "learning_rate": 0.000240586319218241,
+      "loss": 0.7349,
+      "step": 1620
+    },
+    {
+      "epoch": 1.05,
+      "learning_rate": 0.00024019543973941365,
+      "loss": 0.735,
+      "step": 1630
+    },
+    {
+      "epoch": 1.05,
+      "learning_rate": 0.0002398045602605863,
+      "loss": 0.7397,
+      "step": 1640
+    },
+    {
+      "epoch": 1.06,
+      "learning_rate": 0.00023941368078175893,
+      "loss": 0.7393,
+      "step": 1650
+    },
+    {
+      "epoch": 1.07,
+      "learning_rate": 0.0002390228013029316,
+      "loss": 0.751,
+      "step": 1660
+    },
+    {
+      "epoch": 1.07,
+      "learning_rate": 0.0002386319218241042,
+      "loss": 0.7342,
+      "step": 1670
+    },
+    {
+      "epoch": 1.08,
+      "learning_rate": 0.00023824104234527685,
+      "loss": 0.747,
+      "step": 1680
+    },
+    {
+      "epoch": 1.09,
+      "learning_rate": 0.0002378501628664495,
+      "loss": 0.7433,
+      "step": 1690
+    },
+    {
+      "epoch": 1.09,
+      "learning_rate": 0.00023745928338762213,
+      "loss": 0.7217,
+      "step": 1700
+    },
+    {
+      "epoch": 1.1,
+      "learning_rate": 0.00023706840390879478,
+      "loss": 0.7547,
+      "step": 1710
+    },
+    {
+      "epoch": 1.11,
+      "learning_rate": 0.00023667752442996739,
+      "loss": 0.7318,
+      "step": 1720
+    },
+    {
+      "epoch": 1.11,
+      "learning_rate": 0.00023628664495114004,
+      "loss": 0.7477,
+      "step": 1730
+    },
+    {
+      "epoch": 1.12,
+      "learning_rate": 0.00023589576547231267,
+      "loss": 0.7219,
+      "step": 1740
+    },
+    {
+      "epoch": 1.13,
+      "learning_rate": 0.00023550488599348532,
+      "loss": 0.721,
+      "step": 1750
+    },
+    {
+      "epoch": 1.13,
+      "learning_rate": 0.00023511400651465798,
+      "loss": 0.7286,
+      "step": 1760
+    },
+    {
+      "epoch": 1.14,
+      "learning_rate": 0.00023472312703583058,
+      "loss": 0.7326,
+      "step": 1770
+    },
+    {
+      "epoch": 1.14,
+      "learning_rate": 0.00023433224755700324,
+      "loss": 0.7373,
+      "step": 1780
+    },
+    {
+      "epoch": 1.15,
+      "learning_rate": 0.00023394136807817586,
+      "loss": 0.7442,
+      "step": 1790
+    },
+    {
+      "epoch": 1.16,
+      "learning_rate": 0.00023355048859934852,
+      "loss": 0.7191,
+      "step": 1800
+    },
+    {
+      "epoch": 1.16,
+      "eval_loss": 0.7630789279937744,
+      "eval_runtime": 186.1873,
+      "eval_samples_per_second": 10.742,
+      "eval_steps_per_second": 1.343,
+      "step": 1800
+    },
+    {
+      "epoch": 1.16,
+      "learning_rate": 0.00023315960912052117,
+      "loss": 0.7362,
+      "step": 1810
+    },
+    {
+      "epoch": 1.17,
+      "learning_rate": 0.00023276872964169378,
+      "loss": 0.7387,
+      "step": 1820
+    },
+    {
+      "epoch": 1.18,
+      "learning_rate": 0.00023237785016286643,
+      "loss": 0.7343,
+      "step": 1830
+    },
+    {
+      "epoch": 1.18,
+      "learning_rate": 0.00023198697068403906,
+      "loss": 0.7117,
+      "step": 1840
+    },
+    {
+      "epoch": 1.19,
+      "learning_rate": 0.00023159609120521171,
+      "loss": 0.757,
+      "step": 1850
+    },
+    {
+      "epoch": 1.2,
+      "learning_rate": 0.00023120521172638434,
+      "loss": 0.7408,
+      "step": 1860
+    },
+    {
+      "epoch": 1.2,
+      "learning_rate": 0.000230814332247557,
+      "loss": 0.7424,
+      "step": 1870
+    },
+    {
+      "epoch": 1.21,
+      "learning_rate": 0.0002304234527687296,
+      "loss": 0.7284,
+      "step": 1880
+    },
+    {
+      "epoch": 1.22,
+      "learning_rate": 0.00023003257328990225,
+      "loss": 0.7552,
+      "step": 1890
+    },
+    {
+      "epoch": 1.22,
+      "learning_rate": 0.0002296416938110749,
+      "loss": 0.7368,
+      "step": 1900
+    },
+    {
+      "epoch": 1.23,
+      "learning_rate": 0.00022925081433224754,
+      "loss": 0.7248,
+      "step": 1910
+    },
+    {
+      "epoch": 1.23,
+      "learning_rate": 0.0002288599348534202,
+      "loss": 0.7228,
+      "step": 1920
+    },
+    {
+      "epoch": 1.24,
+      "learning_rate": 0.0002284690553745928,
+      "loss": 0.7206,
+      "step": 1930
+    },
+    {
+      "epoch": 1.25,
+      "learning_rate": 0.00022807817589576545,
+      "loss": 0.718,
+      "step": 1940
+    },
+    {
+      "epoch": 1.25,
+      "learning_rate": 0.0002276872964169381,
+      "loss": 0.7415,
+      "step": 1950
+    },
+    {
+      "epoch": 1.26,
+      "learning_rate": 0.00022729641693811073,
+      "loss": 0.724,
+      "step": 1960
+    },
+    {
+      "epoch": 1.27,
+      "learning_rate": 0.0002269055374592834,
+      "loss": 0.7408,
+      "step": 1970
+    },
+    {
+      "epoch": 1.27,
+      "learning_rate": 0.000226514657980456,
+      "loss": 0.7533,
+      "step": 1980
+    },
+    {
+      "epoch": 1.28,
+      "learning_rate": 0.00022612377850162864,
+      "loss": 0.7462,
+      "step": 1990
+    },
+    {
+      "epoch": 1.29,
+      "learning_rate": 0.00022573289902280127,
+      "loss": 0.7,
+      "step": 2000
+    },
+    {
+      "epoch": 1.29,
+      "eval_loss": 0.7622774243354797,
+      "eval_runtime": 185.2663,
+      "eval_samples_per_second": 10.795,
+      "eval_steps_per_second": 1.349,
+      "step": 2000
+    },
+    {
+      "epoch": 1.29,
+      "learning_rate": 0.00022534201954397393,
+      "loss": 0.7393,
+      "step": 2010
+    },
+    {
+      "epoch": 1.3,
+      "learning_rate": 0.00022495114006514658,
+      "loss": 0.7304,
+      "step": 2020
+    },
+    {
+      "epoch": 1.31,
+      "learning_rate": 0.00022456026058631918,
+      "loss": 0.7421,
+      "step": 2030
+    },
+    {
+      "epoch": 1.31,
+      "learning_rate": 0.00022416938110749184,
+      "loss": 0.7484,
+      "step": 2040
+    },
+    {
+      "epoch": 1.32,
+      "learning_rate": 0.00022377850162866447,
+      "loss": 0.7157,
+      "step": 2050
+    },
+    {
+      "epoch": 1.32,
+      "learning_rate": 0.00022338762214983712,
+      "loss": 0.7477,
+      "step": 2060
+    },
+    {
+      "epoch": 1.33,
+      "learning_rate": 0.00022299674267100978,
+      "loss": 0.7533,
+      "step": 2070
+    },
+    {
+      "epoch": 1.34,
+      "learning_rate": 0.00022260586319218238,
+      "loss": 0.732,
+      "step": 2080
+    },
+    {
+      "epoch": 1.34,
+      "learning_rate": 0.00022221498371335503,
+      "loss": 0.7661,
+      "step": 2090
+    },
+    {
+      "epoch": 1.35,
+      "learning_rate": 0.00022182410423452766,
+      "loss": 0.7477,
+      "step": 2100
+    },
+    {
+      "epoch": 1.36,
+      "learning_rate": 0.00022143322475570032,
+      "loss": 0.7431,
+      "step": 2110
+    },
+    {
+      "epoch": 1.36,
+      "learning_rate": 0.00022104234527687295,
+      "loss": 0.734,
+      "step": 2120
+    },
+    {
+      "epoch": 1.37,
+      "learning_rate": 0.00022065146579804557,
+      "loss": 0.7444,
+      "step": 2130
+    },
+    {
+      "epoch": 1.38,
+      "learning_rate": 0.0002202605863192182,
+      "loss": 0.738,
+      "step": 2140
+    },
+    {
+      "epoch": 1.38,
+      "learning_rate": 0.00021986970684039086,
+      "loss": 0.7391,
+      "step": 2150
+    },
+    {
+      "epoch": 1.39,
+      "learning_rate": 0.0002194788273615635,
+      "loss": 0.7383,
+      "step": 2160
+    },
+    {
+      "epoch": 1.4,
+      "learning_rate": 0.00021908794788273614,
+      "loss": 0.7176,
+      "step": 2170
+    },
+    {
+      "epoch": 1.4,
+      "learning_rate": 0.00021873615635179151,
+      "loss": 0.747,
+      "step": 2180
+    },
+    {
+      "epoch": 1.41,
+      "learning_rate": 0.00021834527687296414,
+      "loss": 0.751,
+      "step": 2190
+    },
+    {
+      "epoch": 1.41,
+      "learning_rate": 0.00021795439739413677,
+      "loss": 0.7393,
+      "step": 2200
+    },
+    {
+      "epoch": 1.41,
+      "eval_loss": 0.7590416073799133,
+      "eval_runtime": 185.4936,
+      "eval_samples_per_second": 10.782,
+      "eval_steps_per_second": 1.348,
+      "step": 2200
+    },
+    {
+      "epoch": 1.42,
+      "learning_rate": 0.00021756351791530943,
+      "loss": 0.7333,
+      "step": 2210
+    },
+    {
+      "epoch": 1.43,
+      "learning_rate": 0.00021717263843648208,
+      "loss": 0.7235,
+      "step": 2220
+    },
+    {
+      "epoch": 1.43,
+      "learning_rate": 0.0002167817589576547,
+      "loss": 0.7544,
+      "step": 2230
+    },
+    {
+      "epoch": 1.44,
+      "learning_rate": 0.00021639087947882734,
+      "loss": 0.7369,
+      "step": 2240
+    },
+    {
+      "epoch": 1.45,
+      "learning_rate": 0.00021599999999999996,
+      "loss": 0.7382,
+      "step": 2250
+    },
+    {
+      "epoch": 1.45,
+      "learning_rate": 0.00021560912052117262,
+      "loss": 0.75,
+      "step": 2260
+    },
+    {
+      "epoch": 1.46,
+      "learning_rate": 0.00021521824104234528,
+      "loss": 0.7474,
+      "step": 2270
+    },
+    {
+      "epoch": 1.47,
+      "learning_rate": 0.0002148273615635179,
+      "loss": 0.73,
+      "step": 2280
+    },
+    {
+      "epoch": 1.47,
+      "learning_rate": 0.00021443648208469053,
+      "loss": 0.7401,
+      "step": 2290
+    },
+    {
+      "epoch": 1.48,
+      "learning_rate": 0.00021404560260586316,
+      "loss": 0.7163,
+      "step": 2300
+    },
+    {
+      "epoch": 1.49,
+      "learning_rate": 0.00021365472312703581,
+      "loss": 0.7404,
+      "step": 2310
+    },
+    {
+      "epoch": 1.49,
+      "learning_rate": 0.00021326384364820844,
+      "loss": 0.7339,
+      "step": 2320
+    },
+    {
+      "epoch": 1.5,
+      "learning_rate": 0.0002128729641693811,
+      "loss": 0.7373,
+      "step": 2330
+    },
+    {
+      "epoch": 1.5,
+      "learning_rate": 0.0002124820846905537,
+      "loss": 0.7026,
+      "step": 2340
+    },
+    {
+      "epoch": 1.51,
+      "learning_rate": 0.00021209120521172635,
+      "loss": 0.7349,
+      "step": 2350
+    },
+    {
+      "epoch": 1.52,
+      "learning_rate": 0.000211700325732899,
+      "loss": 0.7561,
+      "step": 2360
+    },
+    {
+      "epoch": 1.52,
+      "learning_rate": 0.00021130944625407164,
+      "loss": 0.7258,
+      "step": 2370
+    },
+    {
+      "epoch": 1.53,
+      "learning_rate": 0.0002109185667752443,
+      "loss": 0.7309,
+      "step": 2380
+    },
+    {
+      "epoch": 1.54,
+      "learning_rate": 0.00021052768729641695,
+      "loss": 0.7343,
+      "step": 2390
+    },
+    {
+      "epoch": 1.54,
+      "learning_rate": 0.00021013680781758955,
+      "loss": 0.7191,
+      "step": 2400
+    },
+    {
+      "epoch": 1.54,
+      "eval_loss": 0.7568334937095642,
+      "eval_runtime": 185.2269,
+      "eval_samples_per_second": 10.798,
+      "eval_steps_per_second": 1.35,
+      "step": 2400
+    },
+    {
+      "epoch": 1.55,
+      "learning_rate": 0.0002097459283387622,
+      "loss": 0.7532,
+      "step": 2410
+    },
+    {
+      "epoch": 1.56,
+      "learning_rate": 0.00020935504885993483,
+      "loss": 0.7308,
+      "step": 2420
+    },
+    {
+      "epoch": 1.56,
+      "learning_rate": 0.0002089641693811075,
+      "loss": 0.7167,
+      "step": 2430
+    },
+    {
+      "epoch": 1.57,
+      "learning_rate": 0.00020857328990228012,
+      "loss": 0.7371,
+      "step": 2440
+    },
+    {
+      "epoch": 1.58,
+      "learning_rate": 0.00020818241042345274,
+      "loss": 0.7317,
+      "step": 2450
+    },
+    {
+      "epoch": 1.58,
+      "learning_rate": 0.00020779153094462537,
+      "loss": 0.7253,
+      "step": 2460
+    },
+    {
+      "epoch": 1.59,
+      "learning_rate": 0.00020740065146579803,
+      "loss": 0.7477,
+      "step": 2470
+    },
+    {
+      "epoch": 1.59,
+      "learning_rate": 0.00020700977198697068,
+      "loss": 0.7493,
+      "step": 2480
+    },
+    {
+      "epoch": 1.6,
+      "learning_rate": 0.0002066188925081433,
+      "loss": 0.7255,
+      "step": 2490
+    },
+    {
+      "epoch": 1.61,
+      "learning_rate": 0.00020622801302931594,
+      "loss": 0.7455,
+      "step": 2500
+    },
+    {
+      "epoch": 1.61,
+      "learning_rate": 0.00020583713355048857,
+      "loss": 0.7205,
+      "step": 2510
+    },
+    {
+      "epoch": 1.62,
+      "learning_rate": 0.00020544625407166122,
+      "loss": 0.7456,
+      "step": 2520
+    },
+    {
+      "epoch": 1.63,
+      "learning_rate": 0.00020505537459283388,
+      "loss": 0.7179,
+      "step": 2530
+    },
+    {
+      "epoch": 1.63,
+      "learning_rate": 0.0002046644951140065,
+      "loss": 0.7284,
+      "step": 2540
+    },
+    {
+      "epoch": 1.64,
+      "learning_rate": 0.00020427361563517913,
+      "loss": 0.7428,
+      "step": 2550
+    },
+    {
+      "epoch": 1.65,
+      "learning_rate": 0.00020388273615635176,
+      "loss": 0.7267,
+      "step": 2560
+    },
+    {
+      "epoch": 1.65,
+      "learning_rate": 0.00020349185667752442,
+      "loss": 0.7395,
+      "step": 2570
+    },
+    {
+      "epoch": 1.66,
+      "learning_rate": 0.00020310097719869705,
+      "loss": 0.7205,
+      "step": 2580
+    },
+    {
+      "epoch": 1.67,
+      "learning_rate": 0.0002027100977198697,
+      "loss": 0.7238,
+      "step": 2590
+    },
+    {
+      "epoch": 1.67,
+      "learning_rate": 0.0002023192182410423,
+      "loss": 0.7572,
+      "step": 2600
+    },
+    {
+      "epoch": 1.67,
+      "eval_loss": 0.7548864483833313,
+      "eval_runtime": 185.0436,
+      "eval_samples_per_second": 10.808,
+      "eval_steps_per_second": 1.351,
+      "step": 2600
+    },
+    {
+      "epoch": 1.68,
+      "learning_rate": 0.00020192833876221496,
+      "loss": 0.7458,
+      "step": 2610
+    },
+    {
+      "epoch": 1.68,
+      "learning_rate": 0.0002015374592833876,
+      "loss": 0.7369,
+      "step": 2620
+    },
+    {
+      "epoch": 1.69,
+      "learning_rate": 0.00020114657980456024,
+      "loss": 0.7367,
+      "step": 2630
+    },
+    {
+      "epoch": 1.7,
+      "learning_rate": 0.0002007557003257329,
+      "loss": 0.7562,
+      "step": 2640
+    },
+    {
+      "epoch": 1.7,
+      "learning_rate": 0.0002003648208469055,
+      "loss": 0.7438,
+      "step": 2650
+    },
+    {
+      "epoch": 1.71,
+      "learning_rate": 0.00019997394136807815,
+      "loss": 0.7323,
+      "step": 2660
+    },
+    {
+      "epoch": 1.72,
+      "learning_rate": 0.0001995830618892508,
+      "loss": 0.7398,
+      "step": 2670
+    },
+    {
+      "epoch": 1.72,
+      "learning_rate": 0.00019919218241042344,
+      "loss": 0.7424,
+      "step": 2680
+    },
+    {
+      "epoch": 1.73,
+      "learning_rate": 0.0001988013029315961,
+      "loss": 0.735,
+      "step": 2690
+    },
+    {
+      "epoch": 1.74,
+      "learning_rate": 0.0001984104234527687,
+      "loss": 0.7102,
+      "step": 2700
+    },
+    {
+      "epoch": 1.74,
+      "learning_rate": 0.00019801954397394135,
+      "loss": 0.7271,
+      "step": 2710
+    },
+    {
+      "epoch": 1.75,
+      "learning_rate": 0.00019762866449511398,
+      "loss": 0.7332,
+      "step": 2720
+    },
+    {
+      "epoch": 1.76,
+      "learning_rate": 0.00019723778501628663,
+      "loss": 0.7062,
+      "step": 2730
+    },
+    {
+      "epoch": 1.76,
+      "learning_rate": 0.00019684690553745929,
+      "loss": 0.7228,
+      "step": 2740
+    },
+    {
+      "epoch": 1.77,
+      "learning_rate": 0.00019645602605863191,
+      "loss": 0.7216,
+      "step": 2750
+    },
+    {
+      "epoch": 1.77,
+      "learning_rate": 0.00019606514657980454,
+      "loss": 0.7358,
+      "step": 2760
+    },
+    {
+      "epoch": 1.78,
+      "learning_rate": 0.00019567426710097717,
+      "loss": 0.7373,
+      "step": 2770
+    },
+    {
+      "epoch": 1.79,
+      "learning_rate": 0.00019528338762214983,
+      "loss": 0.7214,
+      "step": 2780
+    },
+    {
+      "epoch": 1.79,
+      "learning_rate": 0.00019489250814332248,
+      "loss": 0.695,
+      "step": 2790
+    },
+    {
+      "epoch": 1.8,
+      "learning_rate": 0.0001945016286644951,
+      "loss": 0.738,
+      "step": 2800
+    },
+    {
+      "epoch": 1.8,
+      "eval_loss": 0.7532588839530945,
+      "eval_runtime": 185.3248,
+      "eval_samples_per_second": 10.792,
+      "eval_steps_per_second": 1.349,
+      "step": 2800
+    },
+    {
+      "epoch": 1.81,
+      "learning_rate": 0.00019411074918566774,
+      "loss": 0.7249,
+      "step": 2810
+    },
+    {
+      "epoch": 1.81,
+      "learning_rate": 0.00019371986970684037,
+      "loss": 0.7305,
+      "step": 2820
+    },
+    {
+      "epoch": 1.82,
+      "learning_rate": 0.00019332899022801302,
+      "loss": 0.7119,
+      "step": 2830
+    },
+    {
+      "epoch": 1.83,
+      "learning_rate": 0.00019293811074918565,
+      "loss": 0.7215,
+      "step": 2840
+    },
+    {
+      "epoch": 1.83,
+      "learning_rate": 0.0001925472312703583,
+      "loss": 0.7219,
+      "step": 2850
+    },
+    {
+      "epoch": 1.84,
+      "learning_rate": 0.0001921563517915309,
+      "loss": 0.7204,
+      "step": 2860
+    },
+    {
+      "epoch": 1.85,
+      "learning_rate": 0.00019176547231270356,
+      "loss": 0.7209,
+      "step": 2870
+    },
+    {
+      "epoch": 1.85,
+      "learning_rate": 0.00019137459283387622,
+      "loss": 0.7293,
+      "step": 2880
+    },
+    {
+      "epoch": 1.86,
+      "learning_rate": 0.00019098371335504884,
+      "loss": 0.7324,
+      "step": 2890
+    },
+    {
+      "epoch": 1.86,
+      "learning_rate": 0.0001905928338762215,
+      "loss": 0.7256,
+      "step": 2900
+    },
+    {
+      "epoch": 1.87,
+      "learning_rate": 0.0001902019543973941,
+      "loss": 0.7255,
+      "step": 2910
+    },
+    {
+      "epoch": 1.88,
+      "learning_rate": 0.00018981107491856675,
+      "loss": 0.7624,
+      "step": 2920
+    },
+    {
+      "epoch": 1.88,
+      "learning_rate": 0.0001894201954397394,
+      "loss": 0.7005,
+      "step": 2930
+    },
+    {
+      "epoch": 1.89,
+      "learning_rate": 0.00018902931596091204,
+      "loss": 0.7363,
+      "step": 2940
+    },
+    {
+      "epoch": 1.9,
+      "learning_rate": 0.0001886384364820847,
+      "loss": 0.7307,
+      "step": 2950
+    },
+    {
+      "epoch": 1.9,
+      "learning_rate": 0.0001882475570032573,
+      "loss": 0.725,
+      "step": 2960
+    },
+    {
+      "epoch": 1.91,
+      "learning_rate": 0.00018785667752442995,
+      "loss": 0.7456,
+      "step": 2970
+    },
+    {
+      "epoch": 1.92,
+      "learning_rate": 0.00018746579804560258,
+      "loss": 0.7328,
+      "step": 2980
+    },
+    {
+      "epoch": 1.92,
+      "learning_rate": 0.00018707491856677523,
+      "loss": 0.7391,
+      "step": 2990
+    },
+    {
+      "epoch": 1.93,
+      "learning_rate": 0.0001866840390879479,
+      "loss": 0.7163,
+      "step": 3000
+    },
+    {
+      "epoch": 1.93,
+      "eval_loss": 0.7519650459289551,
+      "eval_runtime": 185.5375,
+      "eval_samples_per_second": 10.779,
+      "eval_steps_per_second": 1.347,
+      "step": 3000
+    },
+    {
+      "epoch": 1.94,
+      "learning_rate": 0.0001862931596091205,
+      "loss": 0.7083,
+      "step": 3010
+    },
+    {
+      "epoch": 1.94,
+      "learning_rate": 0.00018590228013029314,
+      "loss": 0.7221,
+      "step": 3020
+    },
+    {
+      "epoch": 1.95,
+      "learning_rate": 0.00018551140065146577,
+      "loss": 0.733,
+      "step": 3030
+    },
+    {
+      "epoch": 1.95,
+      "learning_rate": 0.00018512052117263843,
+      "loss": 0.7427,
+      "step": 3040
+    },
+    {
+      "epoch": 1.96,
+      "learning_rate": 0.00018472964169381108,
+      "loss": 0.7329,
+      "step": 3050
+    },
+    {
+      "epoch": 1.97,
+      "learning_rate": 0.00018433876221498368,
+      "loss": 0.7261,
+      "step": 3060
+    },
+    {
+      "epoch": 1.97,
+      "learning_rate": 0.00018394788273615634,
+      "loss": 0.7267,
+      "step": 3070
+    },
+    {
+      "epoch": 1.98,
+      "learning_rate": 0.00018355700325732897,
+      "loss": 0.7216,
+      "step": 3080
+    },
+    {
+      "epoch": 1.99,
+      "learning_rate": 0.00018316612377850162,
+      "loss": 0.7375,
+      "step": 3090
+    },
+    {
+      "epoch": 1.99,
+      "learning_rate": 0.00018277524429967425,
+      "loss": 0.6999,
+      "step": 3100
+    },
+    {
+      "epoch": 2.0,
+      "learning_rate": 0.00018238436482084688,
+      "loss": 0.7157,
+      "step": 3110
+    },
+    {
+      "epoch": 2.01,
+      "learning_rate": 0.0001819934853420195,
+      "loss": 0.678,
+      "step": 3120
+    },
+    {
+      "epoch": 2.01,
+      "learning_rate": 0.00018160260586319216,
+      "loss": 0.704,
+      "step": 3130
+    },
+    {
+      "epoch": 2.02,
+      "learning_rate": 0.00018121172638436482,
+      "loss": 0.6724,
+      "step": 3140
+    },
+    {
+      "epoch": 2.03,
+      "learning_rate": 0.00018082084690553745,
+      "loss": 0.6877,
+      "step": 3150
+    },
+    {
+      "epoch": 2.03,
+      "learning_rate": 0.0001804299674267101,
+      "loss": 0.7152,
+      "step": 3160
+    },
+    {
+      "epoch": 2.04,
+      "learning_rate": 0.0001800390879478827,
+      "loss": 0.6923,
+      "step": 3170
+    },
+    {
+      "epoch": 2.05,
+      "learning_rate": 0.00017964820846905536,
+      "loss": 0.667,
+      "step": 3180
+    },
+    {
+      "epoch": 2.05,
+      "learning_rate": 0.000179257328990228,
+      "loss": 0.6741,
+      "step": 3190
+    },
+    {
+      "epoch": 2.06,
+      "learning_rate": 0.00017886644951140064,
+      "loss": 0.6879,
+      "step": 3200
+    },
+    {
+      "epoch": 2.06,
+      "eval_loss": 0.7551190257072449,
+      "eval_runtime": 185.6642,
+      "eval_samples_per_second": 10.772,
+      "eval_steps_per_second": 1.347,
+      "step": 3200
+    },
+    {
+      "epoch": 2.06,
+      "learning_rate": 0.0001784755700325733,
+      "loss": 0.6782,
+      "step": 3210
+    },
+    {
+      "epoch": 2.07,
+      "learning_rate": 0.0001780846905537459,
+      "loss": 0.69,
+      "step": 3220
+    },
+    {
+      "epoch": 2.08,
+      "learning_rate": 0.00017769381107491855,
+      "loss": 0.6785,
+      "step": 3230
+    },
+    {
+      "epoch": 2.08,
+      "learning_rate": 0.00017730293159609118,
+      "loss": 0.6908,
+      "step": 3240
+    },
+    {
+      "epoch": 2.09,
+      "learning_rate": 0.00017691205211726384,
+      "loss": 0.6877,
+      "step": 3250
+    },
+    {
+      "epoch": 2.1,
+      "learning_rate": 0.0001765211726384365,
+      "loss": 0.6598,
+      "step": 3260
+    },
+    {
+      "epoch": 2.1,
+      "learning_rate": 0.0001761302931596091,
+      "loss": 0.6786,
+      "step": 3270
+    },
+    {
+      "epoch": 2.11,
+      "learning_rate": 0.00017573941368078175,
+      "loss": 0.699,
+      "step": 3280
+    },
+    {
+      "epoch": 2.12,
+      "learning_rate": 0.00017534853420195438,
+      "loss": 0.6889,
+      "step": 3290
+    },
+    {
+      "epoch": 2.12,
+      "learning_rate": 0.00017495765472312703,
+      "loss": 0.6925,
+      "step": 3300
+    },
+    {
+      "epoch": 2.13,
+      "learning_rate": 0.00017456677524429966,
+      "loss": 0.6823,
+      "step": 3310
+    },
+    {
+      "epoch": 2.14,
+      "learning_rate": 0.0001741758957654723,
+      "loss": 0.672,
+      "step": 3320
+    },
+    {
+      "epoch": 2.14,
+      "learning_rate": 0.00017378501628664494,
+      "loss": 0.702,
+      "step": 3330
+    },
+    {
+      "epoch": 2.15,
+      "learning_rate": 0.00017339413680781757,
+      "loss": 0.6775,
+      "step": 3340
+    },
+    {
+      "epoch": 2.15,
+      "learning_rate": 0.00017300325732899023,
+      "loss": 0.6821,
+      "step": 3350
+    },
+    {
+      "epoch": 2.16,
+      "learning_rate": 0.00017261237785016285,
+      "loss": 0.6746,
+      "step": 3360
+    },
+    {
+      "epoch": 2.17,
+      "learning_rate": 0.00017222149837133548,
+      "loss": 0.6573,
+      "step": 3370
+    },
+    {
+      "epoch": 2.17,
+      "learning_rate": 0.0001718306188925081,
+      "loss": 0.7045,
+      "step": 3380
+    },
+    {
+      "epoch": 2.18,
+      "learning_rate": 0.00017143973941368077,
+      "loss": 0.6952,
+      "step": 3390
+    },
+    {
+      "epoch": 2.19,
+      "learning_rate": 0.00017104885993485342,
+      "loss": 0.6944,
+      "step": 3400
+    },
+    {
+      "epoch": 2.19,
+      "eval_loss": 0.7553194761276245,
+      "eval_runtime": 185.739,
+      "eval_samples_per_second": 10.768,
+      "eval_steps_per_second": 1.346,
+      "step": 3400
+    },
+    {
+      "epoch": 2.19,
+      "learning_rate": 0.00017065798045602605,
+      "loss": 0.682,
+      "step": 3410
+    },
+    {
+      "epoch": 2.2,
+      "learning_rate": 0.00017026710097719868,
+      "loss": 0.7069,
+      "step": 3420
+    },
+    {
+      "epoch": 2.21,
+      "learning_rate": 0.0001698762214983713,
+      "loss": 0.6771,
+      "step": 3430
+    },
+    {
+      "epoch": 2.21,
+      "learning_rate": 0.00016948534201954396,
+      "loss": 0.7152,
+      "step": 3440
+    },
+    {
+      "epoch": 2.22,
+      "learning_rate": 0.0001690944625407166,
+      "loss": 0.6842,
+      "step": 3450
+    },
+    {
+      "epoch": 2.23,
+      "learning_rate": 0.00016870358306188924,
+      "loss": 0.7098,
+      "step": 3460
+    },
+    {
+      "epoch": 2.23,
+      "learning_rate": 0.00016831270358306187,
+      "loss": 0.7196,
+      "step": 3470
+    },
+    {
+      "epoch": 2.24,
+      "learning_rate": 0.0001679218241042345,
+      "loss": 0.695,
+      "step": 3480
+    },
+    {
+      "epoch": 2.24,
+      "learning_rate": 0.00016753094462540716,
+      "loss": 0.706,
+      "step": 3490
+    },
+    {
+      "epoch": 2.25,
+      "learning_rate": 0.00016714006514657978,
+      "loss": 0.6889,
+      "step": 3500
+    },
+    {
+      "epoch": 2.26,
+      "learning_rate": 0.00016674918566775244,
+      "loss": 0.6772,
+      "step": 3510
+    },
+    {
+      "epoch": 2.26,
+      "learning_rate": 0.0001663583061889251,
+      "loss": 0.7214,
+      "step": 3520
+    },
+    {
+      "epoch": 2.27,
+      "learning_rate": 0.0001659674267100977,
+      "loss": 0.6849,
+      "step": 3530
+    },
+    {
+      "epoch": 2.28,
+      "learning_rate": 0.00016557654723127035,
+      "loss": 0.7007,
+      "step": 3540
+    },
+    {
+      "epoch": 2.28,
+      "learning_rate": 0.00016518566775244298,
+      "loss": 0.6973,
+      "step": 3550
+    },
+    {
+      "epoch": 2.29,
+      "learning_rate": 0.00016479478827361563,
+      "loss": 0.7198,
+      "step": 3560
+    },
+    {
+      "epoch": 2.3,
+      "learning_rate": 0.00016440390879478826,
+      "loss": 0.6958,
+      "step": 3570
+    },
+    {
+      "epoch": 2.3,
+      "learning_rate": 0.0001640130293159609,
+      "loss": 0.692,
+      "step": 3580
+    },
+    {
+      "epoch": 2.31,
+      "learning_rate": 0.00016362214983713352,
+      "loss": 0.6965,
+      "step": 3590
+    },
+    {
+      "epoch": 2.32,
+      "learning_rate": 0.00016323127035830617,
+      "loss": 0.6642,
+      "step": 3600
+    },
+    {
+      "epoch": 2.32,
+      "eval_loss": 0.7540327310562134,
+      "eval_runtime": 185.9138,
+      "eval_samples_per_second": 10.758,
+      "eval_steps_per_second": 1.345,
+      "step": 3600
+    },
+    {
+      "epoch": 2.32,
+      "learning_rate": 0.00016284039087947883,
+      "loss": 0.6873,
+      "step": 3610
+    },
+    {
+      "epoch": 2.33,
+      "learning_rate": 0.00016244951140065146,
+      "loss": 0.6797,
+      "step": 3620
+    },
+    {
+      "epoch": 2.33,
+      "learning_rate": 0.00016205863192182408,
+      "loss": 0.6979,
+      "step": 3630
+    },
+    {
+      "epoch": 2.34,
+      "learning_rate": 0.0001616677524429967,
+      "loss": 0.7001,
+      "step": 3640
+    },
+    {
+      "epoch": 2.35,
+      "learning_rate": 0.00016127687296416937,
+      "loss": 0.7082,
+      "step": 3650
+    },
+    {
+      "epoch": 2.35,
+      "learning_rate": 0.00016088599348534202,
+      "loss": 0.7059,
+      "step": 3660
+    },
+    {
+      "epoch": 2.36,
+      "learning_rate": 0.00016049511400651465,
+      "loss": 0.6995,
+      "step": 3670
+    },
+    {
+      "epoch": 2.37,
+      "learning_rate": 0.00016010423452768728,
+      "loss": 0.6789,
+      "step": 3680
+    },
+    {
+      "epoch": 2.37,
+      "learning_rate": 0.0001597133550488599,
+      "loss": 0.6902,
+      "step": 3690
+    },
+    {
+      "epoch": 2.38,
+      "learning_rate": 0.00015932247557003256,
+      "loss": 0.7039,
+      "step": 3700
+    },
+    {
+      "epoch": 2.39,
+      "learning_rate": 0.0001589315960912052,
+      "loss": 0.6914,
+      "step": 3710
+    },
+    {
+      "epoch": 2.39,
+      "learning_rate": 0.00015854071661237785,
+      "loss": 0.6811,
+      "step": 3720
+    },
+    {
+      "epoch": 2.4,
+      "learning_rate": 0.00015814983713355045,
+      "loss": 0.6733,
+      "step": 3730
+    },
+    {
+      "epoch": 2.41,
+      "learning_rate": 0.0001577589576547231,
+      "loss": 0.6626,
+      "step": 3740
+    },
+    {
+      "epoch": 2.41,
+      "learning_rate": 0.00015736807817589576,
+      "loss": 0.704,
+      "step": 3750
+    },
+    {
+      "epoch": 2.42,
+      "learning_rate": 0.00015697719869706839,
+      "loss": 0.6941,
+      "step": 3760
+    },
+    {
+      "epoch": 2.42,
+      "learning_rate": 0.00015658631921824104,
+      "loss": 0.6982,
+      "step": 3770
+    },
+    {
+      "epoch": 2.43,
+      "learning_rate": 0.00015619543973941364,
+      "loss": 0.6998,
+      "step": 3780
+    },
+    {
+      "epoch": 2.44,
+      "learning_rate": 0.0001558045602605863,
+      "loss": 0.7003,
+      "step": 3790
+    },
+    {
+      "epoch": 2.44,
+      "learning_rate": 0.00015541368078175895,
+      "loss": 0.7006,
+      "step": 3800
+    },
+    {
+      "epoch": 2.44,
+      "eval_loss": 0.7535502314567566,
+      "eval_runtime": 185.891,
+      "eval_samples_per_second": 10.759,
+      "eval_steps_per_second": 1.345,
+      "step": 3800
+    },
+    {
+      "epoch": 2.45,
+      "learning_rate": 0.00015502280130293158,
+      "loss": 0.6891,
+      "step": 3810
+    },
+    {
+      "epoch": 2.46,
+      "learning_rate": 0.00015463192182410424,
+      "loss": 0.6828,
+      "step": 3820
+    },
+    {
+      "epoch": 2.46,
+      "learning_rate": 0.00015424104234527684,
+      "loss": 0.6859,
+      "step": 3830
+    },
+    {
+      "epoch": 2.47,
+      "learning_rate": 0.0001538501628664495,
+      "loss": 0.6896,
+      "step": 3840
+    },
+    {
+      "epoch": 2.48,
+      "learning_rate": 0.00015345928338762212,
+      "loss": 0.694,
+      "step": 3850
+    },
+    {
+      "epoch": 2.48,
+      "learning_rate": 0.00015306840390879478,
+      "loss": 0.7111,
+      "step": 3860
+    },
+    {
+      "epoch": 2.49,
+      "learning_rate": 0.00015267752442996743,
+      "loss": 0.6618,
+      "step": 3870
+    },
+    {
+      "epoch": 2.5,
+      "learning_rate": 0.00015228664495114003,
+      "loss": 0.6858,
+      "step": 3880
+    },
+    {
+      "epoch": 2.5,
+      "learning_rate": 0.0001518957654723127,
+      "loss": 0.7063,
+      "step": 3890
+    },
+    {
+      "epoch": 2.51,
+      "learning_rate": 0.00015150488599348532,
+      "loss": 0.6828,
+      "step": 3900
+    },
+    {
+      "epoch": 2.51,
+      "learning_rate": 0.00015111400651465797,
+      "loss": 0.6846,
+      "step": 3910
+    },
+    {
+      "epoch": 2.52,
+      "learning_rate": 0.00015072312703583063,
+      "loss": 0.6835,
+      "step": 3920
+    },
+    {
+      "epoch": 2.53,
+      "learning_rate": 0.00015033224755700325,
+      "loss": 0.6799,
+      "step": 3930
+    },
+    {
+      "epoch": 2.53,
+      "learning_rate": 0.00014994136807817588,
+      "loss": 0.678,
+      "step": 3940
+    },
+    {
+      "epoch": 2.54,
+      "learning_rate": 0.0001495504885993485,
+      "loss": 0.6848,
+      "step": 3950
+    },
+    {
+      "epoch": 2.55,
+      "learning_rate": 0.00014915960912052117,
+      "loss": 0.6839,
+      "step": 3960
+    },
+    {
+      "epoch": 2.55,
+      "learning_rate": 0.0001487687296416938,
+      "loss": 0.7062,
+      "step": 3970
+    },
+    {
+      "epoch": 2.56,
+      "learning_rate": 0.00014837785016286642,
+      "loss": 0.6831,
+      "step": 3980
+    },
+    {
+      "epoch": 2.57,
+      "learning_rate": 0.00014798697068403908,
+      "loss": 0.6938,
+      "step": 3990
+    },
+    {
+      "epoch": 2.57,
+      "learning_rate": 0.0001475960912052117,
+      "loss": 0.6991,
+      "step": 4000
+    },
+    {
+      "epoch": 2.57,
+      "eval_loss": 0.7521887421607971,
+      "eval_runtime": 185.8289,
+      "eval_samples_per_second": 10.763,
+      "eval_steps_per_second": 1.345,
+      "step": 4000
+    },
+    {
+      "epoch": 2.58,
+      "learning_rate": 0.00014720521172638436,
+      "loss": 0.6969,
+      "step": 4010
+    },
+    {
+      "epoch": 2.59,
+      "learning_rate": 0.000146814332247557,
+      "loss": 0.6977,
+      "step": 4020
+    },
+    {
+      "epoch": 2.59,
+      "learning_rate": 0.00014642345276872962,
+      "loss": 0.6786,
+      "step": 4030
+    },
+    {
+      "epoch": 2.6,
+      "learning_rate": 0.00014603257328990227,
+      "loss": 0.7093,
+      "step": 4040
+    },
+    {
+      "epoch": 2.6,
+      "learning_rate": 0.0001456416938110749,
+      "loss": 0.7166,
+      "step": 4050
+    },
+    {
+      "epoch": 2.61,
+      "learning_rate": 0.00014525081433224756,
+      "loss": 0.6865,
+      "step": 4060
+    },
+    {
+      "epoch": 2.62,
+      "learning_rate": 0.00014485993485342018,
+      "loss": 0.7042,
+      "step": 4070
+    },
+    {
+      "epoch": 2.62,
+      "learning_rate": 0.0001444690553745928,
+      "loss": 0.6879,
+      "step": 4080
+    },
+    {
+      "epoch": 2.63,
+      "learning_rate": 0.00014407817589576547,
+      "loss": 0.6843,
+      "step": 4090
+    },
+    {
+      "epoch": 2.64,
+      "learning_rate": 0.0001436872964169381,
+      "loss": 0.6819,
+      "step": 4100
+    },
+    {
+      "epoch": 2.64,
+      "learning_rate": 0.00014329641693811072,
+      "loss": 0.6918,
+      "step": 4110
+    },
+    {
+      "epoch": 2.65,
+      "learning_rate": 0.00014290553745928338,
+      "loss": 0.6747,
+      "step": 4120
+    },
+    {
+      "epoch": 2.66,
+      "learning_rate": 0.000142514657980456,
+      "loss": 0.7034,
+      "step": 4130
+    },
+    {
+      "epoch": 2.66,
+      "learning_rate": 0.00014212377850162866,
+      "loss": 0.6945,
+      "step": 4140
+    },
+    {
+      "epoch": 2.67,
+      "learning_rate": 0.0001417328990228013,
+      "loss": 0.6718,
+      "step": 4150
+    },
+    {
+      "epoch": 2.68,
+      "learning_rate": 0.00014134201954397392,
+      "loss": 0.6768,
+      "step": 4160
+    },
+    {
+      "epoch": 2.68,
+      "learning_rate": 0.00014095114006514657,
+      "loss": 0.6799,
+      "step": 4170
+    },
+    {
+      "epoch": 2.69,
+      "learning_rate": 0.0001405602605863192,
+      "loss": 0.7009,
+      "step": 4180
+    },
+    {
+      "epoch": 2.69,
+      "learning_rate": 0.00014016938110749186,
+      "loss": 0.6575,
+      "step": 4190
+    },
+    {
+      "epoch": 2.7,
+      "learning_rate": 0.00013977850162866449,
+      "loss": 0.6884,
+      "step": 4200
+    },
+    {
+      "epoch": 2.7,
+      "eval_loss": 0.750479519367218,
+      "eval_runtime": 185.8654,
+      "eval_samples_per_second": 10.76,
+      "eval_steps_per_second": 1.345,
+      "step": 4200
+    },
+    {
+      "epoch": 2.71,
+      "learning_rate": 0.0001393876221498371,
+      "loss": 0.6961,
+      "step": 4210
+    },
+    {
+      "epoch": 2.71,
+      "learning_rate": 0.00013899674267100977,
+      "loss": 0.7139,
+      "step": 4220
+    },
+    {
+      "epoch": 2.72,
+      "learning_rate": 0.0001386058631921824,
+      "loss": 0.6777,
+      "step": 4230
+    },
+    {
+      "epoch": 2.73,
+      "learning_rate": 0.00013821498371335502,
+      "loss": 0.6911,
+      "step": 4240
+    },
+    {
+      "epoch": 2.73,
+      "learning_rate": 0.00013782410423452768,
+      "loss": 0.6906,
+      "step": 4250
+    },
+    {
+      "epoch": 2.74,
+      "learning_rate": 0.0001374332247557003,
+      "loss": 0.6955,
+      "step": 4260
+    },
+    {
+      "epoch": 2.75,
+      "learning_rate": 0.00013704234527687296,
+      "loss": 0.68,
+      "step": 4270
+    },
+    {
+      "epoch": 2.75,
+      "learning_rate": 0.0001366514657980456,
+      "loss": 0.6756,
+      "step": 4280
+    },
+    {
+      "epoch": 2.76,
+      "learning_rate": 0.00013626058631921822,
+      "loss": 0.6996,
+      "step": 4290
+    },
+    {
+      "epoch": 2.77,
+      "learning_rate": 0.00013586970684039087,
+      "loss": 0.6608,
+      "step": 4300
+    },
+    {
+      "epoch": 2.77,
+      "learning_rate": 0.0001354788273615635,
+      "loss": 0.6942,
+      "step": 4310
+    },
+    {
+      "epoch": 2.78,
+      "learning_rate": 0.00013508794788273616,
+      "loss": 0.6677,
+      "step": 4320
+    },
+    {
+      "epoch": 2.78,
+      "learning_rate": 0.0001346970684039088,
+      "loss": 0.6876,
+      "step": 4330
+    },
+    {
+      "epoch": 2.79,
+      "learning_rate": 0.00013430618892508141,
+      "loss": 0.6797,
+      "step": 4340
+    },
+    {
+      "epoch": 2.8,
+      "learning_rate": 0.00013391530944625407,
+      "loss": 0.6802,
+      "step": 4350
+    },
+    {
+      "epoch": 2.8,
+      "learning_rate": 0.0001335244299674267,
+      "loss": 0.6918,
+      "step": 4360
+    },
+    {
+      "epoch": 2.81,
+      "learning_rate": 0.00013313355048859933,
+      "loss": 0.6843,
+      "step": 4370
+    },
+    {
+      "epoch": 2.82,
+      "learning_rate": 0.00013274267100977198,
+      "loss": 0.6849,
+      "step": 4380
+    },
+    {
+      "epoch": 2.82,
+      "learning_rate": 0.0001323517915309446,
+      "loss": 0.6743,
+      "step": 4390
+    },
+    {
+      "epoch": 2.83,
+      "learning_rate": 0.00013196091205211726,
+      "loss": 0.6879,
+      "step": 4400
+    },
+    {
+      "epoch": 2.83,
+      "eval_loss": 0.7500499486923218,
+      "eval_runtime": 185.8465,
+      "eval_samples_per_second": 10.762,
+      "eval_steps_per_second": 1.345,
+      "step": 4400
+    },
+    {
+      "epoch": 2.84,
+      "learning_rate": 0.0001315700325732899,
+      "loss": 0.701,
+      "step": 4410
+    },
+    {
+      "epoch": 2.84,
+      "learning_rate": 0.00013117915309446252,
+      "loss": 0.6889,
+      "step": 4420
+    },
+    {
+      "epoch": 2.85,
+      "learning_rate": 0.00013078827361563518,
+      "loss": 0.7037,
+      "step": 4430
+    },
+    {
+      "epoch": 2.86,
+      "learning_rate": 0.0001303973941368078,
+      "loss": 0.6876,
+      "step": 4440
+    },
+    {
+      "epoch": 2.86,
+      "learning_rate": 0.00013000651465798046,
+      "loss": 0.6965,
+      "step": 4450
+    },
+    {
+      "epoch": 2.87,
+      "learning_rate": 0.0001296156351791531,
+      "loss": 0.708,
+      "step": 4460
+    },
+    {
+      "epoch": 2.87,
+      "learning_rate": 0.00012922475570032572,
+      "loss": 0.6795,
+      "step": 4470
+    },
+    {
+      "epoch": 2.88,
+      "learning_rate": 0.00012883387622149837,
+      "loss": 0.6757,
+      "step": 4480
+    },
+    {
+      "epoch": 2.89,
+      "learning_rate": 0.000128442996742671,
+      "loss": 0.6837,
+      "step": 4490
+    },
+    {
+      "epoch": 2.89,
+      "learning_rate": 0.00012805211726384363,
+      "loss": 0.6829,
+      "step": 4500
+    },
+    {
+      "epoch": 2.9,
+      "learning_rate": 0.00012766123778501626,
+      "loss": 0.6873,
+      "step": 4510
+    },
+    {
+      "epoch": 2.91,
+      "learning_rate": 0.0001272703583061889,
+      "loss": 0.7024,
+      "step": 4520
+    },
+    {
+      "epoch": 2.91,
+      "learning_rate": 0.00012687947882736157,
+      "loss": 0.6787,
+      "step": 4530
+    },
+    {
+      "epoch": 2.92,
+      "learning_rate": 0.0001264885993485342,
+      "loss": 0.7035,
+      "step": 4540
+    },
+    {
+      "epoch": 2.93,
+      "learning_rate": 0.00012609771986970682,
+      "loss": 0.6893,
+      "step": 4550
+    },
+    {
+      "epoch": 2.93,
+      "learning_rate": 0.00012570684039087945,
+      "loss": 0.7013,
+      "step": 4560
+    },
+    {
+      "epoch": 2.94,
+      "learning_rate": 0.0001253159609120521,
+      "loss": 0.6746,
+      "step": 4570
+    },
+    {
+      "epoch": 2.95,
+      "learning_rate": 0.00012492508143322476,
+      "loss": 0.6924,
+      "step": 4580
+    },
+    {
+      "epoch": 2.95,
+      "learning_rate": 0.0001245342019543974,
+      "loss": 0.6768,
+      "step": 4590
+    },
+    {
+      "epoch": 2.96,
+      "learning_rate": 0.00012414332247557002,
+      "loss": 0.6815,
+      "step": 4600
+    },
+    {
+      "epoch": 2.96,
+      "eval_loss": 0.7489736080169678,
+      "eval_runtime": 185.9255,
+      "eval_samples_per_second": 10.757,
+      "eval_steps_per_second": 1.345,
+      "step": 4600
+    }
+  ],
+  "logging_steps": 10,
+  "max_steps": 7775,
+  "num_train_epochs": 5,
+  "save_steps": 200,
+  "total_flos": 1.7074957053851075e+18,
+  "trial_name": null,
+  "trial_params": null
+}

with_input/decomp_plan/llama-2-7b/best_model/training_args.bin

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:1502ec45bcfde78e55b2ef8badf6d1db8d39b4f9e5616c9d0c436ebdf8d7504e
+size 4091