Upload 13 files

README.md

@@ -1,16 +1,12 @@
----
-license: cc-by-sa-4.0
-datasets:
-- izumi-lab/llm-japanese-dataset
-language:
-- en
-- ja
----
+## Training procedure
 
-# This model is finetuned by the joint efforts of Sparticle Inc. and A. I. Hakusan Inc.
-## This model is a fine-tuned Llama2-7b model with Japanese dataset with LoRA.
-The training set of this model is about 5% of randomly chosen data from llm-japanese-dataset by izumi-lab.
-For inference, please follow the instructions in https://github.com/tloen/alpaca-lora/ .
 
-Loss = 0.6046
-PPL = 1.5206
+The following `bitsandbytes` quantization config was used during training:
+- load_in_8bit: True
+- llm_int8_threshold: 6.0
+- llm_int8_skip_modules: None
+- llm_int8_enable_fp32_cpu_offload: False
+### Framework versions
+
+
+- PEFT 0.5.0.dev0

adapter_config.json

@@ -1,6 +1,6 @@
 {
   "auto_mapping": null,
-  "base_model_name_or_path": "../llama/Llama-2-7b-hf",
+  "base_model_name_or_path": "../llama/Llama-2-7b-chat-hf",
   "bias": "none",
   "fan_in_fan_out": false,
   "inference_mode": true,

checkpoint-1600/optimizer.pt

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:a8cdadfe13d98d72b8c5b52b0c9e247216b45902febe1c347f8bf9ec19784755
+size 33661637

checkpoint-1600/pytorch_model.bin

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:1337967f27f02a8db07fd9a0eb4c2a884eb143fb9a52cd100b513e3f04344343
+size 16822989

checkpoint-1600/rng_state_0.pth

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:9c647e66c63f8455445708c81235f21165676384fb166b52f28994cd6489be4e
+size 17655

checkpoint-1600/rng_state_1.pth

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:3e9c862bfef6c94a007875ff2ce2a3e912ac6df2e40e678783566b6f9d5aa946
+size 17655

checkpoint-1600/rng_state_2.pth

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:7dae8ec34b426a986693e2a99f72f0ebb1aa81cb2608a69f239c7c42c449f144
+size 17655

checkpoint-1600/rng_state_3.pth

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:94a034bdba4d67402184850929dbfbda245a432d11b2f62f9587e5489bbbe78b
+size 17655

checkpoint-1600/scaler.pt

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:cbbb3b0bb9c64d37bf898d0431b3ed1f0a6f5c9d8c2b563e0f884424fb8bd92a
+size 557

checkpoint-1600/scheduler.pt

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:47c0d57c8e4c5a5bbe8567ed75264238e3147209880bef6733f3b09e0f3e7f2a
+size 627

checkpoint-1600/trainer_state.json

@@ -0,0 +1,1040 @@
+{
+  "best_metric": 0.6304358839988708,
+  "best_model_checkpoint": "../outputs/7b_chat_lora_0.05/checkpoint-1600",
+  "epoch": 0.9644364074743822,
+  "global_step": 1600,
+  "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": 2.544,
+      "step": 10
+    },
+    {
+      "epoch": 0.01,
+      "learning_rate": 5.9999999999999995e-05,
+      "loss": 2.4187,
+      "step": 20
+    },
+    {
+      "epoch": 0.02,
+      "learning_rate": 8.999999999999999e-05,
+      "loss": 2.019,
+      "step": 30
+    },
+    {
+      "epoch": 0.02,
+      "learning_rate": 0.00011999999999999999,
+      "loss": 1.4918,
+      "step": 40
+    },
+    {
+      "epoch": 0.03,
+      "learning_rate": 0.00015,
+      "loss": 1.0844,
+      "step": 50
+    },
+    {
+      "epoch": 0.04,
+      "learning_rate": 0.00017999999999999998,
+      "loss": 0.9015,
+      "step": 60
+    },
+    {
+      "epoch": 0.04,
+      "learning_rate": 0.00020999999999999998,
+      "loss": 0.8297,
+      "step": 70
+    },
+    {
+      "epoch": 0.05,
+      "learning_rate": 0.00023999999999999998,
+      "loss": 0.798,
+      "step": 80
+    },
+    {
+      "epoch": 0.05,
+      "learning_rate": 0.00027,
+      "loss": 0.7381,
+      "step": 90
+    },
+    {
+      "epoch": 0.06,
+      "learning_rate": 0.0003,
+      "loss": 0.7017,
+      "step": 100
+    },
+    {
+      "epoch": 0.07,
+      "learning_rate": 0.0002980756895445798,
+      "loss": 0.7003,
+      "step": 110
+    },
+    {
+      "epoch": 0.07,
+      "learning_rate": 0.0002961513790891597,
+      "loss": 0.6933,
+      "step": 120
+    },
+    {
+      "epoch": 0.08,
+      "learning_rate": 0.0002942270686337396,
+      "loss": 0.6878,
+      "step": 130
+    },
+    {
+      "epoch": 0.08,
+      "learning_rate": 0.00029230275817831937,
+      "loss": 0.685,
+      "step": 140
+    },
+    {
+      "epoch": 0.09,
+      "learning_rate": 0.00029037844772289927,
+      "loss": 0.6865,
+      "step": 150
+    },
+    {
+      "epoch": 0.1,
+      "learning_rate": 0.0002884541372674791,
+      "loss": 0.6768,
+      "step": 160
+    },
+    {
+      "epoch": 0.1,
+      "learning_rate": 0.00028652982681205897,
+      "loss": 0.6737,
+      "step": 170
+    },
+    {
+      "epoch": 0.11,
+      "learning_rate": 0.0002846055163566389,
+      "loss": 0.6704,
+      "step": 180
+    },
+    {
+      "epoch": 0.11,
+      "learning_rate": 0.0002826812059012187,
+      "loss": 0.6741,
+      "step": 190
+    },
+    {
+      "epoch": 0.12,
+      "learning_rate": 0.00028075689544579857,
+      "loss": 0.6699,
+      "step": 200
+    },
+    {
+      "epoch": 0.12,
+      "eval_loss": 0.6688060164451599,
+      "eval_runtime": 25.909,
+      "eval_samples_per_second": 77.193,
+      "eval_steps_per_second": 2.432,
+      "step": 200
+    },
+    {
+      "epoch": 0.13,
+      "learning_rate": 0.0002788325849903784,
+      "loss": 0.6796,
+      "step": 210
+    },
+    {
+      "epoch": 0.13,
+      "learning_rate": 0.00027690827453495827,
+      "loss": 0.6694,
+      "step": 220
+    },
+    {
+      "epoch": 0.14,
+      "learning_rate": 0.00027498396407953817,
+      "loss": 0.6727,
+      "step": 230
+    },
+    {
+      "epoch": 0.14,
+      "learning_rate": 0.000273059653624118,
+      "loss": 0.6636,
+      "step": 240
+    },
+    {
+      "epoch": 0.15,
+      "learning_rate": 0.00027113534316869787,
+      "loss": 0.6602,
+      "step": 250
+    },
+    {
+      "epoch": 0.16,
+      "learning_rate": 0.0002692110327132777,
+      "loss": 0.6617,
+      "step": 260
+    },
+    {
+      "epoch": 0.16,
+      "learning_rate": 0.00026728672225785757,
+      "loss": 0.6475,
+      "step": 270
+    },
+    {
+      "epoch": 0.17,
+      "learning_rate": 0.0002653624118024374,
+      "loss": 0.6634,
+      "step": 280
+    },
+    {
+      "epoch": 0.17,
+      "learning_rate": 0.0002634381013470173,
+      "loss": 0.6611,
+      "step": 290
+    },
+    {
+      "epoch": 0.18,
+      "learning_rate": 0.00026151379089159717,
+      "loss": 0.6636,
+      "step": 300
+    },
+    {
+      "epoch": 0.19,
+      "learning_rate": 0.000259589480436177,
+      "loss": 0.6565,
+      "step": 310
+    },
+    {
+      "epoch": 0.19,
+      "learning_rate": 0.00025766516998075687,
+      "loss": 0.6583,
+      "step": 320
+    },
+    {
+      "epoch": 0.2,
+      "learning_rate": 0.0002557408595253367,
+      "loss": 0.6515,
+      "step": 330
+    },
+    {
+      "epoch": 0.2,
+      "learning_rate": 0.00025381654906991656,
+      "loss": 0.6687,
+      "step": 340
+    },
+    {
+      "epoch": 0.21,
+      "learning_rate": 0.00025189223861449647,
+      "loss": 0.6493,
+      "step": 350
+    },
+    {
+      "epoch": 0.22,
+      "learning_rate": 0.0002499679281590763,
+      "loss": 0.6608,
+      "step": 360
+    },
+    {
+      "epoch": 0.22,
+      "learning_rate": 0.00024804361770365617,
+      "loss": 0.642,
+      "step": 370
+    },
+    {
+      "epoch": 0.23,
+      "learning_rate": 0.000246119307248236,
+      "loss": 0.6549,
+      "step": 380
+    },
+    {
+      "epoch": 0.24,
+      "learning_rate": 0.00024419499679281586,
+      "loss": 0.6521,
+      "step": 390
+    },
+    {
+      "epoch": 0.24,
+      "learning_rate": 0.00024227068633739577,
+      "loss": 0.6517,
+      "step": 400
+    },
+    {
+      "epoch": 0.24,
+      "eval_loss": 0.6501107811927795,
+      "eval_runtime": 26.1226,
+      "eval_samples_per_second": 76.562,
+      "eval_steps_per_second": 2.412,
+      "step": 400
+    },
+    {
+      "epoch": 0.25,
+      "learning_rate": 0.00024034637588197562,
+      "loss": 0.6536,
+      "step": 410
+    },
+    {
+      "epoch": 0.25,
+      "learning_rate": 0.00023842206542655546,
+      "loss": 0.6515,
+      "step": 420
+    },
+    {
+      "epoch": 0.26,
+      "learning_rate": 0.00023649775497113534,
+      "loss": 0.6605,
+      "step": 430
+    },
+    {
+      "epoch": 0.27,
+      "learning_rate": 0.0002345734445157152,
+      "loss": 0.6499,
+      "step": 440
+    },
+    {
+      "epoch": 0.27,
+      "learning_rate": 0.00023264913406029504,
+      "loss": 0.6455,
+      "step": 450
+    },
+    {
+      "epoch": 0.28,
+      "learning_rate": 0.00023072482360487491,
+      "loss": 0.6545,
+      "step": 460
+    },
+    {
+      "epoch": 0.28,
+      "learning_rate": 0.00022880051314945476,
+      "loss": 0.6523,
+      "step": 470
+    },
+    {
+      "epoch": 0.29,
+      "learning_rate": 0.0002268762026940346,
+      "loss": 0.6428,
+      "step": 480
+    },
+    {
+      "epoch": 0.3,
+      "learning_rate": 0.0002249518922386145,
+      "loss": 0.6477,
+      "step": 490
+    },
+    {
+      "epoch": 0.3,
+      "learning_rate": 0.00022302758178319434,
+      "loss": 0.6512,
+      "step": 500
+    },
+    {
+      "epoch": 0.31,
+      "learning_rate": 0.0002211032713277742,
+      "loss": 0.6402,
+      "step": 510
+    },
+    {
+      "epoch": 0.31,
+      "learning_rate": 0.00021917896087235406,
+      "loss": 0.6545,
+      "step": 520
+    },
+    {
+      "epoch": 0.32,
+      "learning_rate": 0.0002172546504169339,
+      "loss": 0.6495,
+      "step": 530
+    },
+    {
+      "epoch": 0.33,
+      "learning_rate": 0.0002153303399615138,
+      "loss": 0.6452,
+      "step": 540
+    },
+    {
+      "epoch": 0.33,
+      "learning_rate": 0.00021340602950609364,
+      "loss": 0.6548,
+      "step": 550
+    },
+    {
+      "epoch": 0.34,
+      "learning_rate": 0.00021148171905067349,
+      "loss": 0.6528,
+      "step": 560
+    },
+    {
+      "epoch": 0.34,
+      "learning_rate": 0.00020955740859525336,
+      "loss": 0.6514,
+      "step": 570
+    },
+    {
+      "epoch": 0.35,
+      "learning_rate": 0.0002076330981398332,
+      "loss": 0.6366,
+      "step": 580
+    },
+    {
+      "epoch": 0.36,
+      "learning_rate": 0.00020570878768441306,
+      "loss": 0.6457,
+      "step": 590
+    },
+    {
+      "epoch": 0.36,
+      "learning_rate": 0.00020378447722899294,
+      "loss": 0.6469,
+      "step": 600
+    },
+    {
+      "epoch": 0.36,
+      "eval_loss": 0.6427023410797119,
+      "eval_runtime": 26.0393,
+      "eval_samples_per_second": 76.807,
+      "eval_steps_per_second": 2.419,
+      "step": 600
+    },
+    {
+      "epoch": 0.37,
+      "learning_rate": 0.00020186016677357278,
+      "loss": 0.654,
+      "step": 610
+    },
+    {
+      "epoch": 0.37,
+      "learning_rate": 0.00019993585631815263,
+      "loss": 0.6449,
+      "step": 620
+    },
+    {
+      "epoch": 0.38,
+      "learning_rate": 0.0001980115458627325,
+      "loss": 0.6459,
+      "step": 630
+    },
+    {
+      "epoch": 0.39,
+      "learning_rate": 0.00019608723540731236,
+      "loss": 0.6454,
+      "step": 640
+    },
+    {
+      "epoch": 0.39,
+      "learning_rate": 0.0001941629249518922,
+      "loss": 0.641,
+      "step": 650
+    },
+    {
+      "epoch": 0.4,
+      "learning_rate": 0.00019223861449647208,
+      "loss": 0.64,
+      "step": 660
+    },
+    {
+      "epoch": 0.4,
+      "learning_rate": 0.00019031430404105193,
+      "loss": 0.6366,
+      "step": 670
+    },
+    {
+      "epoch": 0.41,
+      "learning_rate": 0.00018838999358563178,
+      "loss": 0.6452,
+      "step": 680
+    },
+    {
+      "epoch": 0.42,
+      "learning_rate": 0.00018646568313021166,
+      "loss": 0.6441,
+      "step": 690
+    },
+    {
+      "epoch": 0.42,
+      "learning_rate": 0.0001845413726747915,
+      "loss": 0.6441,
+      "step": 700
+    },
+    {
+      "epoch": 0.43,
+      "learning_rate": 0.00018261706221937138,
+      "loss": 0.6496,
+      "step": 710
+    },
+    {
+      "epoch": 0.43,
+      "learning_rate": 0.00018069275176395123,
+      "loss": 0.6357,
+      "step": 720
+    },
+    {
+      "epoch": 0.44,
+      "learning_rate": 0.00017876844130853108,
+      "loss": 0.6347,
+      "step": 730
+    },
+    {
+      "epoch": 0.45,
+      "learning_rate": 0.00017684413085311096,
+      "loss": 0.6345,
+      "step": 740
+    },
+    {
+      "epoch": 0.45,
+      "learning_rate": 0.0001749198203976908,
+      "loss": 0.6397,
+      "step": 750
+    },
+    {
+      "epoch": 0.46,
+      "learning_rate": 0.00017299550994227065,
+      "loss": 0.6444,
+      "step": 760
+    },
+    {
+      "epoch": 0.46,
+      "learning_rate": 0.00017107119948685053,
+      "loss": 0.64,
+      "step": 770
+    },
+    {
+      "epoch": 0.47,
+      "learning_rate": 0.00016914688903143038,
+      "loss": 0.6514,
+      "step": 780
+    },
+    {
+      "epoch": 0.48,
+      "learning_rate": 0.00016722257857601023,
+      "loss": 0.6424,
+      "step": 790
+    },
+    {
+      "epoch": 0.48,
+      "learning_rate": 0.0001652982681205901,
+      "loss": 0.643,
+      "step": 800
+    },
+    {
+      "epoch": 0.48,
+      "eval_loss": 0.6385083794593811,
+      "eval_runtime": 25.9034,
+      "eval_samples_per_second": 77.21,
+      "eval_steps_per_second": 2.432,
+      "step": 800
+    },
+    {
+      "epoch": 0.49,
+      "learning_rate": 0.00016337395766516995,
+      "loss": 0.6329,
+      "step": 810
+    },
+    {
+      "epoch": 0.49,
+      "learning_rate": 0.0001614496472097498,
+      "loss": 0.6426,
+      "step": 820
+    },
+    {
+      "epoch": 0.5,
+      "learning_rate": 0.00015952533675432968,
+      "loss": 0.6454,
+      "step": 830
+    },
+    {
+      "epoch": 0.51,
+      "learning_rate": 0.00015760102629890953,
+      "loss": 0.6357,
+      "step": 840
+    },
+    {
+      "epoch": 0.51,
+      "learning_rate": 0.00015567671584348943,
+      "loss": 0.6392,
+      "step": 850
+    },
+    {
+      "epoch": 0.52,
+      "learning_rate": 0.00015375240538806928,
+      "loss": 0.6408,
+      "step": 860
+    },
+    {
+      "epoch": 0.52,
+      "learning_rate": 0.0001518280949326491,
+      "loss": 0.6369,
+      "step": 870
+    },
+    {
+      "epoch": 0.53,
+      "learning_rate": 0.00014990378447722898,
+      "loss": 0.6354,
+      "step": 880
+    },
+    {
+      "epoch": 0.54,
+      "learning_rate": 0.00014797947402180885,
+      "loss": 0.6359,
+      "step": 890
+    },
+    {
+      "epoch": 0.54,
+      "learning_rate": 0.0001460551635663887,
+      "loss": 0.6308,
+      "step": 900
+    },
+    {
+      "epoch": 0.55,
+      "learning_rate": 0.00014413085311096855,
+      "loss": 0.6442,
+      "step": 910
+    },
+    {
+      "epoch": 0.55,
+      "learning_rate": 0.00014220654265554843,
+      "loss": 0.6394,
+      "step": 920
+    },
+    {
+      "epoch": 0.56,
+      "learning_rate": 0.00014028223220012828,
+      "loss": 0.6494,
+      "step": 930
+    },
+    {
+      "epoch": 0.57,
+      "learning_rate": 0.00013835792174470812,
+      "loss": 0.6413,
+      "step": 940
+    },
+    {
+      "epoch": 0.57,
+      "learning_rate": 0.000136433611289288,
+      "loss": 0.6373,
+      "step": 950
+    },
+    {
+      "epoch": 0.58,
+      "learning_rate": 0.00013450930083386785,
+      "loss": 0.6367,
+      "step": 960
+    },
+    {
+      "epoch": 0.58,
+      "learning_rate": 0.0001325849903784477,
+      "loss": 0.6398,
+      "step": 970
+    },
+    {
+      "epoch": 0.59,
+      "learning_rate": 0.00013066067992302757,
+      "loss": 0.6413,
+      "step": 980
+    },
+    {
+      "epoch": 0.6,
+      "learning_rate": 0.00012873636946760742,
+      "loss": 0.6334,
+      "step": 990
+    },
+    {
+      "epoch": 0.6,
+      "learning_rate": 0.0001268120590121873,
+      "loss": 0.6331,
+      "step": 1000
+    },
+    {
+      "epoch": 0.6,
+      "eval_loss": 0.6355478763580322,
+      "eval_runtime": 25.8942,
+      "eval_samples_per_second": 77.237,
+      "eval_steps_per_second": 2.433,
+      "step": 1000
+    },
+    {
+      "epoch": 0.61,
+      "learning_rate": 0.00012488774855676715,
+      "loss": 0.6412,
+      "step": 1010
+    },
+    {
+      "epoch": 0.61,
+      "learning_rate": 0.000122963438101347,
+      "loss": 0.6353,
+      "step": 1020
+    },
+    {
+      "epoch": 0.62,
+      "learning_rate": 0.00012103912764592687,
+      "loss": 0.6315,
+      "step": 1030
+    },
+    {
+      "epoch": 0.63,
+      "learning_rate": 0.00011911481719050672,
+      "loss": 0.638,
+      "step": 1040
+    },
+    {
+      "epoch": 0.63,
+      "learning_rate": 0.00011719050673508658,
+      "loss": 0.6305,
+      "step": 1050
+    },
+    {
+      "epoch": 0.64,
+      "learning_rate": 0.00011526619627966645,
+      "loss": 0.6305,
+      "step": 1060
+    },
+    {
+      "epoch": 0.64,
+      "learning_rate": 0.00011334188582424631,
+      "loss": 0.6307,
+      "step": 1070
+    },
+    {
+      "epoch": 0.65,
+      "learning_rate": 0.00011141757536882616,
+      "loss": 0.6394,
+      "step": 1080
+    },
+    {
+      "epoch": 0.66,
+      "learning_rate": 0.00010949326491340602,
+      "loss": 0.6331,
+      "step": 1090
+    },
+    {
+      "epoch": 0.66,
+      "learning_rate": 0.00010756895445798588,
+      "loss": 0.6333,
+      "step": 1100
+    },
+    {
+      "epoch": 0.67,
+      "learning_rate": 0.00010564464400256573,
+      "loss": 0.6434,
+      "step": 1110
+    },
+    {
+      "epoch": 0.68,
+      "learning_rate": 0.0001037203335471456,
+      "loss": 0.6349,
+      "step": 1120
+    },
+    {
+      "epoch": 0.68,
+      "learning_rate": 0.00010179602309172546,
+      "loss": 0.6454,
+      "step": 1130
+    },
+    {
+      "epoch": 0.69,
+      "learning_rate": 9.987171263630532e-05,
+      "loss": 0.6365,
+      "step": 1140
+    },
+    {
+      "epoch": 0.69,
+      "learning_rate": 9.794740218088517e-05,
+      "loss": 0.6338,
+      "step": 1150
+    },
+    {
+      "epoch": 0.7,
+      "learning_rate": 9.602309172546503e-05,
+      "loss": 0.6405,
+      "step": 1160
+    },
+    {
+      "epoch": 0.71,
+      "learning_rate": 9.40987812700449e-05,
+      "loss": 0.6338,
+      "step": 1170
+    },
+    {
+      "epoch": 0.71,
+      "learning_rate": 9.217447081462474e-05,
+      "loss": 0.6363,
+      "step": 1180
+    },
+    {
+      "epoch": 0.72,
+      "learning_rate": 9.02501603592046e-05,
+      "loss": 0.6403,
+      "step": 1190
+    },
+    {
+      "epoch": 0.72,
+      "learning_rate": 8.832584990378447e-05,
+      "loss": 0.6391,
+      "step": 1200
+    },
+    {
+      "epoch": 0.72,
+      "eval_loss": 0.633297860622406,
+      "eval_runtime": 25.8864,
+      "eval_samples_per_second": 77.261,
+      "eval_steps_per_second": 2.434,
+      "step": 1200
+    },
+    {
+      "epoch": 0.73,
+      "learning_rate": 8.640153944836432e-05,
+      "loss": 0.6349,
+      "step": 1210
+    },
+    {
+      "epoch": 0.74,
+      "learning_rate": 8.447722899294418e-05,
+      "loss": 0.6294,
+      "step": 1220
+    },
+    {
+      "epoch": 0.74,
+      "learning_rate": 8.255291853752406e-05,
+      "loss": 0.6292,
+      "step": 1230
+    },
+    {
+      "epoch": 0.75,
+      "learning_rate": 8.062860808210392e-05,
+      "loss": 0.6371,
+      "step": 1240
+    },
+    {
+      "epoch": 0.75,
+      "learning_rate": 7.870429762668377e-05,
+      "loss": 0.6342,
+      "step": 1250
+    },
+    {
+      "epoch": 0.76,
+      "learning_rate": 7.677998717126363e-05,
+      "loss": 0.6387,
+      "step": 1260
+    },
+    {
+      "epoch": 0.77,
+      "learning_rate": 7.485567671584348e-05,
+      "loss": 0.6405,
+      "step": 1270
+    },
+    {
+      "epoch": 0.77,
+      "learning_rate": 7.293136626042334e-05,
+      "loss": 0.627,
+      "step": 1280
+    },
+    {
+      "epoch": 0.78,
+      "learning_rate": 7.10070558050032e-05,
+      "loss": 0.6369,
+      "step": 1290
+    },
+    {
+      "epoch": 0.78,
+      "learning_rate": 6.908274534958307e-05,
+      "loss": 0.6339,
+      "step": 1300
+    },
+    {
+      "epoch": 0.79,
+      "learning_rate": 6.715843489416291e-05,
+      "loss": 0.6353,
+      "step": 1310
+    },
+    {
+      "epoch": 0.8,
+      "learning_rate": 6.523412443874278e-05,
+      "loss": 0.6372,
+      "step": 1320
+    },
+    {
+      "epoch": 0.8,
+      "learning_rate": 6.330981398332264e-05,
+      "loss": 0.632,
+      "step": 1330
+    },
+    {
+      "epoch": 0.81,
+      "learning_rate": 6.138550352790249e-05,
+      "loss": 0.6368,
+      "step": 1340
+    },
+    {
+      "epoch": 0.81,
+      "learning_rate": 5.946119307248236e-05,
+      "loss": 0.6263,
+      "step": 1350
+    },
+    {
+      "epoch": 0.82,
+      "learning_rate": 5.753688261706221e-05,
+      "loss": 0.6305,
+      "step": 1360
+    },
+    {
+      "epoch": 0.83,
+      "learning_rate": 5.5612572161642076e-05,
+      "loss": 0.6441,
+      "step": 1370
+    },
+    {
+      "epoch": 0.83,
+      "learning_rate": 5.368826170622193e-05,
+      "loss": 0.6382,
+      "step": 1380
+    },
+    {
+      "epoch": 0.84,
+      "learning_rate": 5.176395125080179e-05,
+      "loss": 0.6389,
+      "step": 1390
+    },
+    {
+      "epoch": 0.84,
+      "learning_rate": 4.983964079538165e-05,
+      "loss": 0.6277,
+      "step": 1400
+    },
+    {
+      "epoch": 0.84,
+      "eval_loss": 0.6312016844749451,
+      "eval_runtime": 25.972,
+      "eval_samples_per_second": 77.006,
+      "eval_steps_per_second": 2.426,
+      "step": 1400
+    },
+    {
+      "epoch": 0.85,
+      "learning_rate": 4.7915330339961505e-05,
+      "loss": 0.6277,
+      "step": 1410
+    },
+    {
+      "epoch": 0.86,
+      "learning_rate": 4.5991019884541375e-05,
+      "loss": 0.6338,
+      "step": 1420
+    },
+    {
+      "epoch": 0.86,
+      "learning_rate": 4.406670942912123e-05,
+      "loss": 0.6328,
+      "step": 1430
+    },
+    {
+      "epoch": 0.87,
+      "learning_rate": 4.2142398973701086e-05,
+      "loss": 0.6348,
+      "step": 1440
+    },
+    {
+      "epoch": 0.87,
+      "learning_rate": 4.021808851828095e-05,
+      "loss": 0.6229,
+      "step": 1450
+    },
+    {
+      "epoch": 0.88,
+      "learning_rate": 3.8293778062860804e-05,
+      "loss": 0.6326,
+      "step": 1460
+    },
+    {
+      "epoch": 0.89,
+      "learning_rate": 3.636946760744066e-05,
+      "loss": 0.6413,
+      "step": 1470
+    },
+    {
+      "epoch": 0.89,
+      "learning_rate": 3.444515715202052e-05,
+      "loss": 0.633,
+      "step": 1480
+    },
+    {
+      "epoch": 0.9,
+      "learning_rate": 3.2520846696600385e-05,
+      "loss": 0.6201,
+      "step": 1490
+    },
+    {
+      "epoch": 0.9,
+      "learning_rate": 3.059653624118024e-05,
+      "loss": 0.6253,
+      "step": 1500
+    },
+    {
+      "epoch": 0.91,
+      "learning_rate": 2.8672225785760103e-05,
+      "loss": 0.6329,
+      "step": 1510
+    },
+    {
+      "epoch": 0.92,
+      "learning_rate": 2.6747915330339955e-05,
+      "loss": 0.6405,
+      "step": 1520
+    },
+    {
+      "epoch": 0.92,
+      "learning_rate": 2.4823604874919818e-05,
+      "loss": 0.634,
+      "step": 1530
+    },
+    {
+      "epoch": 0.93,
+      "learning_rate": 2.2899294419499677e-05,
+      "loss": 0.6355,
+      "step": 1540
+    },
+    {
+      "epoch": 0.93,
+      "learning_rate": 2.0974983964079536e-05,
+      "loss": 0.6302,
+      "step": 1550
+    },
+    {
+      "epoch": 0.94,
+      "learning_rate": 1.9050673508659395e-05,
+      "loss": 0.6372,
+      "step": 1560
+    },
+    {
+      "epoch": 0.95,
+      "learning_rate": 1.7126363053239254e-05,
+      "loss": 0.6224,
+      "step": 1570
+    },
+    {
+      "epoch": 0.95,
+      "learning_rate": 1.5202052597819113e-05,
+      "loss": 0.6335,
+      "step": 1580
+    },
+    {
+      "epoch": 0.96,
+      "learning_rate": 1.3277742142398974e-05,
+      "loss": 0.6429,
+      "step": 1590
+    },
+    {
+      "epoch": 0.96,
+      "learning_rate": 1.1353431686978832e-05,
+      "loss": 0.6368,
+      "step": 1600
+    },
+    {
+      "epoch": 0.96,
+      "eval_loss": 0.6304358839988708,
+      "eval_runtime": 25.8881,
+      "eval_samples_per_second": 77.255,
+      "eval_steps_per_second": 2.434,
+      "step": 1600
+    }
+  ],
+  "max_steps": 1659,
+  "num_train_epochs": 1,
+  "total_flos": 5.199220197147279e+18,
+  "trial_name": null,
+  "trial_params": null
+}

checkpoint-1600/training_args.bin

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:e452eb637617b1a45640f786a0e85a379ffbe34b5501d894e89bab5a6af5efe0
+size 3899