diff --git a/checkpoint-5700/adapter_model/README.md b/checkpoint-5700/adapter_model/README.md
new file mode 100644
index 0000000000000000000000000000000000000000..f2208b0ded6c10ed47b2ea9df5ab7c8dd721a53c
--- /dev/null
+++ b/checkpoint-5700/adapter_model/README.md
@@ -0,0 +1,20 @@
+---
+library_name: peft
+---
+## Training procedure
+
+
+The following `bitsandbytes` quantization config was used during training:
+- load_in_8bit: False
+- load_in_4bit: True
+- llm_int8_threshold: 6.0
+- llm_int8_skip_modules: None
+- llm_int8_enable_fp32_cpu_offload: False
+- llm_int8_has_fp16_weight: False
+- bnb_4bit_quant_type: nf4
+- bnb_4bit_use_double_quant: True
+- bnb_4bit_compute_dtype: bfloat16
+### Framework versions
+
+
+- PEFT 0.5.0.dev0
diff --git a/checkpoint-5700/adapter_model/adapter_config.json b/checkpoint-5700/adapter_model/adapter_config.json
new file mode 100644
index 0000000000000000000000000000000000000000..7052646debaf453de93d6176727714122c31b64a
--- /dev/null
+++ b/checkpoint-5700/adapter_model/adapter_config.json
@@ -0,0 +1,26 @@
+{
+  "auto_mapping": null,
+  "base_model_name_or_path": "/workspace/webui/models/TheBloke_Llama-2-13B-fp16",
+  "bias": "none",
+  "fan_in_fan_out": null,
+  "inference_mode": true,
+  "init_lora_weights": true,
+  "layers_pattern": null,
+  "layers_to_transform": null,
+  "lora_alpha": 16,
+  "lora_dropout": 0.05,
+  "modules_to_save": null,
+  "peft_type": "LORA",
+  "r": 32,
+  "revision": null,
+  "target_modules": [
+    "q_proj",
+    "v_proj",
+    "gate_proj",
+    "up_proj",
+    "o_proj",
+    "k_proj",
+    "down_proj"
+  ],
+  "task_type": "CAUSAL_LM"
+}
\ No newline at end of file
diff --git a/checkpoint-5700/adapter_model/adapter_model.bin b/checkpoint-5700/adapter_model/adapter_model.bin
new file mode 100644
index 0000000000000000000000000000000000000000..02526e93c7101a3c9d0727acfe024831696398f1
--- /dev/null
+++ b/checkpoint-5700/adapter_model/adapter_model.bin
@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:6bdd5402d919c62a264000ec96b1dd621956e56ddd66679cf8a429f111552d95
+size 500897101
diff --git a/checkpoint-5800/README.md b/checkpoint-5800/README.md
new file mode 100644
index 0000000000000000000000000000000000000000..f2208b0ded6c10ed47b2ea9df5ab7c8dd721a53c
--- /dev/null
+++ b/checkpoint-5800/README.md
@@ -0,0 +1,20 @@
+---
+library_name: peft
+---
+## Training procedure
+
+
+The following `bitsandbytes` quantization config was used during training:
+- load_in_8bit: False
+- load_in_4bit: True
+- llm_int8_threshold: 6.0
+- llm_int8_skip_modules: None
+- llm_int8_enable_fp32_cpu_offload: False
+- llm_int8_has_fp16_weight: False
+- bnb_4bit_quant_type: nf4
+- bnb_4bit_use_double_quant: True
+- bnb_4bit_compute_dtype: bfloat16
+### Framework versions
+
+
+- PEFT 0.5.0.dev0
diff --git a/checkpoint-5800/adapter_config.json b/checkpoint-5800/adapter_config.json
new file mode 100644
index 0000000000000000000000000000000000000000..7052646debaf453de93d6176727714122c31b64a
--- /dev/null
+++ b/checkpoint-5800/adapter_config.json
@@ -0,0 +1,26 @@
+{
+  "auto_mapping": null,
+  "base_model_name_or_path": "/workspace/webui/models/TheBloke_Llama-2-13B-fp16",
+  "bias": "none",
+  "fan_in_fan_out": null,
+  "inference_mode": true,
+  "init_lora_weights": true,
+  "layers_pattern": null,
+  "layers_to_transform": null,
+  "lora_alpha": 16,
+  "lora_dropout": 0.05,
+  "modules_to_save": null,
+  "peft_type": "LORA",
+  "r": 32,
+  "revision": null,
+  "target_modules": [
+    "q_proj",
+    "v_proj",
+    "gate_proj",
+    "up_proj",
+    "o_proj",
+    "k_proj",
+    "down_proj"
+  ],
+  "task_type": "CAUSAL_LM"
+}
\ No newline at end of file
diff --git a/checkpoint-5800/adapter_model.bin b/checkpoint-5800/adapter_model.bin
new file mode 100644
index 0000000000000000000000000000000000000000..b25d3eaace758c6638b956d27554c51cfd758425
--- /dev/null
+++ b/checkpoint-5800/adapter_model.bin
@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:c6c8948b163d7faf80bcb4ea3c791311da3d9b14d4257794aeb4a2b35935a025
+size 500897101
diff --git a/checkpoint-5800/adapter_model/README.md b/checkpoint-5800/adapter_model/README.md
new file mode 100644
index 0000000000000000000000000000000000000000..f2208b0ded6c10ed47b2ea9df5ab7c8dd721a53c
--- /dev/null
+++ b/checkpoint-5800/adapter_model/README.md
@@ -0,0 +1,20 @@
+---
+library_name: peft
+---
+## Training procedure
+
+
+The following `bitsandbytes` quantization config was used during training:
+- load_in_8bit: False
+- load_in_4bit: True
+- llm_int8_threshold: 6.0
+- llm_int8_skip_modules: None
+- llm_int8_enable_fp32_cpu_offload: False
+- llm_int8_has_fp16_weight: False
+- bnb_4bit_quant_type: nf4
+- bnb_4bit_use_double_quant: True
+- bnb_4bit_compute_dtype: bfloat16
+### Framework versions
+
+
+- PEFT 0.5.0.dev0
diff --git a/checkpoint-5800/adapter_model/adapter_config.json b/checkpoint-5800/adapter_model/adapter_config.json
new file mode 100644
index 0000000000000000000000000000000000000000..7052646debaf453de93d6176727714122c31b64a
--- /dev/null
+++ b/checkpoint-5800/adapter_model/adapter_config.json
@@ -0,0 +1,26 @@
+{
+  "auto_mapping": null,
+  "base_model_name_or_path": "/workspace/webui/models/TheBloke_Llama-2-13B-fp16",
+  "bias": "none",
+  "fan_in_fan_out": null,
+  "inference_mode": true,
+  "init_lora_weights": true,
+  "layers_pattern": null,
+  "layers_to_transform": null,
+  "lora_alpha": 16,
+  "lora_dropout": 0.05,
+  "modules_to_save": null,
+  "peft_type": "LORA",
+  "r": 32,
+  "revision": null,
+  "target_modules": [
+    "q_proj",
+    "v_proj",
+    "gate_proj",
+    "up_proj",
+    "o_proj",
+    "k_proj",
+    "down_proj"
+  ],
+  "task_type": "CAUSAL_LM"
+}
\ No newline at end of file
diff --git a/checkpoint-5800/adapter_model/adapter_model.bin b/checkpoint-5800/adapter_model/adapter_model.bin
new file mode 100644
index 0000000000000000000000000000000000000000..b25d3eaace758c6638b956d27554c51cfd758425
--- /dev/null
+++ b/checkpoint-5800/adapter_model/adapter_model.bin
@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:c6c8948b163d7faf80bcb4ea3c791311da3d9b14d4257794aeb4a2b35935a025
+size 500897101
diff --git a/checkpoint-5800/optimizer.pt b/checkpoint-5800/optimizer.pt
new file mode 100644
index 0000000000000000000000000000000000000000..3ff99dd4e8de03fbb3706e831716b20cdc36b029
--- /dev/null
+++ b/checkpoint-5800/optimizer.pt
@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:fdafcf17dcbaf493ce7420b5efc55cc37f121d6d18e479f66229dfb7d3ebfe9a
+size 1001752701
diff --git a/checkpoint-5800/rng_state_0.pth b/checkpoint-5800/rng_state_0.pth
new file mode 100644
index 0000000000000000000000000000000000000000..32576f7fea9a909e430c2379d78ec5bf846b1367
--- /dev/null
+++ b/checkpoint-5800/rng_state_0.pth
@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:fdcc97c3d5ae4ead7b5285c5c5b8dddcbec730d6aced698514214e40163f6c80
+size 27772
diff --git a/checkpoint-5800/rng_state_1.pth b/checkpoint-5800/rng_state_1.pth
new file mode 100644
index 0000000000000000000000000000000000000000..4427827aec35702513d131a1ce0ecb968ec72c8f
--- /dev/null
+++ b/checkpoint-5800/rng_state_1.pth
@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:8f0dfb5773126adda2e927345797eb8babdfc8aa673a963e413f06bda803f6b8
+size 27772
diff --git a/checkpoint-5800/rng_state_10.pth b/checkpoint-5800/rng_state_10.pth
new file mode 100644
index 0000000000000000000000000000000000000000..d7d02c96cf7f3da06779c0b9becde5f1e0927402
--- /dev/null
+++ b/checkpoint-5800/rng_state_10.pth
@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:481ee2d6343e3ee249f8f13ba72e97689ec745d34ca3eb731c5977d455e68087
+size 27789
diff --git a/checkpoint-5800/rng_state_11.pth b/checkpoint-5800/rng_state_11.pth
new file mode 100644
index 0000000000000000000000000000000000000000..02f266ae42f742e7af0f68aa48bb5e1d97f9854a
--- /dev/null
+++ b/checkpoint-5800/rng_state_11.pth
@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:56d4288a9e908f82added428e5a14ff5dd6e86473ebd372733e241f3c2a4e833
+size 27789
diff --git a/checkpoint-5800/rng_state_12.pth b/checkpoint-5800/rng_state_12.pth
new file mode 100644
index 0000000000000000000000000000000000000000..a5bf2100846d316c907c004cdcf4f53dec78dbc4
--- /dev/null
+++ b/checkpoint-5800/rng_state_12.pth
@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:5749fda2bb629d9c3e4a0bb894676f98047fc123c7ada64e2385c1692f194369
+size 27789
diff --git a/checkpoint-5800/rng_state_13.pth b/checkpoint-5800/rng_state_13.pth
new file mode 100644
index 0000000000000000000000000000000000000000..6e4c8ec65ba23f1127297f14df1c0e5a815972bc
--- /dev/null
+++ b/checkpoint-5800/rng_state_13.pth
@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:2ac848792be0fda751cfdebaa1419f845e441f823e55d97dc332cb6ebecab888
+size 27789
diff --git a/checkpoint-5800/rng_state_2.pth b/checkpoint-5800/rng_state_2.pth
new file mode 100644
index 0000000000000000000000000000000000000000..339b13aaf629629acd363207d06b803f3164cc48
--- /dev/null
+++ b/checkpoint-5800/rng_state_2.pth
@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:004ca3f2557bec1000ed01ad0ac091380c145ff4d9054e495042d9673b164cd3
+size 27772
diff --git a/checkpoint-5800/rng_state_3.pth b/checkpoint-5800/rng_state_3.pth
new file mode 100644
index 0000000000000000000000000000000000000000..fdc8655bc3b73b1d48fe9069eff355b73e1cd4de
--- /dev/null
+++ b/checkpoint-5800/rng_state_3.pth
@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:83a6e03a2771bdfefb81f6b70a469b27bf46d9bac7c91cec25fbd8fc2e1bb9fd
+size 27772
diff --git a/checkpoint-5800/rng_state_4.pth b/checkpoint-5800/rng_state_4.pth
new file mode 100644
index 0000000000000000000000000000000000000000..a33983d78492d1d9d9a320f02bab9fd7437fd447
--- /dev/null
+++ b/checkpoint-5800/rng_state_4.pth
@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:027cc230cd1577ece8a43abc1ed88fcc51ac874f84b8bcbe865b8338b17825dc
+size 27772
diff --git a/checkpoint-5800/rng_state_5.pth b/checkpoint-5800/rng_state_5.pth
new file mode 100644
index 0000000000000000000000000000000000000000..9c12ce4de318110b659e68db3eb100c4aa128f72
--- /dev/null
+++ b/checkpoint-5800/rng_state_5.pth
@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:5e15c6848b49b0375761f21215e2d56ea5a16e27885192cc4af7b1a3f9c325a2
+size 27772
diff --git a/checkpoint-5800/rng_state_6.pth b/checkpoint-5800/rng_state_6.pth
new file mode 100644
index 0000000000000000000000000000000000000000..2fcdb815d79ea791d2753a0db4b146458c95e3d7
--- /dev/null
+++ b/checkpoint-5800/rng_state_6.pth
@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:0290c211df0a25ca2b739876f096befd7a2f50010f23fcfb700c648fc7c37b88
+size 27772
diff --git a/checkpoint-5800/rng_state_7.pth b/checkpoint-5800/rng_state_7.pth
new file mode 100644
index 0000000000000000000000000000000000000000..c97a750c2dea4a44ea85d28ef92f4d91c42180b9
--- /dev/null
+++ b/checkpoint-5800/rng_state_7.pth
@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:fe7bdf51b5f2867802a2660bdbfa211be18d28a9f5e7325d4957f7ee65895a3a
+size 27772
diff --git a/checkpoint-5800/rng_state_8.pth b/checkpoint-5800/rng_state_8.pth
new file mode 100644
index 0000000000000000000000000000000000000000..eb02047c3a256a4e8dc65a9507f137abe27d4048
--- /dev/null
+++ b/checkpoint-5800/rng_state_8.pth
@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:0c7043986c3cfa7014abbd2eb87d4f8ccda04f6a243deda922c9522e5c66f2fd
+size 27772
diff --git a/checkpoint-5800/rng_state_9.pth b/checkpoint-5800/rng_state_9.pth
new file mode 100644
index 0000000000000000000000000000000000000000..45ac65bb884854298d454760ce2735e8ecec7624
--- /dev/null
+++ b/checkpoint-5800/rng_state_9.pth
@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:a3371033bac6e8e8c6fb2001ca45e33bd941b91fd370cce2d2edc6bac7df0e55
+size 27772
diff --git a/checkpoint-5800/scheduler.pt b/checkpoint-5800/scheduler.pt
new file mode 100644
index 0000000000000000000000000000000000000000..f744f85c5c20ec4e390e284fde44a23de16db37c
--- /dev/null
+++ b/checkpoint-5800/scheduler.pt
@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:d7e2a15463250851f8f0353fd93191418e2b4e589eb68be6185d12012957a33f
+size 627
diff --git a/checkpoint-5800/trainer_state.json b/checkpoint-5800/trainer_state.json
new file mode 100644
index 0000000000000000000000000000000000000000..253a82686a6ec59beff239685621e7860be41e45
--- /dev/null
+++ b/checkpoint-5800/trainer_state.json
@@ -0,0 +1,1424 @@
+{
+  "best_metric": null,
+  "best_model_checkpoint": null,
+  "epoch": 2.2489336952307095,
+  "global_step": 5800,
+  "is_hyper_param_search": false,
+  "is_local_process_zero": true,
+  "is_world_process_zero": true,
+  "log_history": [
+    {
+      "epoch": 0.02,
+      "learning_rate": 0.0001999867761371633,
+      "loss": 1.0435,
+      "step": 50
+    },
+    {
+      "epoch": 0.04,
+      "learning_rate": 0.00019993306018843102,
+      "loss": 0.8918,
+      "step": 100
+    },
+    {
+      "epoch": 0.06,
+      "learning_rate": 0.00019983804784290833,
+      "loss": 0.8874,
+      "step": 150
+    },
+    {
+      "epoch": 0.08,
+      "learning_rate": 0.00019970177836355307,
+      "loss": 0.8839,
+      "step": 200
+    },
+    {
+      "epoch": 0.09,
+      "learning_rate": 0.00019961818913082012,
+      "loss": 0.8801,
+      "step": 225
+    },
+    {
+      "epoch": 0.1,
+      "learning_rate": 0.00019952430806244534,
+      "loss": 0.8753,
+      "step": 250
+    },
+    {
+      "epoch": 0.11,
+      "learning_rate": 0.00019942014485754635,
+      "loss": 0.8754,
+      "step": 275
+    },
+    {
+      "epoch": 0.12,
+      "learning_rate": 0.00019930571027751713,
+      "loss": 0.8751,
+      "step": 300
+    },
+    {
+      "epoch": 0.13,
+      "learning_rate": 0.0001991810161449164,
+      "loss": 0.8819,
+      "step": 325
+    },
+    {
+      "epoch": 0.14,
+      "learning_rate": 0.00019904607534224612,
+      "loss": 0.8744,
+      "step": 350
+    },
+    {
+      "epoch": 0.15,
+      "learning_rate": 0.00019890090181062063,
+      "loss": 0.8735,
+      "step": 375
+    },
+    {
+      "epoch": 0.16,
+      "learning_rate": 0.00019874551054832625,
+      "loss": 0.8703,
+      "step": 400
+    },
+    {
+      "epoch": 0.16,
+      "learning_rate": 0.00019857991760927193,
+      "loss": 0.8715,
+      "step": 425
+    },
+    {
+      "epoch": 0.17,
+      "learning_rate": 0.00019840414010133045,
+      "loss": 0.8714,
+      "step": 450
+    },
+    {
+      "epoch": 0.18,
+      "learning_rate": 0.00019821819618457114,
+      "loss": 0.8653,
+      "step": 475
+    },
+    {
+      "epoch": 0.19,
+      "learning_rate": 0.0001980221050693837,
+      "loss": 0.8716,
+      "step": 500
+    },
+    {
+      "epoch": 0.2,
+      "learning_rate": 0.00019781588701449338,
+      "loss": 0.8695,
+      "step": 525
+    },
+    {
+      "epoch": 0.21,
+      "learning_rate": 0.0001975995633248682,
+      "loss": 0.8746,
+      "step": 550
+    },
+    {
+      "epoch": 0.22,
+      "learning_rate": 0.00019737315634951762,
+      "loss": 0.8731,
+      "step": 575
+    },
+    {
+      "epoch": 0.23,
+      "learning_rate": 0.00019713668947918386,
+      "loss": 0.867,
+      "step": 600
+    },
+    {
+      "epoch": 0.24,
+      "learning_rate": 0.0001968901871439252,
+      "loss": 0.8706,
+      "step": 625
+    },
+    {
+      "epoch": 0.25,
+      "learning_rate": 0.000196633674810592,
+      "loss": 0.8595,
+      "step": 650
+    },
+    {
+      "epoch": 0.26,
+      "learning_rate": 0.0001963671789801958,
+      "loss": 0.8627,
+      "step": 675
+    },
+    {
+      "epoch": 0.27,
+      "learning_rate": 0.0001960907271851712,
+      "loss": 0.8607,
+      "step": 700
+    },
+    {
+      "epoch": 0.28,
+      "learning_rate": 0.00019580434798653173,
+      "loss": 0.858,
+      "step": 725
+    },
+    {
+      "epoch": 0.29,
+      "learning_rate": 0.00019550807097091876,
+      "loss": 0.8589,
+      "step": 750
+    },
+    {
+      "epoch": 0.3,
+      "learning_rate": 0.00019520192674754515,
+      "loss": 0.8561,
+      "step": 775
+    },
+    {
+      "epoch": 0.31,
+      "learning_rate": 0.00019488594694503264,
+      "loss": 0.8576,
+      "step": 800
+    },
+    {
+      "epoch": 0.32,
+      "learning_rate": 0.00019456016420814446,
+      "loss": 0.8597,
+      "step": 825
+    },
+    {
+      "epoch": 0.33,
+      "learning_rate": 0.00019422461219441254,
+      "loss": 0.862,
+      "step": 850
+    },
+    {
+      "epoch": 0.34,
+      "learning_rate": 0.00019387932557066035,
+      "loss": 0.8577,
+      "step": 875
+    },
+    {
+      "epoch": 0.35,
+      "learning_rate": 0.00019352434000942127,
+      "loss": 0.8632,
+      "step": 900
+    },
+    {
+      "epoch": 0.36,
+      "learning_rate": 0.00019315969218525333,
+      "loss": 0.8567,
+      "step": 925
+    },
+    {
+      "epoch": 0.37,
+      "learning_rate": 0.00019278541977095005,
+      "loss": 0.8501,
+      "step": 950
+    },
+    {
+      "epoch": 0.38,
+      "learning_rate": 0.00019240156143364844,
+      "loss": 0.8596,
+      "step": 975
+    },
+    {
+      "epoch": 0.39,
+      "learning_rate": 0.00019200815683083434,
+      "loss": 0.8556,
+      "step": 1000
+    },
+    {
+      "epoch": 0.39,
+      "eval_loss": 0.8521950244903564,
+      "eval_runtime": 59.8838,
+      "eval_samples_per_second": 12.19,
+      "eval_steps_per_second": 0.885,
+      "step": 1000
+    },
+    {
+      "epoch": 0.4,
+      "learning_rate": 0.00019160524660624505,
+      "loss": 0.8531,
+      "step": 1025
+    },
+    {
+      "epoch": 0.41,
+      "learning_rate": 0.00019119287238567045,
+      "loss": 0.8513,
+      "step": 1050
+    },
+    {
+      "epoch": 0.42,
+      "learning_rate": 0.00019077107677265253,
+      "loss": 0.8502,
+      "step": 1075
+    },
+    {
+      "epoch": 0.43,
+      "learning_rate": 0.00019033990334408384,
+      "loss": 0.8469,
+      "step": 1100
+    },
+    {
+      "epoch": 0.44,
+      "learning_rate": 0.00018989939664570545,
+      "loss": 0.8495,
+      "step": 1125
+    },
+    {
+      "epoch": 0.45,
+      "learning_rate": 0.00018944960218750484,
+      "loss": 0.8485,
+      "step": 1150
+    },
+    {
+      "epoch": 0.46,
+      "learning_rate": 0.00018899056643901404,
+      "loss": 0.8534,
+      "step": 1175
+    },
+    {
+      "epoch": 0.47,
+      "learning_rate": 0.00018852233682450893,
+      "loss": 0.8531,
+      "step": 1200
+    },
+    {
+      "epoch": 0.47,
+      "learning_rate": 0.00018804496171810948,
+      "loss": 0.8509,
+      "step": 1225
+    },
+    {
+      "epoch": 0.48,
+      "learning_rate": 0.00018755849043878222,
+      "loss": 0.8445,
+      "step": 1250
+    },
+    {
+      "epoch": 0.49,
+      "learning_rate": 0.0001870629732452449,
+      "loss": 0.8548,
+      "step": 1275
+    },
+    {
+      "epoch": 0.5,
+      "learning_rate": 0.00018655846133077417,
+      "loss": 0.8441,
+      "step": 1300
+    },
+    {
+      "epoch": 0.51,
+      "learning_rate": 0.00018604500681791656,
+      "loss": 0.8533,
+      "step": 1325
+    },
+    {
+      "epoch": 0.52,
+      "learning_rate": 0.00018552266275310373,
+      "loss": 0.8505,
+      "step": 1350
+    },
+    {
+      "epoch": 0.53,
+      "learning_rate": 0.0001849914831011719,
+      "loss": 0.8544,
+      "step": 1375
+    },
+    {
+      "epoch": 0.54,
+      "learning_rate": 0.00018445152273978668,
+      "loss": 0.845,
+      "step": 1400
+    },
+    {
+      "epoch": 0.55,
+      "learning_rate": 0.00018390283745377354,
+      "loss": 0.8376,
+      "step": 1425
+    },
+    {
+      "epoch": 0.56,
+      "learning_rate": 0.0001833454839293545,
+      "loss": 0.847,
+      "step": 1450
+    },
+    {
+      "epoch": 0.57,
+      "learning_rate": 0.00018277951974829163,
+      "loss": 0.8473,
+      "step": 1475
+    },
+    {
+      "epoch": 0.58,
+      "learning_rate": 0.0001822050033819382,
+      "loss": 0.8438,
+      "step": 1500
+    },
+    {
+      "epoch": 0.59,
+      "learning_rate": 0.00018162199418519785,
+      "loss": 0.8418,
+      "step": 1525
+    },
+    {
+      "epoch": 0.6,
+      "learning_rate": 0.00018103055239039243,
+      "loss": 0.842,
+      "step": 1550
+    },
+    {
+      "epoch": 0.61,
+      "learning_rate": 0.0001804307391010393,
+      "loss": 0.8435,
+      "step": 1575
+    },
+    {
+      "epoch": 0.62,
+      "learning_rate": 0.00017982261628553842,
+      "loss": 0.8349,
+      "step": 1600
+    },
+    {
+      "epoch": 0.63,
+      "learning_rate": 0.0001792062467707703,
+      "loss": 0.8483,
+      "step": 1625
+    },
+    {
+      "epoch": 0.64,
+      "learning_rate": 0.0001785816942356052,
+      "loss": 0.8387,
+      "step": 1650
+    },
+    {
+      "epoch": 0.65,
+      "learning_rate": 0.00017794902320432429,
+      "loss": 0.843,
+      "step": 1675
+    },
+    {
+      "epoch": 0.66,
+      "learning_rate": 0.00017730829903995333,
+      "loss": 0.8424,
+      "step": 1700
+    },
+    {
+      "epoch": 0.67,
+      "learning_rate": 0.00017665958793751006,
+      "loss": 0.8418,
+      "step": 1725
+    },
+    {
+      "epoch": 0.68,
+      "learning_rate": 0.00017600295691716522,
+      "loss": 0.8384,
+      "step": 1750
+    },
+    {
+      "epoch": 0.69,
+      "learning_rate": 0.00017533847381731856,
+      "loss": 0.8445,
+      "step": 1775
+    },
+    {
+      "epoch": 0.7,
+      "learning_rate": 0.00017466620728759033,
+      "loss": 0.8446,
+      "step": 1800
+    },
+    {
+      "epoch": 0.71,
+      "learning_rate": 0.00017398622678172878,
+      "loss": 0.838,
+      "step": 1825
+    },
+    {
+      "epoch": 0.72,
+      "learning_rate": 0.0001732986025504348,
+      "loss": 0.8415,
+      "step": 1850
+    },
+    {
+      "epoch": 0.73,
+      "learning_rate": 0.000172603405634104,
+      "loss": 0.8357,
+      "step": 1875
+    },
+    {
+      "epoch": 0.74,
+      "learning_rate": 0.00017190070785548755,
+      "loss": 0.8311,
+      "step": 1900
+    },
+    {
+      "epoch": 0.75,
+      "learning_rate": 0.0001711905818122717,
+      "loss": 0.8333,
+      "step": 1925
+    },
+    {
+      "epoch": 0.76,
+      "learning_rate": 0.0001704731008695777,
+      "loss": 0.8387,
+      "step": 1950
+    },
+    {
+      "epoch": 0.77,
+      "learning_rate": 0.0001697483391523821,
+      "loss": 0.8442,
+      "step": 1975
+    },
+    {
+      "epoch": 0.78,
+      "learning_rate": 0.00016901637153785885,
+      "loss": 0.8399,
+      "step": 2000
+    },
+    {
+      "epoch": 0.78,
+      "eval_loss": 0.8339959383010864,
+      "eval_runtime": 58.5829,
+      "eval_samples_per_second": 12.461,
+      "eval_steps_per_second": 0.905,
+      "step": 2000
+    },
+    {
+      "epoch": 0.79,
+      "learning_rate": 0.0001682772736476434,
+      "loss": 0.8334,
+      "step": 2025
+    },
+    {
+      "epoch": 0.79,
+      "learning_rate": 0.0001675311218400201,
+      "loss": 0.835,
+      "step": 2050
+    },
+    {
+      "epoch": 0.8,
+      "learning_rate": 0.00016677799320203332,
+      "loss": 0.8368,
+      "step": 2075
+    },
+    {
+      "epoch": 0.81,
+      "learning_rate": 0.00016601796554152344,
+      "loss": 0.8278,
+      "step": 2100
+    },
+    {
+      "epoch": 0.82,
+      "learning_rate": 0.00016525111737908827,
+      "loss": 0.8334,
+      "step": 2125
+    },
+    {
+      "epoch": 0.83,
+      "learning_rate": 0.00016447752793997096,
+      "loss": 0.8416,
+      "step": 2150
+    },
+    {
+      "epoch": 0.84,
+      "learning_rate": 0.00016369727714587483,
+      "loss": 0.8297,
+      "step": 2175
+    },
+    {
+      "epoch": 0.85,
+      "learning_rate": 0.0001629104456067066,
+      "loss": 0.8327,
+      "step": 2200
+    },
+    {
+      "epoch": 0.86,
+      "learning_rate": 0.00016211711461224825,
+      "loss": 0.8324,
+      "step": 2225
+    },
+    {
+      "epoch": 0.87,
+      "learning_rate": 0.0001613173661237589,
+      "loss": 0.8313,
+      "step": 2250
+    },
+    {
+      "epoch": 0.88,
+      "learning_rate": 0.0001605112827655069,
+      "loss": 0.8292,
+      "step": 2275
+    },
+    {
+      "epoch": 0.89,
+      "learning_rate": 0.0001596989478162339,
+      "loss": 0.8334,
+      "step": 2300
+    },
+    {
+      "epoch": 0.9,
+      "learning_rate": 0.00015888044520055106,
+      "loss": 0.8352,
+      "step": 2325
+    },
+    {
+      "epoch": 0.91,
+      "learning_rate": 0.00015805585948026852,
+      "loss": 0.823,
+      "step": 2350
+    },
+    {
+      "epoch": 0.92,
+      "learning_rate": 0.000157225275845659,
+      "loss": 0.8293,
+      "step": 2375
+    },
+    {
+      "epoch": 0.93,
+      "learning_rate": 0.00015638878010665672,
+      "loss": 0.8289,
+      "step": 2400
+    },
+    {
+      "epoch": 0.94,
+      "learning_rate": 0.00015554645868399205,
+      "loss": 0.832,
+      "step": 2425
+    },
+    {
+      "epoch": 0.95,
+      "learning_rate": 0.00015469839860026308,
+      "loss": 0.8294,
+      "step": 2450
+    },
+    {
+      "epoch": 0.96,
+      "learning_rate": 0.0001538446874709452,
+      "loss": 0.8281,
+      "step": 2475
+    },
+    {
+      "epoch": 0.97,
+      "learning_rate": 0.00015298541349533925,
+      "loss": 0.8314,
+      "step": 2500
+    },
+    {
+      "epoch": 0.98,
+      "learning_rate": 0.00015212066544745926,
+      "loss": 0.831,
+      "step": 2525
+    },
+    {
+      "epoch": 0.99,
+      "learning_rate": 0.00015125053266686124,
+      "loss": 0.8319,
+      "step": 2550
+    },
+    {
+      "epoch": 1.0,
+      "learning_rate": 0.00015037510504941303,
+      "loss": 0.8259,
+      "step": 2575
+    },
+    {
+      "epoch": 1.01,
+      "learning_rate": 0.00014949447303800695,
+      "loss": 0.8133,
+      "step": 2600
+    },
+    {
+      "epoch": 1.02,
+      "learning_rate": 0.00014860872761321593,
+      "loss": 0.8139,
+      "step": 2625
+    },
+    {
+      "epoch": 1.03,
+      "learning_rate": 0.00014771796028389405,
+      "loss": 0.804,
+      "step": 2650
+    },
+    {
+      "epoch": 1.04,
+      "learning_rate": 0.0001468222630777225,
+      "loss": 0.8011,
+      "step": 2675
+    },
+    {
+      "epoch": 1.05,
+      "learning_rate": 0.00014592172853170193,
+      "loss": 0.8037,
+      "step": 2700
+    },
+    {
+      "epoch": 1.06,
+      "learning_rate": 0.00014501644968259212,
+      "loss": 0.8063,
+      "step": 2725
+    },
+    {
+      "epoch": 1.07,
+      "learning_rate": 0.00014410652005730025,
+      "loss": 0.8155,
+      "step": 2750
+    },
+    {
+      "epoch": 1.08,
+      "learning_rate": 0.00014319203366321826,
+      "loss": 0.8066,
+      "step": 2775
+    },
+    {
+      "epoch": 1.09,
+      "learning_rate": 0.0001422730849785107,
+      "loss": 0.8091,
+      "step": 2800
+    },
+    {
+      "epoch": 1.1,
+      "learning_rate": 0.0001413497689423539,
+      "loss": 0.8067,
+      "step": 2825
+    },
+    {
+      "epoch": 1.11,
+      "learning_rate": 0.00014042218094512755,
+      "loss": 0.8046,
+      "step": 2850
+    },
+    {
+      "epoch": 1.11,
+      "learning_rate": 0.00013949041681855985,
+      "loss": 0.8053,
+      "step": 2875
+    },
+    {
+      "epoch": 1.12,
+      "learning_rate": 0.0001385545728258264,
+      "loss": 0.8075,
+      "step": 2900
+    },
+    {
+      "epoch": 1.13,
+      "learning_rate": 0.0001376147456516055,
+      "loss": 0.8015,
+      "step": 2925
+    },
+    {
+      "epoch": 1.14,
+      "learning_rate": 0.00013667103239208903,
+      "loss": 0.8016,
+      "step": 2950
+    },
+    {
+      "epoch": 1.15,
+      "learning_rate": 0.00013572353054495126,
+      "loss": 0.8029,
+      "step": 2975
+    },
+    {
+      "epoch": 1.16,
+      "learning_rate": 0.0001347723379992762,
+      "loss": 0.8017,
+      "step": 3000
+    },
+    {
+      "epoch": 1.16,
+      "eval_loss": 0.8229297995567322,
+      "eval_runtime": 59.3398,
+      "eval_samples_per_second": 12.302,
+      "eval_steps_per_second": 0.893,
+      "step": 3000
+    },
+    {
+      "epoch": 1.17,
+      "learning_rate": 0.0001338175530254443,
+      "loss": 0.8049,
+      "step": 3025
+    },
+    {
+      "epoch": 1.18,
+      "learning_rate": 0.00013285927426497985,
+      "loss": 0.8027,
+      "step": 3050
+    },
+    {
+      "epoch": 1.19,
+      "learning_rate": 0.00013189760072036008,
+      "loss": 0.8028,
+      "step": 3075
+    },
+    {
+      "epoch": 1.2,
+      "learning_rate": 0.0001309326317447869,
+      "loss": 0.8021,
+      "step": 3100
+    },
+    {
+      "epoch": 1.21,
+      "learning_rate": 0.00012996446703192257,
+      "loss": 0.8033,
+      "step": 3125
+    },
+    {
+      "epoch": 1.22,
+      "learning_rate": 0.00012899320660558986,
+      "loss": 0.8016,
+      "step": 3150
+    },
+    {
+      "epoch": 1.23,
+      "learning_rate": 0.00012801895080943846,
+      "loss": 0.7995,
+      "step": 3175
+    },
+    {
+      "epoch": 1.24,
+      "learning_rate": 0.0001270418002965782,
+      "loss": 0.799,
+      "step": 3200
+    },
+    {
+      "epoch": 1.25,
+      "learning_rate": 0.0001260618560191802,
+      "loss": 0.8002,
+      "step": 3225
+    },
+    {
+      "epoch": 1.26,
+      "learning_rate": 0.00012507921921804717,
+      "loss": 0.8068,
+      "step": 3250
+    },
+    {
+      "epoch": 1.27,
+      "learning_rate": 0.00012409399141215423,
+      "loss": 0.8041,
+      "step": 3275
+    },
+    {
+      "epoch": 1.28,
+      "learning_rate": 0.0001231062743881603,
+      "loss": 0.7999,
+      "step": 3300
+    },
+    {
+      "epoch": 1.29,
+      "learning_rate": 0.0001221161701898926,
+      "loss": 0.7995,
+      "step": 3325
+    },
+    {
+      "epoch": 1.3,
+      "learning_rate": 0.00012112378110780391,
+      "loss": 0.7959,
+      "step": 3350
+    },
+    {
+      "epoch": 1.31,
+      "learning_rate": 0.00012012920966840486,
+      "loss": 0.7999,
+      "step": 3375
+    },
+    {
+      "epoch": 1.32,
+      "learning_rate": 0.00011913255862367151,
+      "loss": 0.8016,
+      "step": 3400
+    },
+    {
+      "epoch": 1.33,
+      "learning_rate": 0.00011813393094042993,
+      "loss": 0.7944,
+      "step": 3425
+    },
+    {
+      "epoch": 1.34,
+      "learning_rate": 0.0001171334297897181,
+      "loss": 0.8026,
+      "step": 3450
+    },
+    {
+      "epoch": 1.35,
+      "learning_rate": 0.00011613115853612734,
+      "loss": 0.8004,
+      "step": 3475
+    },
+    {
+      "epoch": 1.36,
+      "learning_rate": 0.00011512722072712321,
+      "loss": 0.7992,
+      "step": 3500
+    },
+    {
+      "epoch": 1.37,
+      "learning_rate": 0.00011412172008234785,
+      "loss": 0.8004,
+      "step": 3525
+    },
+    {
+      "epoch": 1.38,
+      "learning_rate": 0.0001131147604829043,
+      "loss": 0.8009,
+      "step": 3550
+    },
+    {
+      "epoch": 1.39,
+      "learning_rate": 0.00011210644596062439,
+      "loss": 0.7993,
+      "step": 3575
+    },
+    {
+      "epoch": 1.4,
+      "learning_rate": 0.00011109688068732081,
+      "loss": 0.7965,
+      "step": 3600
+    },
+    {
+      "epoch": 1.41,
+      "learning_rate": 0.00011008616896402482,
+      "loss": 0.7991,
+      "step": 3625
+    },
+    {
+      "epoch": 1.42,
+      "learning_rate": 0.00010907441521021072,
+      "loss": 0.8026,
+      "step": 3650
+    },
+    {
+      "epoch": 1.42,
+      "learning_rate": 0.00010806172395300789,
+      "loss": 0.7941,
+      "step": 3675
+    },
+    {
+      "epoch": 1.43,
+      "learning_rate": 0.00010704819981640186,
+      "loss": 0.7989,
+      "step": 3700
+    },
+    {
+      "epoch": 1.44,
+      "learning_rate": 0.00010603394751042522,
+      "loss": 0.7981,
+      "step": 3725
+    },
+    {
+      "epoch": 1.45,
+      "learning_rate": 0.00010501907182033979,
+      "loss": 0.7985,
+      "step": 3750
+    },
+    {
+      "epoch": 1.46,
+      "learning_rate": 0.000104003677595811,
+      "loss": 0.7921,
+      "step": 3775
+    },
+    {
+      "epoch": 1.47,
+      "learning_rate": 0.00010298786974007555,
+      "loss": 0.8012,
+      "step": 3800
+    },
+    {
+      "epoch": 1.48,
+      "learning_rate": 0.00010197175319910343,
+      "loss": 0.7906,
+      "step": 3825
+    },
+    {
+      "epoch": 1.49,
+      "learning_rate": 0.00010095543295075593,
+      "loss": 0.7928,
+      "step": 3850
+    },
+    {
+      "epoch": 1.5,
+      "learning_rate": 9.993901399393979e-05,
+      "loss": 0.8018,
+      "step": 3875
+    },
+    {
+      "epoch": 1.51,
+      "learning_rate": 9.892260133775968e-05,
+      "loss": 0.7991,
+      "step": 3900
+    },
+    {
+      "epoch": 1.52,
+      "learning_rate": 9.79062999906693e-05,
+      "loss": 0.795,
+      "step": 3925
+    },
+    {
+      "epoch": 1.53,
+      "learning_rate": 9.68902149496227e-05,
+      "loss": 0.7977,
+      "step": 3950
+    },
+    {
+      "epoch": 1.54,
+      "learning_rate": 9.587445118922674e-05,
+      "loss": 0.8013,
+      "step": 3975
+    },
+    {
+      "epoch": 1.55,
+      "learning_rate": 9.485911365089589e-05,
+      "loss": 0.7978,
+      "step": 4000
+    },
+    {
+      "epoch": 1.55,
+      "eval_loss": 0.8142631649971008,
+      "eval_runtime": 59.4108,
+      "eval_samples_per_second": 12.287,
+      "eval_steps_per_second": 0.892,
+      "step": 4000
+    },
+    {
+      "epoch": 1.56,
+      "learning_rate": 9.384430723201036e-05,
+      "loss": 0.7912,
+      "step": 4025
+    },
+    {
+      "epoch": 1.57,
+      "learning_rate": 9.283013677507902e-05,
+      "loss": 0.7919,
+      "step": 4050
+    },
+    {
+      "epoch": 1.58,
+      "learning_rate": 9.181670705690761e-05,
+      "loss": 0.7919,
+      "step": 4075
+    },
+    {
+      "epoch": 1.59,
+      "learning_rate": 9.080412277777413e-05,
+      "loss": 0.8018,
+      "step": 4100
+    },
+    {
+      "epoch": 1.6,
+      "learning_rate": 8.979248855061188e-05,
+      "loss": 0.7811,
+      "step": 4125
+    },
+    {
+      "epoch": 1.61,
+      "learning_rate": 8.878190889020159e-05,
+      "loss": 0.7919,
+      "step": 4150
+    },
+    {
+      "epoch": 1.62,
+      "learning_rate": 8.777248820237376e-05,
+      "loss": 0.7994,
+      "step": 4175
+    },
+    {
+      "epoch": 1.63,
+      "learning_rate": 8.676433077322215e-05,
+      "loss": 0.7956,
+      "step": 4200
+    },
+    {
+      "epoch": 1.64,
+      "learning_rate": 8.575754075832973e-05,
+      "loss": 0.7968,
+      "step": 4225
+    },
+    {
+      "epoch": 1.65,
+      "learning_rate": 8.475222217200801e-05,
+      "loss": 0.7905,
+      "step": 4250
+    },
+    {
+      "epoch": 1.66,
+      "learning_rate": 8.374847887655112e-05,
+      "loss": 0.7889,
+      "step": 4275
+    },
+    {
+      "epoch": 1.67,
+      "learning_rate": 8.274641457150543e-05,
+      "loss": 0.7988,
+      "step": 4300
+    },
+    {
+      "epoch": 1.68,
+      "learning_rate": 8.174613278295608e-05,
+      "loss": 0.7947,
+      "step": 4325
+    },
+    {
+      "epoch": 1.69,
+      "learning_rate": 8.074773685283137e-05,
+      "loss": 0.7929,
+      "step": 4350
+    },
+    {
+      "epoch": 1.7,
+      "learning_rate": 7.97513299282264e-05,
+      "loss": 0.7949,
+      "step": 4375
+    },
+    {
+      "epoch": 1.71,
+      "learning_rate": 7.875701495074638e-05,
+      "loss": 0.7925,
+      "step": 4400
+    },
+    {
+      "epoch": 1.72,
+      "learning_rate": 7.776489464587158e-05,
+      "loss": 0.7917,
+      "step": 4425
+    },
+    {
+      "epoch": 1.73,
+      "learning_rate": 7.677507151234448e-05,
+      "loss": 0.7905,
+      "step": 4450
+    },
+    {
+      "epoch": 1.74,
+      "learning_rate": 7.578764781158034e-05,
+      "loss": 0.7912,
+      "step": 4475
+    },
+    {
+      "epoch": 1.74,
+      "learning_rate": 7.480272555710227e-05,
+      "loss": 0.8006,
+      "step": 4500
+    },
+    {
+      "epoch": 1.75,
+      "learning_rate": 7.382040650400185e-05,
+      "loss": 0.7937,
+      "step": 4525
+    },
+    {
+      "epoch": 1.76,
+      "learning_rate": 7.28407921384267e-05,
+      "loss": 0.794,
+      "step": 4550
+    },
+    {
+      "epoch": 1.77,
+      "learning_rate": 7.186398366709545e-05,
+      "loss": 0.7931,
+      "step": 4575
+    },
+    {
+      "epoch": 1.78,
+      "learning_rate": 7.089008200684197e-05,
+      "loss": 0.7982,
+      "step": 4600
+    },
+    {
+      "epoch": 1.79,
+      "learning_rate": 6.991918777418928e-05,
+      "loss": 0.7916,
+      "step": 4625
+    },
+    {
+      "epoch": 1.8,
+      "learning_rate": 6.895140127495455e-05,
+      "loss": 0.7919,
+      "step": 4650
+    },
+    {
+      "epoch": 1.81,
+      "learning_rate": 6.798682249388631e-05,
+      "loss": 0.7863,
+      "step": 4675
+    },
+    {
+      "epoch": 1.82,
+      "learning_rate": 6.702555108433461e-05,
+      "loss": 0.789,
+      "step": 4700
+    },
+    {
+      "epoch": 1.83,
+      "learning_rate": 6.606768635795574e-05,
+      "loss": 0.7902,
+      "step": 4725
+    },
+    {
+      "epoch": 1.84,
+      "learning_rate": 6.511332727445191e-05,
+      "loss": 0.7924,
+      "step": 4750
+    },
+    {
+      "epoch": 1.85,
+      "learning_rate": 6.416257243134747e-05,
+      "loss": 0.7957,
+      "step": 4775
+    },
+    {
+      "epoch": 1.86,
+      "learning_rate": 6.321552005380256e-05,
+      "loss": 0.7916,
+      "step": 4800
+    },
+    {
+      "epoch": 1.87,
+      "learning_rate": 6.22722679844652e-05,
+      "loss": 0.7867,
+      "step": 4825
+    },
+    {
+      "epoch": 1.88,
+      "learning_rate": 6.133291367336284e-05,
+      "loss": 0.7944,
+      "step": 4850
+    },
+    {
+      "epoch": 1.89,
+      "learning_rate": 6.039755416783457e-05,
+      "loss": 0.7982,
+      "step": 4875
+    },
+    {
+      "epoch": 1.9,
+      "learning_rate": 5.946628610250484e-05,
+      "loss": 0.7918,
+      "step": 4900
+    },
+    {
+      "epoch": 1.91,
+      "learning_rate": 5.853920568929996e-05,
+      "loss": 0.7921,
+      "step": 4925
+    },
+    {
+      "epoch": 1.92,
+      "learning_rate": 5.761640870750799e-05,
+      "loss": 0.7878,
+      "step": 4950
+    },
+    {
+      "epoch": 1.93,
+      "learning_rate": 5.669799049388375e-05,
+      "loss": 0.7901,
+      "step": 4975
+    },
+    {
+      "epoch": 1.94,
+      "learning_rate": 5.578404593279911e-05,
+      "loss": 0.7858,
+      "step": 5000
+    },
+    {
+      "epoch": 1.94,
+      "eval_loss": 0.807844877243042,
+      "eval_runtime": 59.586,
+      "eval_samples_per_second": 12.251,
+      "eval_steps_per_second": 0.889,
+      "step": 5000
+    },
+    {
+      "epoch": 1.95,
+      "learning_rate": 5.487466944644033e-05,
+      "loss": 0.7902,
+      "step": 5025
+    },
+    {
+      "epoch": 1.96,
+      "learning_rate": 5.3969954985052996e-05,
+      "loss": 0.7979,
+      "step": 5050
+    },
+    {
+      "epoch": 1.97,
+      "learning_rate": 5.306999601723579e-05,
+      "loss": 0.7931,
+      "step": 5075
+    },
+    {
+      "epoch": 1.98,
+      "learning_rate": 5.21748855202839e-05,
+      "loss": 0.7868,
+      "step": 5100
+    },
+    {
+      "epoch": 1.99,
+      "learning_rate": 5.128471597058342e-05,
+      "loss": 0.7993,
+      "step": 5125
+    },
+    {
+      "epoch": 2.0,
+      "learning_rate": 5.03995793340572e-05,
+      "loss": 0.7892,
+      "step": 5150
+    },
+    {
+      "epoch": 2.01,
+      "learning_rate": 4.9519567056663694e-05,
+      "loss": 0.7788,
+      "step": 5175
+    },
+    {
+      "epoch": 2.02,
+      "learning_rate": 4.864477005494938e-05,
+      "loss": 0.7654,
+      "step": 5200
+    },
+    {
+      "epoch": 2.03,
+      "learning_rate": 4.777527870665592e-05,
+      "loss": 0.7468,
+      "step": 5225
+    },
+    {
+      "epoch": 2.04,
+      "learning_rate": 4.691118284138296e-05,
+      "loss": 0.7359,
+      "step": 5250
+    },
+    {
+      "epoch": 2.05,
+      "learning_rate": 4.605257173130763e-05,
+      "loss": 0.7422,
+      "step": 5275
+    },
+    {
+      "epoch": 2.06,
+      "learning_rate": 4.519953408196152e-05,
+      "loss": 0.7424,
+      "step": 5300
+    },
+    {
+      "epoch": 2.06,
+      "learning_rate": 4.435215802306635e-05,
+      "loss": 0.7521,
+      "step": 5325
+    },
+    {
+      "epoch": 2.07,
+      "learning_rate": 4.351053109942894e-05,
+      "loss": 0.7477,
+      "step": 5350
+    },
+    {
+      "epoch": 2.08,
+      "learning_rate": 4.2674740261896776e-05,
+      "loss": 0.7456,
+      "step": 5375
+    },
+    {
+      "epoch": 2.09,
+      "learning_rate": 4.1844871858374844e-05,
+      "loss": 0.766,
+      "step": 5400
+    },
+    {
+      "epoch": 2.1,
+      "learning_rate": 4.1021011624904814e-05,
+      "loss": 0.7664,
+      "step": 5425
+    },
+    {
+      "epoch": 2.11,
+      "learning_rate": 4.0203244676807353e-05,
+      "loss": 0.7703,
+      "step": 5450
+    },
+    {
+      "epoch": 2.12,
+      "learning_rate": 3.939165549988873e-05,
+      "loss": 0.7674,
+      "step": 5475
+    },
+    {
+      "epoch": 2.13,
+      "learning_rate": 3.858632794171222e-05,
+      "loss": 0.7722,
+      "step": 5500
+    },
+    {
+      "epoch": 2.14,
+      "learning_rate": 3.778734520293562e-05,
+      "loss": 0.7716,
+      "step": 5525
+    },
+    {
+      "epoch": 2.15,
+      "learning_rate": 3.699478982871561e-05,
+      "loss": 0.7795,
+      "step": 5550
+    },
+    {
+      "epoch": 2.16,
+      "learning_rate": 3.62087437001797e-05,
+      "loss": 0.7728,
+      "step": 5575
+    },
+    {
+      "epoch": 2.17,
+      "learning_rate": 3.5429288025966944e-05,
+      "loss": 0.7709,
+      "step": 5600
+    },
+    {
+      "epoch": 2.18,
+      "learning_rate": 3.4656503333837956e-05,
+      "loss": 0.7682,
+      "step": 5625
+    },
+    {
+      "epoch": 2.19,
+      "learning_rate": 3.389046946235542e-05,
+      "loss": 0.7734,
+      "step": 5650
+    },
+    {
+      "epoch": 2.2,
+      "learning_rate": 3.313126555263576e-05,
+      "loss": 0.7716,
+      "step": 5675
+    },
+    {
+      "epoch": 2.21,
+      "learning_rate": 3.237897004017276e-05,
+      "loss": 0.7716,
+      "step": 5700
+    },
+    {
+      "epoch": 2.22,
+      "learning_rate": 3.163366064673427e-05,
+      "loss": 0.7721,
+      "step": 5725
+    },
+    {
+      "epoch": 2.23,
+      "learning_rate": 3.089541437233252e-05,
+      "loss": 0.7658,
+      "step": 5750
+    },
+    {
+      "epoch": 2.24,
+      "learning_rate": 3.0164307487268996e-05,
+      "loss": 0.7716,
+      "step": 5775
+    },
+    {
+      "epoch": 2.25,
+      "learning_rate": 2.944041552425475e-05,
+      "loss": 0.7687,
+      "step": 5800
+    }
+  ],
+  "max_steps": 7737,
+  "num_train_epochs": 3,
+  "total_flos": 2.4980935510562177e+19,
+  "trial_name": null,
+  "trial_params": null
+}
diff --git a/checkpoint-5800/training_args.bin b/checkpoint-5800/training_args.bin
new file mode 100644
index 0000000000000000000000000000000000000000..c456116f688fe6cb6deecb1e0a1cf8d153d349fb
--- /dev/null
+++ b/checkpoint-5800/training_args.bin
@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:df5d13b3f1b9942f80afde79010ef0947feee3df761d245fef1699bc397648b2
+size 4027
diff --git a/checkpoint-5900/README.md b/checkpoint-5900/README.md
new file mode 100644
index 0000000000000000000000000000000000000000..f2208b0ded6c10ed47b2ea9df5ab7c8dd721a53c
--- /dev/null
+++ b/checkpoint-5900/README.md
@@ -0,0 +1,20 @@
+---
+library_name: peft
+---
+## Training procedure
+
+
+The following `bitsandbytes` quantization config was used during training:
+- load_in_8bit: False
+- load_in_4bit: True
+- llm_int8_threshold: 6.0
+- llm_int8_skip_modules: None
+- llm_int8_enable_fp32_cpu_offload: False
+- llm_int8_has_fp16_weight: False
+- bnb_4bit_quant_type: nf4
+- bnb_4bit_use_double_quant: True
+- bnb_4bit_compute_dtype: bfloat16
+### Framework versions
+
+
+- PEFT 0.5.0.dev0
diff --git a/checkpoint-5900/adapter_config.json b/checkpoint-5900/adapter_config.json
new file mode 100644
index 0000000000000000000000000000000000000000..7052646debaf453de93d6176727714122c31b64a
--- /dev/null
+++ b/checkpoint-5900/adapter_config.json
@@ -0,0 +1,26 @@
+{
+  "auto_mapping": null,
+  "base_model_name_or_path": "/workspace/webui/models/TheBloke_Llama-2-13B-fp16",
+  "bias": "none",
+  "fan_in_fan_out": null,
+  "inference_mode": true,
+  "init_lora_weights": true,
+  "layers_pattern": null,
+  "layers_to_transform": null,
+  "lora_alpha": 16,
+  "lora_dropout": 0.05,
+  "modules_to_save": null,
+  "peft_type": "LORA",
+  "r": 32,
+  "revision": null,
+  "target_modules": [
+    "q_proj",
+    "v_proj",
+    "gate_proj",
+    "up_proj",
+    "o_proj",
+    "k_proj",
+    "down_proj"
+  ],
+  "task_type": "CAUSAL_LM"
+}
\ No newline at end of file
diff --git a/checkpoint-5900/adapter_model.bin b/checkpoint-5900/adapter_model.bin
new file mode 100644
index 0000000000000000000000000000000000000000..a5a6e355eff258e831ac8972e31e0b7b97f5a0a4
--- /dev/null
+++ b/checkpoint-5900/adapter_model.bin
@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:d3b3da1c54329907c1913ad730b0e8c5d24d4c2d49e4f95aa7976664afc0a98c
+size 500897101
diff --git a/checkpoint-5900/adapter_model/README.md b/checkpoint-5900/adapter_model/README.md
new file mode 100644
index 0000000000000000000000000000000000000000..f2208b0ded6c10ed47b2ea9df5ab7c8dd721a53c
--- /dev/null
+++ b/checkpoint-5900/adapter_model/README.md
@@ -0,0 +1,20 @@
+---
+library_name: peft
+---
+## Training procedure
+
+
+The following `bitsandbytes` quantization config was used during training:
+- load_in_8bit: False
+- load_in_4bit: True
+- llm_int8_threshold: 6.0
+- llm_int8_skip_modules: None
+- llm_int8_enable_fp32_cpu_offload: False
+- llm_int8_has_fp16_weight: False
+- bnb_4bit_quant_type: nf4
+- bnb_4bit_use_double_quant: True
+- bnb_4bit_compute_dtype: bfloat16
+### Framework versions
+
+
+- PEFT 0.5.0.dev0
diff --git a/checkpoint-5900/adapter_model/adapter_config.json b/checkpoint-5900/adapter_model/adapter_config.json
new file mode 100644
index 0000000000000000000000000000000000000000..7052646debaf453de93d6176727714122c31b64a
--- /dev/null
+++ b/checkpoint-5900/adapter_model/adapter_config.json
@@ -0,0 +1,26 @@
+{
+  "auto_mapping": null,
+  "base_model_name_or_path": "/workspace/webui/models/TheBloke_Llama-2-13B-fp16",
+  "bias": "none",
+  "fan_in_fan_out": null,
+  "inference_mode": true,
+  "init_lora_weights": true,
+  "layers_pattern": null,
+  "layers_to_transform": null,
+  "lora_alpha": 16,
+  "lora_dropout": 0.05,
+  "modules_to_save": null,
+  "peft_type": "LORA",
+  "r": 32,
+  "revision": null,
+  "target_modules": [
+    "q_proj",
+    "v_proj",
+    "gate_proj",
+    "up_proj",
+    "o_proj",
+    "k_proj",
+    "down_proj"
+  ],
+  "task_type": "CAUSAL_LM"
+}
\ No newline at end of file
diff --git a/checkpoint-5900/adapter_model/adapter_model.bin b/checkpoint-5900/adapter_model/adapter_model.bin
new file mode 100644
index 0000000000000000000000000000000000000000..a5a6e355eff258e831ac8972e31e0b7b97f5a0a4
--- /dev/null
+++ b/checkpoint-5900/adapter_model/adapter_model.bin
@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:d3b3da1c54329907c1913ad730b0e8c5d24d4c2d49e4f95aa7976664afc0a98c
+size 500897101
diff --git a/checkpoint-5900/optimizer.pt b/checkpoint-5900/optimizer.pt
new file mode 100644
index 0000000000000000000000000000000000000000..72266aa65efccd48202e3fb60bdb4aeb59f6ba26
--- /dev/null
+++ b/checkpoint-5900/optimizer.pt
@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:1197bafa9ae84ea7a453b741f03a67406157df8c322c0f2e87f2e9c99c7e6415
+size 1001752701
diff --git a/checkpoint-5900/rng_state_0.pth b/checkpoint-5900/rng_state_0.pth
new file mode 100644
index 0000000000000000000000000000000000000000..18663756587b8bd910f595fe23fb24123cbf1bed
--- /dev/null
+++ b/checkpoint-5900/rng_state_0.pth
@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:c6b970d7fefc137c6fa7c1f3ec5751405f46258d274a85e6d4372f5c1b430100
+size 27772
diff --git a/checkpoint-5900/rng_state_1.pth b/checkpoint-5900/rng_state_1.pth
new file mode 100644
index 0000000000000000000000000000000000000000..e6dead9aa0a1048087f6cdc6949859fc9a9bd458
--- /dev/null
+++ b/checkpoint-5900/rng_state_1.pth
@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:b7c346a11c7d7489d10f88696f09c70a4abe9b4d22e70d081daf287152ba6758
+size 27772
diff --git a/checkpoint-5900/rng_state_10.pth b/checkpoint-5900/rng_state_10.pth
new file mode 100644
index 0000000000000000000000000000000000000000..3e1f2dffc2f5d8d9447fbb85330e4bfd2af94428
--- /dev/null
+++ b/checkpoint-5900/rng_state_10.pth
@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:28cc2f86fa3f892a60fcd12ec1c513fe4892758b164d8b6dcaf3ffad379de4bf
+size 27789
diff --git a/checkpoint-5900/rng_state_11.pth b/checkpoint-5900/rng_state_11.pth
new file mode 100644
index 0000000000000000000000000000000000000000..d0a94b7884f61885024b4ce786c9cd49560ff4a8
--- /dev/null
+++ b/checkpoint-5900/rng_state_11.pth
@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:e7e9eaa5d2c6ecd694c6005f784826c1ef125cd869dc4ae397dc6622d332750e
+size 27789
diff --git a/checkpoint-5900/rng_state_12.pth b/checkpoint-5900/rng_state_12.pth
new file mode 100644
index 0000000000000000000000000000000000000000..46ee92749473d23751444f701c8fdd98c7b6b03a
--- /dev/null
+++ b/checkpoint-5900/rng_state_12.pth
@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:38acd1e9cddbb15c52d3524a7f3e3b0813e9d4cd4aa950fe53314f0b60178040
+size 27789
diff --git a/checkpoint-5900/rng_state_13.pth b/checkpoint-5900/rng_state_13.pth
new file mode 100644
index 0000000000000000000000000000000000000000..ec9105e08842cbba5b65f8457b1b03541bb29903
--- /dev/null
+++ b/checkpoint-5900/rng_state_13.pth
@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:f0c474d8b008c60069849057cd5597c90d8f9e8db24500adaab180041a48449f
+size 27789
diff --git a/checkpoint-5900/rng_state_2.pth b/checkpoint-5900/rng_state_2.pth
new file mode 100644
index 0000000000000000000000000000000000000000..6a9b5e7010270c29b101f59decb40b4f8e99955a
--- /dev/null
+++ b/checkpoint-5900/rng_state_2.pth
@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:b70d819dbeb60759d92edd51269d8cee2700d47a9ed7c6a47d97f93849e55bb0
+size 27772
diff --git a/checkpoint-5900/rng_state_3.pth b/checkpoint-5900/rng_state_3.pth
new file mode 100644
index 0000000000000000000000000000000000000000..766c27d9f7bfd190376f77d0549e323ac4dafb34
--- /dev/null
+++ b/checkpoint-5900/rng_state_3.pth
@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:1d28ba4337f3d0ed31c728bf8fbbd7c00c8014b3f57ffbf29863e421ce59080c
+size 27772
diff --git a/checkpoint-5900/rng_state_4.pth b/checkpoint-5900/rng_state_4.pth
new file mode 100644
index 0000000000000000000000000000000000000000..58d9ca8b3ee6e9a5b21de233887d2752e6347a7d
--- /dev/null
+++ b/checkpoint-5900/rng_state_4.pth
@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:00c43f507f234617c2e9bcf5dd72636cec19310803bce07ad6514e0e90814858
+size 27772
diff --git a/checkpoint-5900/rng_state_5.pth b/checkpoint-5900/rng_state_5.pth
new file mode 100644
index 0000000000000000000000000000000000000000..091228cf96b09b25b5d2a6eb2e968700c19f61b8
--- /dev/null
+++ b/checkpoint-5900/rng_state_5.pth
@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:29979a24bd6bef289a10c8e120d56f27b5f058b309d98caf16f123a7a861dce9
+size 27772
diff --git a/checkpoint-5900/rng_state_6.pth b/checkpoint-5900/rng_state_6.pth
new file mode 100644
index 0000000000000000000000000000000000000000..1f7862d3d3ffda1d283d94272f9bd3dc88e7c55c
--- /dev/null
+++ b/checkpoint-5900/rng_state_6.pth
@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:78aa8d7c2497285c4d5739872ca5ceac97b5b636a89d97a0185297b5bbd049d6
+size 27772
diff --git a/checkpoint-5900/rng_state_7.pth b/checkpoint-5900/rng_state_7.pth
new file mode 100644
index 0000000000000000000000000000000000000000..bf5c1d123706c7d2647e69f7e1af9b77b9fb1599
--- /dev/null
+++ b/checkpoint-5900/rng_state_7.pth
@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:04a7b8a1abaea1776e2ddd83a1b3d050ed23897ede63223f377eb777702f7a0f
+size 27772
diff --git a/checkpoint-5900/rng_state_8.pth b/checkpoint-5900/rng_state_8.pth
new file mode 100644
index 0000000000000000000000000000000000000000..f202d39b150ac8803b4f06ea54b6b25b9cb7a07e
--- /dev/null
+++ b/checkpoint-5900/rng_state_8.pth
@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:acca6763654c1dff02cf8af98fff17739233e457ad75b6ff4297b68efc173984
+size 27772
diff --git a/checkpoint-5900/rng_state_9.pth b/checkpoint-5900/rng_state_9.pth
new file mode 100644
index 0000000000000000000000000000000000000000..e5b73e77dc5a4623884fa5dc6cc8865042a65e03
--- /dev/null
+++ b/checkpoint-5900/rng_state_9.pth
@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:a71a0b91f398dff81646dcd42cd8df3dfca54a277c75bab4cc51332fdf4a4768
+size 27772
diff --git a/checkpoint-5900/scheduler.pt b/checkpoint-5900/scheduler.pt
new file mode 100644
index 0000000000000000000000000000000000000000..6398293689882cc1555627d627d6ecde7fca623c
--- /dev/null
+++ b/checkpoint-5900/scheduler.pt
@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:f769241bb104f866dbb417f19005672d382c2f1d72ce0cc8284c37b0a8dd3e0e
+size 627
diff --git a/checkpoint-5900/trainer_state.json b/checkpoint-5900/trainer_state.json
new file mode 100644
index 0000000000000000000000000000000000000000..f25e9351d5d160b59e18ca4b6901b53e3e6a665f
--- /dev/null
+++ b/checkpoint-5900/trainer_state.json
@@ -0,0 +1,1448 @@
+{
+  "best_metric": null,
+  "best_model_checkpoint": null,
+  "epoch": 2.2877084141139976,
+  "global_step": 5900,
+  "is_hyper_param_search": false,
+  "is_local_process_zero": true,
+  "is_world_process_zero": true,
+  "log_history": [
+    {
+      "epoch": 0.02,
+      "learning_rate": 0.0001999867761371633,
+      "loss": 1.0435,
+      "step": 50
+    },
+    {
+      "epoch": 0.04,
+      "learning_rate": 0.00019993306018843102,
+      "loss": 0.8918,
+      "step": 100
+    },
+    {
+      "epoch": 0.06,
+      "learning_rate": 0.00019983804784290833,
+      "loss": 0.8874,
+      "step": 150
+    },
+    {
+      "epoch": 0.08,
+      "learning_rate": 0.00019970177836355307,
+      "loss": 0.8839,
+      "step": 200
+    },
+    {
+      "epoch": 0.09,
+      "learning_rate": 0.00019961818913082012,
+      "loss": 0.8801,
+      "step": 225
+    },
+    {
+      "epoch": 0.1,
+      "learning_rate": 0.00019952430806244534,
+      "loss": 0.8753,
+      "step": 250
+    },
+    {
+      "epoch": 0.11,
+      "learning_rate": 0.00019942014485754635,
+      "loss": 0.8754,
+      "step": 275
+    },
+    {
+      "epoch": 0.12,
+      "learning_rate": 0.00019930571027751713,
+      "loss": 0.8751,
+      "step": 300
+    },
+    {
+      "epoch": 0.13,
+      "learning_rate": 0.0001991810161449164,
+      "loss": 0.8819,
+      "step": 325
+    },
+    {
+      "epoch": 0.14,
+      "learning_rate": 0.00019904607534224612,
+      "loss": 0.8744,
+      "step": 350
+    },
+    {
+      "epoch": 0.15,
+      "learning_rate": 0.00019890090181062063,
+      "loss": 0.8735,
+      "step": 375
+    },
+    {
+      "epoch": 0.16,
+      "learning_rate": 0.00019874551054832625,
+      "loss": 0.8703,
+      "step": 400
+    },
+    {
+      "epoch": 0.16,
+      "learning_rate": 0.00019857991760927193,
+      "loss": 0.8715,
+      "step": 425
+    },
+    {
+      "epoch": 0.17,
+      "learning_rate": 0.00019840414010133045,
+      "loss": 0.8714,
+      "step": 450
+    },
+    {
+      "epoch": 0.18,
+      "learning_rate": 0.00019821819618457114,
+      "loss": 0.8653,
+      "step": 475
+    },
+    {
+      "epoch": 0.19,
+      "learning_rate": 0.0001980221050693837,
+      "loss": 0.8716,
+      "step": 500
+    },
+    {
+      "epoch": 0.2,
+      "learning_rate": 0.00019781588701449338,
+      "loss": 0.8695,
+      "step": 525
+    },
+    {
+      "epoch": 0.21,
+      "learning_rate": 0.0001975995633248682,
+      "loss": 0.8746,
+      "step": 550
+    },
+    {
+      "epoch": 0.22,
+      "learning_rate": 0.00019737315634951762,
+      "loss": 0.8731,
+      "step": 575
+    },
+    {
+      "epoch": 0.23,
+      "learning_rate": 0.00019713668947918386,
+      "loss": 0.867,
+      "step": 600
+    },
+    {
+      "epoch": 0.24,
+      "learning_rate": 0.0001968901871439252,
+      "loss": 0.8706,
+      "step": 625
+    },
+    {
+      "epoch": 0.25,
+      "learning_rate": 0.000196633674810592,
+      "loss": 0.8595,
+      "step": 650
+    },
+    {
+      "epoch": 0.26,
+      "learning_rate": 0.0001963671789801958,
+      "loss": 0.8627,
+      "step": 675
+    },
+    {
+      "epoch": 0.27,
+      "learning_rate": 0.0001960907271851712,
+      "loss": 0.8607,
+      "step": 700
+    },
+    {
+      "epoch": 0.28,
+      "learning_rate": 0.00019580434798653173,
+      "loss": 0.858,
+      "step": 725
+    },
+    {
+      "epoch": 0.29,
+      "learning_rate": 0.00019550807097091876,
+      "loss": 0.8589,
+      "step": 750
+    },
+    {
+      "epoch": 0.3,
+      "learning_rate": 0.00019520192674754515,
+      "loss": 0.8561,
+      "step": 775
+    },
+    {
+      "epoch": 0.31,
+      "learning_rate": 0.00019488594694503264,
+      "loss": 0.8576,
+      "step": 800
+    },
+    {
+      "epoch": 0.32,
+      "learning_rate": 0.00019456016420814446,
+      "loss": 0.8597,
+      "step": 825
+    },
+    {
+      "epoch": 0.33,
+      "learning_rate": 0.00019422461219441254,
+      "loss": 0.862,
+      "step": 850
+    },
+    {
+      "epoch": 0.34,
+      "learning_rate": 0.00019387932557066035,
+      "loss": 0.8577,
+      "step": 875
+    },
+    {
+      "epoch": 0.35,
+      "learning_rate": 0.00019352434000942127,
+      "loss": 0.8632,
+      "step": 900
+    },
+    {
+      "epoch": 0.36,
+      "learning_rate": 0.00019315969218525333,
+      "loss": 0.8567,
+      "step": 925
+    },
+    {
+      "epoch": 0.37,
+      "learning_rate": 0.00019278541977095005,
+      "loss": 0.8501,
+      "step": 950
+    },
+    {
+      "epoch": 0.38,
+      "learning_rate": 0.00019240156143364844,
+      "loss": 0.8596,
+      "step": 975
+    },
+    {
+      "epoch": 0.39,
+      "learning_rate": 0.00019200815683083434,
+      "loss": 0.8556,
+      "step": 1000
+    },
+    {
+      "epoch": 0.39,
+      "eval_loss": 0.8521950244903564,
+      "eval_runtime": 59.8838,
+      "eval_samples_per_second": 12.19,
+      "eval_steps_per_second": 0.885,
+      "step": 1000
+    },
+    {
+      "epoch": 0.4,
+      "learning_rate": 0.00019160524660624505,
+      "loss": 0.8531,
+      "step": 1025
+    },
+    {
+      "epoch": 0.41,
+      "learning_rate": 0.00019119287238567045,
+      "loss": 0.8513,
+      "step": 1050
+    },
+    {
+      "epoch": 0.42,
+      "learning_rate": 0.00019077107677265253,
+      "loss": 0.8502,
+      "step": 1075
+    },
+    {
+      "epoch": 0.43,
+      "learning_rate": 0.00019033990334408384,
+      "loss": 0.8469,
+      "step": 1100
+    },
+    {
+      "epoch": 0.44,
+      "learning_rate": 0.00018989939664570545,
+      "loss": 0.8495,
+      "step": 1125
+    },
+    {
+      "epoch": 0.45,
+      "learning_rate": 0.00018944960218750484,
+      "loss": 0.8485,
+      "step": 1150
+    },
+    {
+      "epoch": 0.46,
+      "learning_rate": 0.00018899056643901404,
+      "loss": 0.8534,
+      "step": 1175
+    },
+    {
+      "epoch": 0.47,
+      "learning_rate": 0.00018852233682450893,
+      "loss": 0.8531,
+      "step": 1200
+    },
+    {
+      "epoch": 0.47,
+      "learning_rate": 0.00018804496171810948,
+      "loss": 0.8509,
+      "step": 1225
+    },
+    {
+      "epoch": 0.48,
+      "learning_rate": 0.00018755849043878222,
+      "loss": 0.8445,
+      "step": 1250
+    },
+    {
+      "epoch": 0.49,
+      "learning_rate": 0.0001870629732452449,
+      "loss": 0.8548,
+      "step": 1275
+    },
+    {
+      "epoch": 0.5,
+      "learning_rate": 0.00018655846133077417,
+      "loss": 0.8441,
+      "step": 1300
+    },
+    {
+      "epoch": 0.51,
+      "learning_rate": 0.00018604500681791656,
+      "loss": 0.8533,
+      "step": 1325
+    },
+    {
+      "epoch": 0.52,
+      "learning_rate": 0.00018552266275310373,
+      "loss": 0.8505,
+      "step": 1350
+    },
+    {
+      "epoch": 0.53,
+      "learning_rate": 0.0001849914831011719,
+      "loss": 0.8544,
+      "step": 1375
+    },
+    {
+      "epoch": 0.54,
+      "learning_rate": 0.00018445152273978668,
+      "loss": 0.845,
+      "step": 1400
+    },
+    {
+      "epoch": 0.55,
+      "learning_rate": 0.00018390283745377354,
+      "loss": 0.8376,
+      "step": 1425
+    },
+    {
+      "epoch": 0.56,
+      "learning_rate": 0.0001833454839293545,
+      "loss": 0.847,
+      "step": 1450
+    },
+    {
+      "epoch": 0.57,
+      "learning_rate": 0.00018277951974829163,
+      "loss": 0.8473,
+      "step": 1475
+    },
+    {
+      "epoch": 0.58,
+      "learning_rate": 0.0001822050033819382,
+      "loss": 0.8438,
+      "step": 1500
+    },
+    {
+      "epoch": 0.59,
+      "learning_rate": 0.00018162199418519785,
+      "loss": 0.8418,
+      "step": 1525
+    },
+    {
+      "epoch": 0.6,
+      "learning_rate": 0.00018103055239039243,
+      "loss": 0.842,
+      "step": 1550
+    },
+    {
+      "epoch": 0.61,
+      "learning_rate": 0.0001804307391010393,
+      "loss": 0.8435,
+      "step": 1575
+    },
+    {
+      "epoch": 0.62,
+      "learning_rate": 0.00017982261628553842,
+      "loss": 0.8349,
+      "step": 1600
+    },
+    {
+      "epoch": 0.63,
+      "learning_rate": 0.0001792062467707703,
+      "loss": 0.8483,
+      "step": 1625
+    },
+    {
+      "epoch": 0.64,
+      "learning_rate": 0.0001785816942356052,
+      "loss": 0.8387,
+      "step": 1650
+    },
+    {
+      "epoch": 0.65,
+      "learning_rate": 0.00017794902320432429,
+      "loss": 0.843,
+      "step": 1675
+    },
+    {
+      "epoch": 0.66,
+      "learning_rate": 0.00017730829903995333,
+      "loss": 0.8424,
+      "step": 1700
+    },
+    {
+      "epoch": 0.67,
+      "learning_rate": 0.00017665958793751006,
+      "loss": 0.8418,
+      "step": 1725
+    },
+    {
+      "epoch": 0.68,
+      "learning_rate": 0.00017600295691716522,
+      "loss": 0.8384,
+      "step": 1750
+    },
+    {
+      "epoch": 0.69,
+      "learning_rate": 0.00017533847381731856,
+      "loss": 0.8445,
+      "step": 1775
+    },
+    {
+      "epoch": 0.7,
+      "learning_rate": 0.00017466620728759033,
+      "loss": 0.8446,
+      "step": 1800
+    },
+    {
+      "epoch": 0.71,
+      "learning_rate": 0.00017398622678172878,
+      "loss": 0.838,
+      "step": 1825
+    },
+    {
+      "epoch": 0.72,
+      "learning_rate": 0.0001732986025504348,
+      "loss": 0.8415,
+      "step": 1850
+    },
+    {
+      "epoch": 0.73,
+      "learning_rate": 0.000172603405634104,
+      "loss": 0.8357,
+      "step": 1875
+    },
+    {
+      "epoch": 0.74,
+      "learning_rate": 0.00017190070785548755,
+      "loss": 0.8311,
+      "step": 1900
+    },
+    {
+      "epoch": 0.75,
+      "learning_rate": 0.0001711905818122717,
+      "loss": 0.8333,
+      "step": 1925
+    },
+    {
+      "epoch": 0.76,
+      "learning_rate": 0.0001704731008695777,
+      "loss": 0.8387,
+      "step": 1950
+    },
+    {
+      "epoch": 0.77,
+      "learning_rate": 0.0001697483391523821,
+      "loss": 0.8442,
+      "step": 1975
+    },
+    {
+      "epoch": 0.78,
+      "learning_rate": 0.00016901637153785885,
+      "loss": 0.8399,
+      "step": 2000
+    },
+    {
+      "epoch": 0.78,
+      "eval_loss": 0.8339959383010864,
+      "eval_runtime": 58.5829,
+      "eval_samples_per_second": 12.461,
+      "eval_steps_per_second": 0.905,
+      "step": 2000
+    },
+    {
+      "epoch": 0.79,
+      "learning_rate": 0.0001682772736476434,
+      "loss": 0.8334,
+      "step": 2025
+    },
+    {
+      "epoch": 0.79,
+      "learning_rate": 0.0001675311218400201,
+      "loss": 0.835,
+      "step": 2050
+    },
+    {
+      "epoch": 0.8,
+      "learning_rate": 0.00016677799320203332,
+      "loss": 0.8368,
+      "step": 2075
+    },
+    {
+      "epoch": 0.81,
+      "learning_rate": 0.00016601796554152344,
+      "loss": 0.8278,
+      "step": 2100
+    },
+    {
+      "epoch": 0.82,
+      "learning_rate": 0.00016525111737908827,
+      "loss": 0.8334,
+      "step": 2125
+    },
+    {
+      "epoch": 0.83,
+      "learning_rate": 0.00016447752793997096,
+      "loss": 0.8416,
+      "step": 2150
+    },
+    {
+      "epoch": 0.84,
+      "learning_rate": 0.00016369727714587483,
+      "loss": 0.8297,
+      "step": 2175
+    },
+    {
+      "epoch": 0.85,
+      "learning_rate": 0.0001629104456067066,
+      "loss": 0.8327,
+      "step": 2200
+    },
+    {
+      "epoch": 0.86,
+      "learning_rate": 0.00016211711461224825,
+      "loss": 0.8324,
+      "step": 2225
+    },
+    {
+      "epoch": 0.87,
+      "learning_rate": 0.0001613173661237589,
+      "loss": 0.8313,
+      "step": 2250
+    },
+    {
+      "epoch": 0.88,
+      "learning_rate": 0.0001605112827655069,
+      "loss": 0.8292,
+      "step": 2275
+    },
+    {
+      "epoch": 0.89,
+      "learning_rate": 0.0001596989478162339,
+      "loss": 0.8334,
+      "step": 2300
+    },
+    {
+      "epoch": 0.9,
+      "learning_rate": 0.00015888044520055106,
+      "loss": 0.8352,
+      "step": 2325
+    },
+    {
+      "epoch": 0.91,
+      "learning_rate": 0.00015805585948026852,
+      "loss": 0.823,
+      "step": 2350
+    },
+    {
+      "epoch": 0.92,
+      "learning_rate": 0.000157225275845659,
+      "loss": 0.8293,
+      "step": 2375
+    },
+    {
+      "epoch": 0.93,
+      "learning_rate": 0.00015638878010665672,
+      "loss": 0.8289,
+      "step": 2400
+    },
+    {
+      "epoch": 0.94,
+      "learning_rate": 0.00015554645868399205,
+      "loss": 0.832,
+      "step": 2425
+    },
+    {
+      "epoch": 0.95,
+      "learning_rate": 0.00015469839860026308,
+      "loss": 0.8294,
+      "step": 2450
+    },
+    {
+      "epoch": 0.96,
+      "learning_rate": 0.0001538446874709452,
+      "loss": 0.8281,
+      "step": 2475
+    },
+    {
+      "epoch": 0.97,
+      "learning_rate": 0.00015298541349533925,
+      "loss": 0.8314,
+      "step": 2500
+    },
+    {
+      "epoch": 0.98,
+      "learning_rate": 0.00015212066544745926,
+      "loss": 0.831,
+      "step": 2525
+    },
+    {
+      "epoch": 0.99,
+      "learning_rate": 0.00015125053266686124,
+      "loss": 0.8319,
+      "step": 2550
+    },
+    {
+      "epoch": 1.0,
+      "learning_rate": 0.00015037510504941303,
+      "loss": 0.8259,
+      "step": 2575
+    },
+    {
+      "epoch": 1.01,
+      "learning_rate": 0.00014949447303800695,
+      "loss": 0.8133,
+      "step": 2600
+    },
+    {
+      "epoch": 1.02,
+      "learning_rate": 0.00014860872761321593,
+      "loss": 0.8139,
+      "step": 2625
+    },
+    {
+      "epoch": 1.03,
+      "learning_rate": 0.00014771796028389405,
+      "loss": 0.804,
+      "step": 2650
+    },
+    {
+      "epoch": 1.04,
+      "learning_rate": 0.0001468222630777225,
+      "loss": 0.8011,
+      "step": 2675
+    },
+    {
+      "epoch": 1.05,
+      "learning_rate": 0.00014592172853170193,
+      "loss": 0.8037,
+      "step": 2700
+    },
+    {
+      "epoch": 1.06,
+      "learning_rate": 0.00014501644968259212,
+      "loss": 0.8063,
+      "step": 2725
+    },
+    {
+      "epoch": 1.07,
+      "learning_rate": 0.00014410652005730025,
+      "loss": 0.8155,
+      "step": 2750
+    },
+    {
+      "epoch": 1.08,
+      "learning_rate": 0.00014319203366321826,
+      "loss": 0.8066,
+      "step": 2775
+    },
+    {
+      "epoch": 1.09,
+      "learning_rate": 0.0001422730849785107,
+      "loss": 0.8091,
+      "step": 2800
+    },
+    {
+      "epoch": 1.1,
+      "learning_rate": 0.0001413497689423539,
+      "loss": 0.8067,
+      "step": 2825
+    },
+    {
+      "epoch": 1.11,
+      "learning_rate": 0.00014042218094512755,
+      "loss": 0.8046,
+      "step": 2850
+    },
+    {
+      "epoch": 1.11,
+      "learning_rate": 0.00013949041681855985,
+      "loss": 0.8053,
+      "step": 2875
+    },
+    {
+      "epoch": 1.12,
+      "learning_rate": 0.0001385545728258264,
+      "loss": 0.8075,
+      "step": 2900
+    },
+    {
+      "epoch": 1.13,
+      "learning_rate": 0.0001376147456516055,
+      "loss": 0.8015,
+      "step": 2925
+    },
+    {
+      "epoch": 1.14,
+      "learning_rate": 0.00013667103239208903,
+      "loss": 0.8016,
+      "step": 2950
+    },
+    {
+      "epoch": 1.15,
+      "learning_rate": 0.00013572353054495126,
+      "loss": 0.8029,
+      "step": 2975
+    },
+    {
+      "epoch": 1.16,
+      "learning_rate": 0.0001347723379992762,
+      "loss": 0.8017,
+      "step": 3000
+    },
+    {
+      "epoch": 1.16,
+      "eval_loss": 0.8229297995567322,
+      "eval_runtime": 59.3398,
+      "eval_samples_per_second": 12.302,
+      "eval_steps_per_second": 0.893,
+      "step": 3000
+    },
+    {
+      "epoch": 1.17,
+      "learning_rate": 0.0001338175530254443,
+      "loss": 0.8049,
+      "step": 3025
+    },
+    {
+      "epoch": 1.18,
+      "learning_rate": 0.00013285927426497985,
+      "loss": 0.8027,
+      "step": 3050
+    },
+    {
+      "epoch": 1.19,
+      "learning_rate": 0.00013189760072036008,
+      "loss": 0.8028,
+      "step": 3075
+    },
+    {
+      "epoch": 1.2,
+      "learning_rate": 0.0001309326317447869,
+      "loss": 0.8021,
+      "step": 3100
+    },
+    {
+      "epoch": 1.21,
+      "learning_rate": 0.00012996446703192257,
+      "loss": 0.8033,
+      "step": 3125
+    },
+    {
+      "epoch": 1.22,
+      "learning_rate": 0.00012899320660558986,
+      "loss": 0.8016,
+      "step": 3150
+    },
+    {
+      "epoch": 1.23,
+      "learning_rate": 0.00012801895080943846,
+      "loss": 0.7995,
+      "step": 3175
+    },
+    {
+      "epoch": 1.24,
+      "learning_rate": 0.0001270418002965782,
+      "loss": 0.799,
+      "step": 3200
+    },
+    {
+      "epoch": 1.25,
+      "learning_rate": 0.0001260618560191802,
+      "loss": 0.8002,
+      "step": 3225
+    },
+    {
+      "epoch": 1.26,
+      "learning_rate": 0.00012507921921804717,
+      "loss": 0.8068,
+      "step": 3250
+    },
+    {
+      "epoch": 1.27,
+      "learning_rate": 0.00012409399141215423,
+      "loss": 0.8041,
+      "step": 3275
+    },
+    {
+      "epoch": 1.28,
+      "learning_rate": 0.0001231062743881603,
+      "loss": 0.7999,
+      "step": 3300
+    },
+    {
+      "epoch": 1.29,
+      "learning_rate": 0.0001221161701898926,
+      "loss": 0.7995,
+      "step": 3325
+    },
+    {
+      "epoch": 1.3,
+      "learning_rate": 0.00012112378110780391,
+      "loss": 0.7959,
+      "step": 3350
+    },
+    {
+      "epoch": 1.31,
+      "learning_rate": 0.00012012920966840486,
+      "loss": 0.7999,
+      "step": 3375
+    },
+    {
+      "epoch": 1.32,
+      "learning_rate": 0.00011913255862367151,
+      "loss": 0.8016,
+      "step": 3400
+    },
+    {
+      "epoch": 1.33,
+      "learning_rate": 0.00011813393094042993,
+      "loss": 0.7944,
+      "step": 3425
+    },
+    {
+      "epoch": 1.34,
+      "learning_rate": 0.0001171334297897181,
+      "loss": 0.8026,
+      "step": 3450
+    },
+    {
+      "epoch": 1.35,
+      "learning_rate": 0.00011613115853612734,
+      "loss": 0.8004,
+      "step": 3475
+    },
+    {
+      "epoch": 1.36,
+      "learning_rate": 0.00011512722072712321,
+      "loss": 0.7992,
+      "step": 3500
+    },
+    {
+      "epoch": 1.37,
+      "learning_rate": 0.00011412172008234785,
+      "loss": 0.8004,
+      "step": 3525
+    },
+    {
+      "epoch": 1.38,
+      "learning_rate": 0.0001131147604829043,
+      "loss": 0.8009,
+      "step": 3550
+    },
+    {
+      "epoch": 1.39,
+      "learning_rate": 0.00011210644596062439,
+      "loss": 0.7993,
+      "step": 3575
+    },
+    {
+      "epoch": 1.4,
+      "learning_rate": 0.00011109688068732081,
+      "loss": 0.7965,
+      "step": 3600
+    },
+    {
+      "epoch": 1.41,
+      "learning_rate": 0.00011008616896402482,
+      "loss": 0.7991,
+      "step": 3625
+    },
+    {
+      "epoch": 1.42,
+      "learning_rate": 0.00010907441521021072,
+      "loss": 0.8026,
+      "step": 3650
+    },
+    {
+      "epoch": 1.42,
+      "learning_rate": 0.00010806172395300789,
+      "loss": 0.7941,
+      "step": 3675
+    },
+    {
+      "epoch": 1.43,
+      "learning_rate": 0.00010704819981640186,
+      "loss": 0.7989,
+      "step": 3700
+    },
+    {
+      "epoch": 1.44,
+      "learning_rate": 0.00010603394751042522,
+      "loss": 0.7981,
+      "step": 3725
+    },
+    {
+      "epoch": 1.45,
+      "learning_rate": 0.00010501907182033979,
+      "loss": 0.7985,
+      "step": 3750
+    },
+    {
+      "epoch": 1.46,
+      "learning_rate": 0.000104003677595811,
+      "loss": 0.7921,
+      "step": 3775
+    },
+    {
+      "epoch": 1.47,
+      "learning_rate": 0.00010298786974007555,
+      "loss": 0.8012,
+      "step": 3800
+    },
+    {
+      "epoch": 1.48,
+      "learning_rate": 0.00010197175319910343,
+      "loss": 0.7906,
+      "step": 3825
+    },
+    {
+      "epoch": 1.49,
+      "learning_rate": 0.00010095543295075593,
+      "loss": 0.7928,
+      "step": 3850
+    },
+    {
+      "epoch": 1.5,
+      "learning_rate": 9.993901399393979e-05,
+      "loss": 0.8018,
+      "step": 3875
+    },
+    {
+      "epoch": 1.51,
+      "learning_rate": 9.892260133775968e-05,
+      "loss": 0.7991,
+      "step": 3900
+    },
+    {
+      "epoch": 1.52,
+      "learning_rate": 9.79062999906693e-05,
+      "loss": 0.795,
+      "step": 3925
+    },
+    {
+      "epoch": 1.53,
+      "learning_rate": 9.68902149496227e-05,
+      "loss": 0.7977,
+      "step": 3950
+    },
+    {
+      "epoch": 1.54,
+      "learning_rate": 9.587445118922674e-05,
+      "loss": 0.8013,
+      "step": 3975
+    },
+    {
+      "epoch": 1.55,
+      "learning_rate": 9.485911365089589e-05,
+      "loss": 0.7978,
+      "step": 4000
+    },
+    {
+      "epoch": 1.55,
+      "eval_loss": 0.8142631649971008,
+      "eval_runtime": 59.4108,
+      "eval_samples_per_second": 12.287,
+      "eval_steps_per_second": 0.892,
+      "step": 4000
+    },
+    {
+      "epoch": 1.56,
+      "learning_rate": 9.384430723201036e-05,
+      "loss": 0.7912,
+      "step": 4025
+    },
+    {
+      "epoch": 1.57,
+      "learning_rate": 9.283013677507902e-05,
+      "loss": 0.7919,
+      "step": 4050
+    },
+    {
+      "epoch": 1.58,
+      "learning_rate": 9.181670705690761e-05,
+      "loss": 0.7919,
+      "step": 4075
+    },
+    {
+      "epoch": 1.59,
+      "learning_rate": 9.080412277777413e-05,
+      "loss": 0.8018,
+      "step": 4100
+    },
+    {
+      "epoch": 1.6,
+      "learning_rate": 8.979248855061188e-05,
+      "loss": 0.7811,
+      "step": 4125
+    },
+    {
+      "epoch": 1.61,
+      "learning_rate": 8.878190889020159e-05,
+      "loss": 0.7919,
+      "step": 4150
+    },
+    {
+      "epoch": 1.62,
+      "learning_rate": 8.777248820237376e-05,
+      "loss": 0.7994,
+      "step": 4175
+    },
+    {
+      "epoch": 1.63,
+      "learning_rate": 8.676433077322215e-05,
+      "loss": 0.7956,
+      "step": 4200
+    },
+    {
+      "epoch": 1.64,
+      "learning_rate": 8.575754075832973e-05,
+      "loss": 0.7968,
+      "step": 4225
+    },
+    {
+      "epoch": 1.65,
+      "learning_rate": 8.475222217200801e-05,
+      "loss": 0.7905,
+      "step": 4250
+    },
+    {
+      "epoch": 1.66,
+      "learning_rate": 8.374847887655112e-05,
+      "loss": 0.7889,
+      "step": 4275
+    },
+    {
+      "epoch": 1.67,
+      "learning_rate": 8.274641457150543e-05,
+      "loss": 0.7988,
+      "step": 4300
+    },
+    {
+      "epoch": 1.68,
+      "learning_rate": 8.174613278295608e-05,
+      "loss": 0.7947,
+      "step": 4325
+    },
+    {
+      "epoch": 1.69,
+      "learning_rate": 8.074773685283137e-05,
+      "loss": 0.7929,
+      "step": 4350
+    },
+    {
+      "epoch": 1.7,
+      "learning_rate": 7.97513299282264e-05,
+      "loss": 0.7949,
+      "step": 4375
+    },
+    {
+      "epoch": 1.71,
+      "learning_rate": 7.875701495074638e-05,
+      "loss": 0.7925,
+      "step": 4400
+    },
+    {
+      "epoch": 1.72,
+      "learning_rate": 7.776489464587158e-05,
+      "loss": 0.7917,
+      "step": 4425
+    },
+    {
+      "epoch": 1.73,
+      "learning_rate": 7.677507151234448e-05,
+      "loss": 0.7905,
+      "step": 4450
+    },
+    {
+      "epoch": 1.74,
+      "learning_rate": 7.578764781158034e-05,
+      "loss": 0.7912,
+      "step": 4475
+    },
+    {
+      "epoch": 1.74,
+      "learning_rate": 7.480272555710227e-05,
+      "loss": 0.8006,
+      "step": 4500
+    },
+    {
+      "epoch": 1.75,
+      "learning_rate": 7.382040650400185e-05,
+      "loss": 0.7937,
+      "step": 4525
+    },
+    {
+      "epoch": 1.76,
+      "learning_rate": 7.28407921384267e-05,
+      "loss": 0.794,
+      "step": 4550
+    },
+    {
+      "epoch": 1.77,
+      "learning_rate": 7.186398366709545e-05,
+      "loss": 0.7931,
+      "step": 4575
+    },
+    {
+      "epoch": 1.78,
+      "learning_rate": 7.089008200684197e-05,
+      "loss": 0.7982,
+      "step": 4600
+    },
+    {
+      "epoch": 1.79,
+      "learning_rate": 6.991918777418928e-05,
+      "loss": 0.7916,
+      "step": 4625
+    },
+    {
+      "epoch": 1.8,
+      "learning_rate": 6.895140127495455e-05,
+      "loss": 0.7919,
+      "step": 4650
+    },
+    {
+      "epoch": 1.81,
+      "learning_rate": 6.798682249388631e-05,
+      "loss": 0.7863,
+      "step": 4675
+    },
+    {
+      "epoch": 1.82,
+      "learning_rate": 6.702555108433461e-05,
+      "loss": 0.789,
+      "step": 4700
+    },
+    {
+      "epoch": 1.83,
+      "learning_rate": 6.606768635795574e-05,
+      "loss": 0.7902,
+      "step": 4725
+    },
+    {
+      "epoch": 1.84,
+      "learning_rate": 6.511332727445191e-05,
+      "loss": 0.7924,
+      "step": 4750
+    },
+    {
+      "epoch": 1.85,
+      "learning_rate": 6.416257243134747e-05,
+      "loss": 0.7957,
+      "step": 4775
+    },
+    {
+      "epoch": 1.86,
+      "learning_rate": 6.321552005380256e-05,
+      "loss": 0.7916,
+      "step": 4800
+    },
+    {
+      "epoch": 1.87,
+      "learning_rate": 6.22722679844652e-05,
+      "loss": 0.7867,
+      "step": 4825
+    },
+    {
+      "epoch": 1.88,
+      "learning_rate": 6.133291367336284e-05,
+      "loss": 0.7944,
+      "step": 4850
+    },
+    {
+      "epoch": 1.89,
+      "learning_rate": 6.039755416783457e-05,
+      "loss": 0.7982,
+      "step": 4875
+    },
+    {
+      "epoch": 1.9,
+      "learning_rate": 5.946628610250484e-05,
+      "loss": 0.7918,
+      "step": 4900
+    },
+    {
+      "epoch": 1.91,
+      "learning_rate": 5.853920568929996e-05,
+      "loss": 0.7921,
+      "step": 4925
+    },
+    {
+      "epoch": 1.92,
+      "learning_rate": 5.761640870750799e-05,
+      "loss": 0.7878,
+      "step": 4950
+    },
+    {
+      "epoch": 1.93,
+      "learning_rate": 5.669799049388375e-05,
+      "loss": 0.7901,
+      "step": 4975
+    },
+    {
+      "epoch": 1.94,
+      "learning_rate": 5.578404593279911e-05,
+      "loss": 0.7858,
+      "step": 5000
+    },
+    {
+      "epoch": 1.94,
+      "eval_loss": 0.807844877243042,
+      "eval_runtime": 59.586,
+      "eval_samples_per_second": 12.251,
+      "eval_steps_per_second": 0.889,
+      "step": 5000
+    },
+    {
+      "epoch": 1.95,
+      "learning_rate": 5.487466944644033e-05,
+      "loss": 0.7902,
+      "step": 5025
+    },
+    {
+      "epoch": 1.96,
+      "learning_rate": 5.3969954985052996e-05,
+      "loss": 0.7979,
+      "step": 5050
+    },
+    {
+      "epoch": 1.97,
+      "learning_rate": 5.306999601723579e-05,
+      "loss": 0.7931,
+      "step": 5075
+    },
+    {
+      "epoch": 1.98,
+      "learning_rate": 5.21748855202839e-05,
+      "loss": 0.7868,
+      "step": 5100
+    },
+    {
+      "epoch": 1.99,
+      "learning_rate": 5.128471597058342e-05,
+      "loss": 0.7993,
+      "step": 5125
+    },
+    {
+      "epoch": 2.0,
+      "learning_rate": 5.03995793340572e-05,
+      "loss": 0.7892,
+      "step": 5150
+    },
+    {
+      "epoch": 2.01,
+      "learning_rate": 4.9519567056663694e-05,
+      "loss": 0.7788,
+      "step": 5175
+    },
+    {
+      "epoch": 2.02,
+      "learning_rate": 4.864477005494938e-05,
+      "loss": 0.7654,
+      "step": 5200
+    },
+    {
+      "epoch": 2.03,
+      "learning_rate": 4.777527870665592e-05,
+      "loss": 0.7468,
+      "step": 5225
+    },
+    {
+      "epoch": 2.04,
+      "learning_rate": 4.691118284138296e-05,
+      "loss": 0.7359,
+      "step": 5250
+    },
+    {
+      "epoch": 2.05,
+      "learning_rate": 4.605257173130763e-05,
+      "loss": 0.7422,
+      "step": 5275
+    },
+    {
+      "epoch": 2.06,
+      "learning_rate": 4.519953408196152e-05,
+      "loss": 0.7424,
+      "step": 5300
+    },
+    {
+      "epoch": 2.06,
+      "learning_rate": 4.435215802306635e-05,
+      "loss": 0.7521,
+      "step": 5325
+    },
+    {
+      "epoch": 2.07,
+      "learning_rate": 4.351053109942894e-05,
+      "loss": 0.7477,
+      "step": 5350
+    },
+    {
+      "epoch": 2.08,
+      "learning_rate": 4.2674740261896776e-05,
+      "loss": 0.7456,
+      "step": 5375
+    },
+    {
+      "epoch": 2.09,
+      "learning_rate": 4.1844871858374844e-05,
+      "loss": 0.766,
+      "step": 5400
+    },
+    {
+      "epoch": 2.1,
+      "learning_rate": 4.1021011624904814e-05,
+      "loss": 0.7664,
+      "step": 5425
+    },
+    {
+      "epoch": 2.11,
+      "learning_rate": 4.0203244676807353e-05,
+      "loss": 0.7703,
+      "step": 5450
+    },
+    {
+      "epoch": 2.12,
+      "learning_rate": 3.939165549988873e-05,
+      "loss": 0.7674,
+      "step": 5475
+    },
+    {
+      "epoch": 2.13,
+      "learning_rate": 3.858632794171222e-05,
+      "loss": 0.7722,
+      "step": 5500
+    },
+    {
+      "epoch": 2.14,
+      "learning_rate": 3.778734520293562e-05,
+      "loss": 0.7716,
+      "step": 5525
+    },
+    {
+      "epoch": 2.15,
+      "learning_rate": 3.699478982871561e-05,
+      "loss": 0.7795,
+      "step": 5550
+    },
+    {
+      "epoch": 2.16,
+      "learning_rate": 3.62087437001797e-05,
+      "loss": 0.7728,
+      "step": 5575
+    },
+    {
+      "epoch": 2.17,
+      "learning_rate": 3.5429288025966944e-05,
+      "loss": 0.7709,
+      "step": 5600
+    },
+    {
+      "epoch": 2.18,
+      "learning_rate": 3.4656503333837956e-05,
+      "loss": 0.7682,
+      "step": 5625
+    },
+    {
+      "epoch": 2.19,
+      "learning_rate": 3.389046946235542e-05,
+      "loss": 0.7734,
+      "step": 5650
+    },
+    {
+      "epoch": 2.2,
+      "learning_rate": 3.313126555263576e-05,
+      "loss": 0.7716,
+      "step": 5675
+    },
+    {
+      "epoch": 2.21,
+      "learning_rate": 3.237897004017276e-05,
+      "loss": 0.7716,
+      "step": 5700
+    },
+    {
+      "epoch": 2.22,
+      "learning_rate": 3.163366064673427e-05,
+      "loss": 0.7721,
+      "step": 5725
+    },
+    {
+      "epoch": 2.23,
+      "learning_rate": 3.089541437233252e-05,
+      "loss": 0.7658,
+      "step": 5750
+    },
+    {
+      "epoch": 2.24,
+      "learning_rate": 3.0164307487268996e-05,
+      "loss": 0.7716,
+      "step": 5775
+    },
+    {
+      "epoch": 2.25,
+      "learning_rate": 2.944041552425475e-05,
+      "loss": 0.7687,
+      "step": 5800
+    },
+    {
+      "epoch": 2.26,
+      "learning_rate": 2.8723813270606982e-05,
+      "loss": 0.7698,
+      "step": 5825
+    },
+    {
+      "epoch": 2.27,
+      "learning_rate": 2.8014574760522416e-05,
+      "loss": 0.7641,
+      "step": 5850
+    },
+    {
+      "epoch": 2.28,
+      "learning_rate": 2.731277326742876e-05,
+      "loss": 0.7746,
+      "step": 5875
+    },
+    {
+      "epoch": 2.29,
+      "learning_rate": 2.6618481296414522e-05,
+      "loss": 0.7722,
+      "step": 5900
+    }
+  ],
+  "max_steps": 7737,
+  "num_train_epochs": 3,
+  "total_flos": 2.5410887248997515e+19,
+  "trial_name": null,
+  "trial_params": null
+}
diff --git a/checkpoint-5900/training_args.bin b/checkpoint-5900/training_args.bin
new file mode 100644
index 0000000000000000000000000000000000000000..c456116f688fe6cb6deecb1e0a1cf8d153d349fb
--- /dev/null
+++ b/checkpoint-5900/training_args.bin
@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:df5d13b3f1b9942f80afde79010ef0947feee3df761d245fef1699bc397648b2
+size 4027