Upload folder using huggingface_hub (#9)

- Upload folder using huggingface_hub (e68ee205210ce2889c0c3007229e19a4d715d7c1)

README.md

@@ -4,18 +4,6 @@ library_name: peft
 ## Training procedure
 
 
-The following `bitsandbytes` quantization config was used during training:
-- quant_method: bitsandbytes
-- load_in_8bit: True
-- load_in_4bit: False
-- 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: fp4
-- bnb_4bit_use_double_quant: False
-- bnb_4bit_compute_dtype: float32
-
 The following `bitsandbytes` quantization config was used during training:
 - quant_method: bitsandbytes
 - load_in_8bit: True
@@ -29,6 +17,5 @@ The following `bitsandbytes` quantization config was used during training:
 - bnb_4bit_compute_dtype: float32
 ### Framework versions
 
-- PEFT 0.6.0.dev0
 
 - PEFT 0.6.0.dev0

adapter_model.bin

@@ -1,3 +1,3 @@
 version https://git-lfs.github.com/spec/v1
-oid sha256:d17f997023cf8f985a547661db8c0ff488c11f01eaeaca062f1c51fd98fd299d
+oid sha256:a3b13e1a616710fea5e375eaba4c717e346333ccc3d748617ed5bb54d4724e79
 size 39409357

checkpoint-1000/README.md

@@ -0,0 +1,34 @@
+---
+library_name: peft
+---
+## Training procedure
+
+
+The following `bitsandbytes` quantization config was used during training:
+- quant_method: bitsandbytes
+- load_in_8bit: True
+- load_in_4bit: False
+- 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: fp4
+- bnb_4bit_use_double_quant: False
+- bnb_4bit_compute_dtype: float32
+
+The following `bitsandbytes` quantization config was used during training:
+- quant_method: bitsandbytes
+- load_in_8bit: True
+- load_in_4bit: False
+- 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: fp4
+- bnb_4bit_use_double_quant: False
+- bnb_4bit_compute_dtype: float32
+### Framework versions
+
+- PEFT 0.6.0.dev0
+
+- PEFT 0.6.0.dev0

checkpoint-1000/adapter_config.json

@@ -0,0 +1,23 @@
+{
+  "auto_mapping": null,
+  "base_model_name_or_path": "bigscience/bloomz-3b",
+  "bias": "none",
+  "fan_in_fan_out": false,
+  "inference_mode": true,
+  "init_lora_weights": true,
+  "layers_pattern": null,
+  "layers_to_transform": null,
+  "lora_alpha": 16,
+  "lora_dropout": 0.0,
+  "modules_to_save": null,
+  "peft_type": "LORA",
+  "r": 8,
+  "revision": null,
+  "target_modules": [
+    "dense_4h_to_h",
+    "dense",
+    "dense_h_to_4h",
+    "query_key_value"
+  ],
+  "task_type": "CAUSAL_LM"
+}

checkpoint-1000/adapter_model.bin

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:442784f9988ccf98a4c80729f764248b7aece00d5cbe4899fc23821ab36912c5
+size 39409357

checkpoint-1000/optimizer.pt

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:a86df865574f7737280e79335d7211d01b31b136c41c37c56c6ff9c2dbed8bdc
+size 78844421

checkpoint-1000/rng_state.pth

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:b0dfb171332ffdf5185078b01e6c5599c2aa6657e6b96cd2596b1da1f86c2d1f
+size 14575

checkpoint-1000/scheduler.pt

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:f90c270e464cbba9ce3ac99c0baf5b79e003ab516981bec7ad9324d178b9d534
+size 627

checkpoint-1000/special_tokens_map.json

@@ -0,0 +1,6 @@
+{
+  "bos_token": "<s>",
+  "eos_token": "</s>",
+  "pad_token": "<pad>",
+  "unk_token": "<unk>"
+}

checkpoint-1000/tokenizer.json

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:17a208233d2ee8d8c83b23bc214df737c44806a1919f444e89b31e586cd956ba
+size 14500471

checkpoint-1000/tokenizer_config.json

@@ -0,0 +1,10 @@
+{
+  "add_prefix_space": false,
+  "bos_token": "<s>",
+  "clean_up_tokenization_spaces": false,
+  "eos_token": "</s>",
+  "model_max_length": 1000000000000000019884624838656,
+  "pad_token": "<pad>",
+  "tokenizer_class": "BloomTokenizer",
+  "unk_token": "<unk>"
+}

checkpoint-1000/training_args.bin

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:7015d16f0f7058d5e16036c5259111dacee395cddfb5fc5ae08d917da3d67f65
+size 4155

checkpoint-1100/README.md

@@ -0,0 +1,34 @@
+---
+library_name: peft
+---
+## Training procedure
+
+
+The following `bitsandbytes` quantization config was used during training:
+- quant_method: bitsandbytes
+- load_in_8bit: True
+- load_in_4bit: False
+- 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: fp4
+- bnb_4bit_use_double_quant: False
+- bnb_4bit_compute_dtype: float32
+
+The following `bitsandbytes` quantization config was used during training:
+- quant_method: bitsandbytes
+- load_in_8bit: True
+- load_in_4bit: False
+- 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: fp4
+- bnb_4bit_use_double_quant: False
+- bnb_4bit_compute_dtype: float32
+### Framework versions
+
+- PEFT 0.6.0.dev0
+
+- PEFT 0.6.0.dev0

checkpoint-1100/adapter_config.json

@@ -0,0 +1,23 @@
+{
+  "auto_mapping": null,
+  "base_model_name_or_path": "bigscience/bloomz-3b",
+  "bias": "none",
+  "fan_in_fan_out": false,
+  "inference_mode": true,
+  "init_lora_weights": true,
+  "layers_pattern": null,
+  "layers_to_transform": null,
+  "lora_alpha": 16,
+  "lora_dropout": 0.0,
+  "modules_to_save": null,
+  "peft_type": "LORA",
+  "r": 8,
+  "revision": null,
+  "target_modules": [
+    "dense_4h_to_h",
+    "dense",
+    "dense_h_to_4h",
+    "query_key_value"
+  ],
+  "task_type": "CAUSAL_LM"
+}

checkpoint-1100/adapter_model.bin

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:6c88ad543ad0cd581cce028fd3dc8d689a3d5d1f1655d648f44d44b3f227ec50
+size 39409357

checkpoint-1100/optimizer.pt

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:5b80027248fc1a38c437e756d546038dbc041930932c6a4ce5ee2f0941b779e7
+size 78844421

checkpoint-1100/rng_state.pth

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:82632deb3eef79ffa0456776bb9d6e5a9a57b9d7022aadd020f4d487704bacab
+size 14575

checkpoint-1100/scheduler.pt

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:8dc44d9525e28449178817fd7456c24041b688fdd22e4d5015874b3026a0e057
+size 627

checkpoint-1100/special_tokens_map.json

@@ -0,0 +1,6 @@
+{
+  "bos_token": "<s>",
+  "eos_token": "</s>",
+  "pad_token": "<pad>",
+  "unk_token": "<unk>"
+}

checkpoint-1100/tokenizer.json

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:17a208233d2ee8d8c83b23bc214df737c44806a1919f444e89b31e586cd956ba
+size 14500471

checkpoint-1100/tokenizer_config.json

@@ -0,0 +1,10 @@
+{
+  "add_prefix_space": false,
+  "bos_token": "<s>",
+  "clean_up_tokenization_spaces": false,
+  "eos_token": "</s>",
+  "model_max_length": 1000000000000000019884624838656,
+  "pad_token": "<pad>",
+  "tokenizer_class": "BloomTokenizer",
+  "unk_token": "<unk>"
+}

checkpoint-1100/training_args.bin

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:7015d16f0f7058d5e16036c5259111dacee395cddfb5fc5ae08d917da3d67f65
+size 4155

checkpoint-1200/README.md

@@ -0,0 +1,34 @@
+---
+library_name: peft
+---
+## Training procedure
+
+
+The following `bitsandbytes` quantization config was used during training:
+- quant_method: bitsandbytes
+- load_in_8bit: True
+- load_in_4bit: False
+- 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: fp4
+- bnb_4bit_use_double_quant: False
+- bnb_4bit_compute_dtype: float32
+
+The following `bitsandbytes` quantization config was used during training:
+- quant_method: bitsandbytes
+- load_in_8bit: True
+- load_in_4bit: False
+- 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: fp4
+- bnb_4bit_use_double_quant: False
+- bnb_4bit_compute_dtype: float32
+### Framework versions
+
+- PEFT 0.6.0.dev0
+
+- PEFT 0.6.0.dev0

checkpoint-1200/adapter_config.json

@@ -0,0 +1,23 @@
+{
+  "auto_mapping": null,
+  "base_model_name_or_path": "bigscience/bloomz-3b",
+  "bias": "none",
+  "fan_in_fan_out": false,
+  "inference_mode": true,
+  "init_lora_weights": true,
+  "layers_pattern": null,
+  "layers_to_transform": null,
+  "lora_alpha": 16,
+  "lora_dropout": 0.0,
+  "modules_to_save": null,
+  "peft_type": "LORA",
+  "r": 8,
+  "revision": null,
+  "target_modules": [
+    "dense_4h_to_h",
+    "dense",
+    "dense_h_to_4h",
+    "query_key_value"
+  ],
+  "task_type": "CAUSAL_LM"
+}

checkpoint-1200/adapter_model.bin

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:12968dd3374591da9108d33e9902eb7d308edc5381a57c32c4be86471025fe20
+size 39409357

checkpoint-1200/optimizer.pt

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:3441b8e2f6f31ce90cfae67eb2109bfdede30ba5ad2d8e44cbbe7c61214bd3f2
+size 78844421

checkpoint-1200/rng_state.pth

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:91db4cb8b4c509d633633237f7fc5865ed0256714ee41edb2e45f150f9672a02
+size 14575

checkpoint-1200/scheduler.pt

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:d75dac35039ac49dba103114c80c66a12f59c7fd524db7ff27a94928098ce8c4
+size 627

checkpoint-1200/special_tokens_map.json

@@ -0,0 +1,6 @@
+{
+  "bos_token": "<s>",
+  "eos_token": "</s>",
+  "pad_token": "<pad>",
+  "unk_token": "<unk>"
+}

checkpoint-1200/tokenizer.json

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:17a208233d2ee8d8c83b23bc214df737c44806a1919f444e89b31e586cd956ba
+size 14500471

checkpoint-1200/tokenizer_config.json

@@ -0,0 +1,10 @@
+{
+  "add_prefix_space": false,
+  "bos_token": "<s>",
+  "clean_up_tokenization_spaces": false,
+  "eos_token": "</s>",
+  "model_max_length": 1000000000000000019884624838656,
+  "pad_token": "<pad>",
+  "tokenizer_class": "BloomTokenizer",
+  "unk_token": "<unk>"
+}

checkpoint-1200/training_args.bin

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:7015d16f0f7058d5e16036c5259111dacee395cddfb5fc5ae08d917da3d67f65
+size 4155

checkpoint-300/README.md

@@ -0,0 +1,34 @@
+---
+library_name: peft
+---
+## Training procedure
+
+
+The following `bitsandbytes` quantization config was used during training:
+- quant_method: bitsandbytes
+- load_in_8bit: True
+- load_in_4bit: False
+- 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: fp4
+- bnb_4bit_use_double_quant: False
+- bnb_4bit_compute_dtype: float32
+
+The following `bitsandbytes` quantization config was used during training:
+- quant_method: bitsandbytes
+- load_in_8bit: True
+- load_in_4bit: False
+- 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: fp4
+- bnb_4bit_use_double_quant: False
+- bnb_4bit_compute_dtype: float32
+### Framework versions
+
+- PEFT 0.6.0.dev0
+
+- PEFT 0.6.0.dev0

checkpoint-300/adapter_config.json

@@ -0,0 +1,23 @@
+{
+  "auto_mapping": null,
+  "base_model_name_or_path": "bigscience/bloomz-3b",
+  "bias": "none",
+  "fan_in_fan_out": false,
+  "inference_mode": true,
+  "init_lora_weights": true,
+  "layers_pattern": null,
+  "layers_to_transform": null,
+  "lora_alpha": 16,
+  "lora_dropout": 0.0,
+  "modules_to_save": null,
+  "peft_type": "LORA",
+  "r": 8,
+  "revision": null,
+  "target_modules": [
+    "dense_4h_to_h",
+    "dense",
+    "dense_h_to_4h",
+    "query_key_value"
+  ],
+  "task_type": "CAUSAL_LM"
+}

checkpoint-300/adapter_model.bin

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:5dfc1b8458803ffd860d90eb4599c2284b60aa39a418a44a3966aac16492e688
+size 39409357

checkpoint-300/optimizer.pt

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:4050c9e817810a65825f67449896912c0286675a222fadfa4cf24a2ea357f5aa
+size 78844421

checkpoint-300/rng_state.pth

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:36fc71bd44bd7f04f2599c5dface64c517de1a7ab7bac3600f3f6470c6c72673
+size 14575

checkpoint-300/scheduler.pt

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:2173731890e5f0214a48e0accca0e2767cccd021afad070b099ba172c002c0b0
+size 627

checkpoint-300/special_tokens_map.json

@@ -0,0 +1,6 @@
+{
+  "bos_token": "<s>",
+  "eos_token": "</s>",
+  "pad_token": "<pad>",
+  "unk_token": "<unk>"
+}

checkpoint-300/tokenizer.json

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:17a208233d2ee8d8c83b23bc214df737c44806a1919f444e89b31e586cd956ba
+size 14500471

checkpoint-300/tokenizer_config.json

@@ -0,0 +1,10 @@
+{
+  "add_prefix_space": false,
+  "bos_token": "<s>",
+  "clean_up_tokenization_spaces": false,
+  "eos_token": "</s>",
+  "model_max_length": 1000000000000000019884624838656,
+  "pad_token": "<pad>",
+  "tokenizer_class": "BloomTokenizer",
+  "unk_token": "<unk>"
+}

checkpoint-300/trainer_state.json

@@ -0,0 +1,1819 @@
+{
+  "best_metric": null,
+  "best_model_checkpoint": null,
+  "epoch": 7.559055118110236,
+  "eval_steps": 500,
+  "global_step": 300,
+  "is_hyper_param_search": false,
+  "is_local_process_zero": true,
+  "is_world_process_zero": true,
+  "log_history": [
+    {
+      "epoch": 0.03,
+      "learning_rate": 0.000499599358974359,
+      "loss": 3.1389,
+      "step": 1
+    },
+    {
+      "epoch": 0.05,
+      "learning_rate": 0.000499198717948718,
+      "loss": 2.8569,
+      "step": 2
+    },
+    {
+      "epoch": 0.08,
+      "learning_rate": 0.0004987980769230769,
+      "loss": 2.75,
+      "step": 3
+    },
+    {
+      "epoch": 0.1,
+      "learning_rate": 0.0004983974358974359,
+      "loss": 2.7375,
+      "step": 4
+    },
+    {
+      "epoch": 0.13,
+      "learning_rate": 0.0004979967948717949,
+      "loss": 2.7342,
+      "step": 5
+    },
+    {
+      "epoch": 0.15,
+      "learning_rate": 0.0004975961538461539,
+      "loss": 2.753,
+      "step": 6
+    },
+    {
+      "epoch": 0.18,
+      "learning_rate": 0.0004971955128205128,
+      "loss": 2.7124,
+      "step": 7
+    },
+    {
+      "epoch": 0.2,
+      "learning_rate": 0.0004967948717948718,
+      "loss": 2.7038,
+      "step": 8
+    },
+    {
+      "epoch": 0.23,
+      "learning_rate": 0.0004963942307692308,
+      "loss": 2.6142,
+      "step": 9
+    },
+    {
+      "epoch": 0.25,
+      "learning_rate": 0.0004959935897435898,
+      "loss": 2.6427,
+      "step": 10
+    },
+    {
+      "epoch": 0.28,
+      "learning_rate": 0.0004955929487179487,
+      "loss": 2.6265,
+      "step": 11
+    },
+    {
+      "epoch": 0.3,
+      "learning_rate": 0.0004951923076923077,
+      "loss": 2.6352,
+      "step": 12
+    },
+    {
+      "epoch": 0.33,
+      "learning_rate": 0.0004947916666666667,
+      "loss": 2.616,
+      "step": 13
+    },
+    {
+      "epoch": 0.35,
+      "learning_rate": 0.0004943910256410257,
+      "loss": 2.6937,
+      "step": 14
+    },
+    {
+      "epoch": 0.38,
+      "learning_rate": 0.0004939903846153846,
+      "loss": 2.6719,
+      "step": 15
+    },
+    {
+      "epoch": 0.4,
+      "learning_rate": 0.0004935897435897436,
+      "loss": 2.6043,
+      "step": 16
+    },
+    {
+      "epoch": 0.43,
+      "learning_rate": 0.0004931891025641026,
+      "loss": 2.565,
+      "step": 17
+    },
+    {
+      "epoch": 0.45,
+      "learning_rate": 0.0004927884615384616,
+      "loss": 2.5776,
+      "step": 18
+    },
+    {
+      "epoch": 0.48,
+      "learning_rate": 0.0004923878205128205,
+      "loss": 2.6326,
+      "step": 19
+    },
+    {
+      "epoch": 0.5,
+      "learning_rate": 0.0004919871794871795,
+      "loss": 2.5868,
+      "step": 20
+    },
+    {
+      "epoch": 0.53,
+      "learning_rate": 0.0004915865384615384,
+      "loss": 2.5499,
+      "step": 21
+    },
+    {
+      "epoch": 0.55,
+      "learning_rate": 0.0004911858974358974,
+      "loss": 2.5823,
+      "step": 22
+    },
+    {
+      "epoch": 0.58,
+      "learning_rate": 0.0004907852564102564,
+      "loss": 2.5858,
+      "step": 23
+    },
+    {
+      "epoch": 0.6,
+      "learning_rate": 0.0004903846153846154,
+      "loss": 2.5688,
+      "step": 24
+    },
+    {
+      "epoch": 0.63,
+      "learning_rate": 0.0004899839743589743,
+      "loss": 2.57,
+      "step": 25
+    },
+    {
+      "epoch": 0.66,
+      "learning_rate": 0.0004895833333333333,
+      "loss": 2.5569,
+      "step": 26
+    },
+    {
+      "epoch": 0.68,
+      "learning_rate": 0.0004891826923076923,
+      "loss": 2.5067,
+      "step": 27
+    },
+    {
+      "epoch": 0.71,
+      "learning_rate": 0.0004887820512820513,
+      "loss": 2.4936,
+      "step": 28
+    },
+    {
+      "epoch": 0.73,
+      "learning_rate": 0.0004883814102564103,
+      "loss": 2.5958,
+      "step": 29
+    },
+    {
+      "epoch": 0.76,
+      "learning_rate": 0.00048798076923076925,
+      "loss": 2.5213,
+      "step": 30
+    },
+    {
+      "epoch": 0.78,
+      "learning_rate": 0.0004875801282051282,
+      "loss": 2.6369,
+      "step": 31
+    },
+    {
+      "epoch": 0.81,
+      "learning_rate": 0.0004871794871794872,
+      "loss": 2.578,
+      "step": 32
+    },
+    {
+      "epoch": 0.83,
+      "learning_rate": 0.00048677884615384615,
+      "loss": 2.6214,
+      "step": 33
+    },
+    {
+      "epoch": 0.86,
+      "learning_rate": 0.00048637820512820515,
+      "loss": 2.6193,
+      "step": 34
+    },
+    {
+      "epoch": 0.88,
+      "learning_rate": 0.0004859775641025641,
+      "loss": 2.5948,
+      "step": 35
+    },
+    {
+      "epoch": 0.91,
+      "learning_rate": 0.0004855769230769231,
+      "loss": 2.5615,
+      "step": 36
+    },
+    {
+      "epoch": 0.93,
+      "learning_rate": 0.00048517628205128205,
+      "loss": 2.5246,
+      "step": 37
+    },
+    {
+      "epoch": 0.96,
+      "learning_rate": 0.00048477564102564105,
+      "loss": 2.4585,
+      "step": 38
+    },
+    {
+      "epoch": 0.98,
+      "learning_rate": 0.000484375,
+      "loss": 2.5621,
+      "step": 39
+    },
+    {
+      "epoch": 1.01,
+      "learning_rate": 0.000483974358974359,
+      "loss": 2.571,
+      "step": 40
+    },
+    {
+      "epoch": 1.03,
+      "learning_rate": 0.00048357371794871795,
+      "loss": 2.5382,
+      "step": 41
+    },
+    {
+      "epoch": 1.06,
+      "learning_rate": 0.00048317307692307695,
+      "loss": 2.5028,
+      "step": 42
+    },
+    {
+      "epoch": 1.08,
+      "learning_rate": 0.0004827724358974359,
+      "loss": 2.4812,
+      "step": 43
+    },
+    {
+      "epoch": 1.11,
+      "learning_rate": 0.0004823717948717949,
+      "loss": 2.5161,
+      "step": 44
+    },
+    {
+      "epoch": 1.13,
+      "learning_rate": 0.00048197115384615384,
+      "loss": 2.4557,
+      "step": 45
+    },
+    {
+      "epoch": 1.16,
+      "learning_rate": 0.00048157051282051285,
+      "loss": 2.4617,
+      "step": 46
+    },
+    {
+      "epoch": 1.18,
+      "learning_rate": 0.0004811698717948718,
+      "loss": 2.5267,
+      "step": 47
+    },
+    {
+      "epoch": 1.21,
+      "learning_rate": 0.0004807692307692308,
+      "loss": 2.4555,
+      "step": 48
+    },
+    {
+      "epoch": 1.23,
+      "learning_rate": 0.00048036858974358974,
+      "loss": 2.37,
+      "step": 49
+    },
+    {
+      "epoch": 1.26,
+      "learning_rate": 0.00047996794871794875,
+      "loss": 2.4829,
+      "step": 50
+    },
+    {
+      "epoch": 1.29,
+      "learning_rate": 0.0004795673076923077,
+      "loss": 2.5022,
+      "step": 51
+    },
+    {
+      "epoch": 1.31,
+      "learning_rate": 0.0004791666666666667,
+      "loss": 2.4958,
+      "step": 52
+    },
+    {
+      "epoch": 1.34,
+      "learning_rate": 0.0004787660256410257,
+      "loss": 2.4358,
+      "step": 53
+    },
+    {
+      "epoch": 1.36,
+      "learning_rate": 0.00047836538461538464,
+      "loss": 2.4742,
+      "step": 54
+    },
+    {
+      "epoch": 1.39,
+      "learning_rate": 0.00047796474358974365,
+      "loss": 2.4158,
+      "step": 55
+    },
+    {
+      "epoch": 1.41,
+      "learning_rate": 0.0004775641025641026,
+      "loss": 2.5216,
+      "step": 56
+    },
+    {
+      "epoch": 1.44,
+      "learning_rate": 0.0004771634615384616,
+      "loss": 2.5239,
+      "step": 57
+    },
+    {
+      "epoch": 1.46,
+      "learning_rate": 0.00047676282051282054,
+      "loss": 2.3946,
+      "step": 58
+    },
+    {
+      "epoch": 1.49,
+      "learning_rate": 0.0004763621794871795,
+      "loss": 2.4457,
+      "step": 59
+    },
+    {
+      "epoch": 1.51,
+      "learning_rate": 0.00047596153846153844,
+      "loss": 2.4379,
+      "step": 60
+    },
+    {
+      "epoch": 1.54,
+      "learning_rate": 0.00047556089743589744,
+      "loss": 2.4812,
+      "step": 61
+    },
+    {
+      "epoch": 1.56,
+      "learning_rate": 0.0004751602564102564,
+      "loss": 2.4153,
+      "step": 62
+    },
+    {
+      "epoch": 1.59,
+      "learning_rate": 0.0004747596153846154,
+      "loss": 2.4115,
+      "step": 63
+    },
+    {
+      "epoch": 1.61,
+      "learning_rate": 0.00047435897435897434,
+      "loss": 2.3637,
+      "step": 64
+    },
+    {
+      "epoch": 1.64,
+      "learning_rate": 0.00047395833333333334,
+      "loss": 2.3741,
+      "step": 65
+    },
+    {
+      "epoch": 1.66,
+      "learning_rate": 0.0004735576923076923,
+      "loss": 2.4461,
+      "step": 66
+    },
+    {
+      "epoch": 1.69,
+      "learning_rate": 0.0004731570512820513,
+      "loss": 2.4694,
+      "step": 67
+    },
+    {
+      "epoch": 1.71,
+      "learning_rate": 0.00047275641025641024,
+      "loss": 2.5177,
+      "step": 68
+    },
+    {
+      "epoch": 1.74,
+      "learning_rate": 0.00047235576923076924,
+      "loss": 2.4964,
+      "step": 69
+    },
+    {
+      "epoch": 1.76,
+      "learning_rate": 0.0004719551282051282,
+      "loss": 2.5114,
+      "step": 70
+    },
+    {
+      "epoch": 1.79,
+      "learning_rate": 0.0004715544871794872,
+      "loss": 2.4447,
+      "step": 71
+    },
+    {
+      "epoch": 1.81,
+      "learning_rate": 0.00047115384615384613,
+      "loss": 2.4112,
+      "step": 72
+    },
+    {
+      "epoch": 1.84,
+      "learning_rate": 0.00047075320512820514,
+      "loss": 2.4585,
+      "step": 73
+    },
+    {
+      "epoch": 1.86,
+      "learning_rate": 0.0004703525641025641,
+      "loss": 2.4696,
+      "step": 74
+    },
+    {
+      "epoch": 1.89,
+      "learning_rate": 0.0004699519230769231,
+      "loss": 2.3813,
+      "step": 75
+    },
+    {
+      "epoch": 1.91,
+      "learning_rate": 0.00046955128205128203,
+      "loss": 2.4962,
+      "step": 76
+    },
+    {
+      "epoch": 1.94,
+      "learning_rate": 0.00046915064102564104,
+      "loss": 2.4392,
+      "step": 77
+    },
+    {
+      "epoch": 1.97,
+      "learning_rate": 0.00046875,
+      "loss": 2.3436,
+      "step": 78
+    },
+    {
+      "epoch": 1.99,
+      "learning_rate": 0.000468349358974359,
+      "loss": 2.4758,
+      "step": 79
+    },
+    {
+      "epoch": 2.02,
+      "learning_rate": 0.000467948717948718,
+      "loss": 2.4026,
+      "step": 80
+    },
+    {
+      "epoch": 2.04,
+      "learning_rate": 0.00046754807692307693,
+      "loss": 2.3728,
+      "step": 81
+    },
+    {
+      "epoch": 2.07,
+      "learning_rate": 0.00046714743589743594,
+      "loss": 2.395,
+      "step": 82
+    },
+    {
+      "epoch": 2.09,
+      "learning_rate": 0.0004667467948717949,
+      "loss": 2.2767,
+      "step": 83
+    },
+    {
+      "epoch": 2.12,
+      "learning_rate": 0.0004663461538461539,
+      "loss": 2.3483,
+      "step": 84
+    },
+    {
+      "epoch": 2.14,
+      "learning_rate": 0.00046594551282051283,
+      "loss": 2.2911,
+      "step": 85
+    },
+    {
+      "epoch": 2.17,
+      "learning_rate": 0.00046554487179487183,
+      "loss": 2.3502,
+      "step": 86
+    },
+    {
+      "epoch": 2.19,
+      "learning_rate": 0.0004651442307692308,
+      "loss": 2.3532,
+      "step": 87
+    },
+    {
+      "epoch": 2.22,
+      "learning_rate": 0.0004647435897435898,
+      "loss": 2.266,
+      "step": 88
+    },
+    {
+      "epoch": 2.24,
+      "learning_rate": 0.00046434294871794873,
+      "loss": 2.3046,
+      "step": 89
+    },
+    {
+      "epoch": 2.27,
+      "learning_rate": 0.00046394230769230773,
+      "loss": 2.2458,
+      "step": 90
+    },
+    {
+      "epoch": 2.29,
+      "learning_rate": 0.0004635416666666667,
+      "loss": 2.3324,
+      "step": 91
+    },
+    {
+      "epoch": 2.32,
+      "learning_rate": 0.0004631410256410257,
+      "loss": 2.3016,
+      "step": 92
+    },
+    {
+      "epoch": 2.34,
+      "learning_rate": 0.00046274038461538463,
+      "loss": 2.379,
+      "step": 93
+    },
+    {
+      "epoch": 2.37,
+      "learning_rate": 0.00046233974358974363,
+      "loss": 2.3075,
+      "step": 94
+    },
+    {
+      "epoch": 2.39,
+      "learning_rate": 0.0004619391025641026,
+      "loss": 2.3942,
+      "step": 95
+    },
+    {
+      "epoch": 2.42,
+      "learning_rate": 0.0004615384615384616,
+      "loss": 2.2891,
+      "step": 96
+    },
+    {
+      "epoch": 2.44,
+      "learning_rate": 0.00046113782051282053,
+      "loss": 2.4199,
+      "step": 97
+    },
+    {
+      "epoch": 2.47,
+      "learning_rate": 0.0004607371794871795,
+      "loss": 2.3662,
+      "step": 98
+    },
+    {
+      "epoch": 2.49,
+      "learning_rate": 0.0004603365384615384,
+      "loss": 2.3346,
+      "step": 99
+    },
+    {
+      "epoch": 2.52,
+      "learning_rate": 0.0004599358974358974,
+      "loss": 2.2607,
+      "step": 100
+    },
+    {
+      "epoch": 2.54,
+      "learning_rate": 0.0004595352564102564,
+      "loss": 2.3446,
+      "step": 101
+    },
+    {
+      "epoch": 2.57,
+      "learning_rate": 0.0004591346153846154,
+      "loss": 2.3649,
+      "step": 102
+    },
+    {
+      "epoch": 2.6,
+      "learning_rate": 0.0004587339743589743,
+      "loss": 2.2394,
+      "step": 103
+    },
+    {
+      "epoch": 2.62,
+      "learning_rate": 0.0004583333333333333,
+      "loss": 2.3278,
+      "step": 104
+    },
+    {
+      "epoch": 2.65,
+      "learning_rate": 0.0004579326923076923,
+      "loss": 2.2725,
+      "step": 105
+    },
+    {
+      "epoch": 2.67,
+      "learning_rate": 0.0004575320512820513,
+      "loss": 2.3328,
+      "step": 106
+    },
+    {
+      "epoch": 2.7,
+      "learning_rate": 0.0004571314102564103,
+      "loss": 2.296,
+      "step": 107
+    },
+    {
+      "epoch": 2.72,
+      "learning_rate": 0.0004567307692307692,
+      "loss": 2.2562,
+      "step": 108
+    },
+    {
+      "epoch": 2.75,
+      "learning_rate": 0.0004563301282051282,
+      "loss": 2.3062,
+      "step": 109
+    },
+    {
+      "epoch": 2.77,
+      "learning_rate": 0.0004559294871794872,
+      "loss": 2.3125,
+      "step": 110
+    },
+    {
+      "epoch": 2.8,
+      "learning_rate": 0.0004555288461538462,
+      "loss": 2.3257,
+      "step": 111
+    },
+    {
+      "epoch": 2.82,
+      "learning_rate": 0.0004551282051282051,
+      "loss": 2.3431,
+      "step": 112
+    },
+    {
+      "epoch": 2.85,
+      "learning_rate": 0.0004547275641025641,
+      "loss": 2.3507,
+      "step": 113
+    },
+    {
+      "epoch": 2.87,
+      "learning_rate": 0.00045432692307692307,
+      "loss": 2.3621,
+      "step": 114
+    },
+    {
+      "epoch": 2.9,
+      "learning_rate": 0.0004539262820512821,
+      "loss": 2.2937,
+      "step": 115
+    },
+    {
+      "epoch": 2.92,
+      "learning_rate": 0.000453525641025641,
+      "loss": 2.3356,
+      "step": 116
+    },
+    {
+      "epoch": 2.95,
+      "learning_rate": 0.000453125,
+      "loss": 2.2228,
+      "step": 117
+    },
+    {
+      "epoch": 2.97,
+      "learning_rate": 0.00045272435897435897,
+      "loss": 2.3189,
+      "step": 118
+    },
+    {
+      "epoch": 3.0,
+      "learning_rate": 0.00045232371794871797,
+      "loss": 2.3605,
+      "step": 119
+    },
+    {
+      "epoch": 3.02,
+      "learning_rate": 0.0004519230769230769,
+      "loss": 2.1857,
+      "step": 120
+    },
+    {
+      "epoch": 3.05,
+      "learning_rate": 0.0004515224358974359,
+      "loss": 2.1538,
+      "step": 121
+    },
+    {
+      "epoch": 3.07,
+      "learning_rate": 0.00045112179487179487,
+      "loss": 2.2092,
+      "step": 122
+    },
+    {
+      "epoch": 3.1,
+      "learning_rate": 0.00045072115384615387,
+      "loss": 2.2089,
+      "step": 123
+    },
+    {
+      "epoch": 3.12,
+      "learning_rate": 0.0004503205128205128,
+      "loss": 2.1746,
+      "step": 124
+    },
+    {
+      "epoch": 3.15,
+      "learning_rate": 0.0004499198717948718,
+      "loss": 2.1455,
+      "step": 125
+    },
+    {
+      "epoch": 3.17,
+      "learning_rate": 0.00044951923076923077,
+      "loss": 2.119,
+      "step": 126
+    },
+    {
+      "epoch": 3.2,
+      "learning_rate": 0.00044911858974358977,
+      "loss": 2.2341,
+      "step": 127
+    },
+    {
+      "epoch": 3.23,
+      "learning_rate": 0.0004487179487179487,
+      "loss": 2.2275,
+      "step": 128
+    },
+    {
+      "epoch": 3.25,
+      "learning_rate": 0.0004483173076923077,
+      "loss": 2.176,
+      "step": 129
+    },
+    {
+      "epoch": 3.28,
+      "learning_rate": 0.0004479166666666667,
+      "loss": 2.1187,
+      "step": 130
+    },
+    {
+      "epoch": 3.3,
+      "learning_rate": 0.00044751602564102567,
+      "loss": 2.1175,
+      "step": 131
+    },
+    {
+      "epoch": 3.33,
+      "learning_rate": 0.00044711538461538467,
+      "loss": 2.1818,
+      "step": 132
+    },
+    {
+      "epoch": 3.35,
+      "learning_rate": 0.0004467147435897436,
+      "loss": 2.1193,
+      "step": 133
+    },
+    {
+      "epoch": 3.38,
+      "learning_rate": 0.0004463141025641026,
+      "loss": 2.2079,
+      "step": 134
+    },
+    {
+      "epoch": 3.4,
+      "learning_rate": 0.00044591346153846157,
+      "loss": 2.1361,
+      "step": 135
+    },
+    {
+      "epoch": 3.43,
+      "learning_rate": 0.00044551282051282057,
+      "loss": 2.2246,
+      "step": 136
+    },
+    {
+      "epoch": 3.45,
+      "learning_rate": 0.00044511217948717946,
+      "loss": 2.1139,
+      "step": 137
+    },
+    {
+      "epoch": 3.48,
+      "learning_rate": 0.00044471153846153846,
+      "loss": 2.1058,
+      "step": 138
+    },
+    {
+      "epoch": 3.5,
+      "learning_rate": 0.0004443108974358974,
+      "loss": 2.1442,
+      "step": 139
+    },
+    {
+      "epoch": 3.53,
+      "learning_rate": 0.0004439102564102564,
+      "loss": 2.1039,
+      "step": 140
+    },
+    {
+      "epoch": 3.55,
+      "learning_rate": 0.00044350961538461536,
+      "loss": 2.1884,
+      "step": 141
+    },
+    {
+      "epoch": 3.58,
+      "learning_rate": 0.00044310897435897436,
+      "loss": 2.21,
+      "step": 142
+    },
+    {
+      "epoch": 3.6,
+      "learning_rate": 0.0004427083333333333,
+      "loss": 2.1139,
+      "step": 143
+    },
+    {
+      "epoch": 3.63,
+      "learning_rate": 0.0004423076923076923,
+      "loss": 2.1049,
+      "step": 144
+    },
+    {
+      "epoch": 3.65,
+      "learning_rate": 0.00044190705128205126,
+      "loss": 2.1282,
+      "step": 145
+    },
+    {
+      "epoch": 3.68,
+      "learning_rate": 0.00044150641025641026,
+      "loss": 2.1803,
+      "step": 146
+    },
+    {
+      "epoch": 3.7,
+      "learning_rate": 0.0004411057692307692,
+      "loss": 2.1406,
+      "step": 147
+    },
+    {
+      "epoch": 3.73,
+      "learning_rate": 0.0004407051282051282,
+      "loss": 2.1841,
+      "step": 148
+    },
+    {
+      "epoch": 3.75,
+      "learning_rate": 0.00044030448717948716,
+      "loss": 2.1961,
+      "step": 149
+    },
+    {
+      "epoch": 3.78,
+      "learning_rate": 0.00043990384615384616,
+      "loss": 2.1265,
+      "step": 150
+    },
+    {
+      "epoch": 3.8,
+      "learning_rate": 0.0004395032051282051,
+      "loss": 2.1708,
+      "step": 151
+    },
+    {
+      "epoch": 3.83,
+      "learning_rate": 0.0004391025641025641,
+      "loss": 2.1555,
+      "step": 152
+    },
+    {
+      "epoch": 3.86,
+      "learning_rate": 0.00043870192307692306,
+      "loss": 2.1463,
+      "step": 153
+    },
+    {
+      "epoch": 3.88,
+      "learning_rate": 0.00043830128205128206,
+      "loss": 2.1947,
+      "step": 154
+    },
+    {
+      "epoch": 3.91,
+      "learning_rate": 0.000437900641025641,
+      "loss": 2.1962,
+      "step": 155
+    },
+    {
+      "epoch": 3.93,
+      "learning_rate": 0.0004375,
+      "loss": 2.2109,
+      "step": 156
+    },
+    {
+      "epoch": 3.96,
+      "learning_rate": 0.000437099358974359,
+      "loss": 2.1556,
+      "step": 157
+    },
+    {
+      "epoch": 3.98,
+      "learning_rate": 0.00043669871794871796,
+      "loss": 2.1605,
+      "step": 158
+    },
+    {
+      "epoch": 4.01,
+      "learning_rate": 0.00043629807692307696,
+      "loss": 2.155,
+      "step": 159
+    },
+    {
+      "epoch": 4.03,
+      "learning_rate": 0.0004358974358974359,
+      "loss": 2.0347,
+      "step": 160
+    },
+    {
+      "epoch": 4.06,
+      "learning_rate": 0.0004354967948717949,
+      "loss": 1.9321,
+      "step": 161
+    },
+    {
+      "epoch": 4.08,
+      "learning_rate": 0.00043509615384615386,
+      "loss": 2.0034,
+      "step": 162
+    },
+    {
+      "epoch": 4.11,
+      "learning_rate": 0.00043469551282051286,
+      "loss": 2.0203,
+      "step": 163
+    },
+    {
+      "epoch": 4.13,
+      "learning_rate": 0.0004342948717948718,
+      "loss": 1.9189,
+      "step": 164
+    },
+    {
+      "epoch": 4.16,
+      "learning_rate": 0.0004338942307692308,
+      "loss": 1.9868,
+      "step": 165
+    },
+    {
+      "epoch": 4.18,
+      "learning_rate": 0.00043349358974358976,
+      "loss": 1.9343,
+      "step": 166
+    },
+    {
+      "epoch": 4.21,
+      "learning_rate": 0.00043309294871794876,
+      "loss": 1.8894,
+      "step": 167
+    },
+    {
+      "epoch": 4.23,
+      "learning_rate": 0.0004326923076923077,
+      "loss": 1.9735,
+      "step": 168
+    },
+    {
+      "epoch": 4.26,
+      "learning_rate": 0.0004322916666666667,
+      "loss": 2.0012,
+      "step": 169
+    },
+    {
+      "epoch": 4.28,
+      "learning_rate": 0.00043189102564102565,
+      "loss": 1.9087,
+      "step": 170
+    },
+    {
+      "epoch": 4.31,
+      "learning_rate": 0.00043149038461538466,
+      "loss": 1.9213,
+      "step": 171
+    },
+    {
+      "epoch": 4.33,
+      "learning_rate": 0.0004310897435897436,
+      "loss": 2.0195,
+      "step": 172
+    },
+    {
+      "epoch": 4.36,
+      "learning_rate": 0.0004306891025641026,
+      "loss": 1.9806,
+      "step": 173
+    },
+    {
+      "epoch": 4.38,
+      "learning_rate": 0.00043028846153846155,
+      "loss": 1.9731,
+      "step": 174
+    },
+    {
+      "epoch": 4.41,
+      "learning_rate": 0.00042988782051282056,
+      "loss": 1.9282,
+      "step": 175
+    },
+    {
+      "epoch": 4.43,
+      "learning_rate": 0.00042948717948717945,
+      "loss": 1.9433,
+      "step": 176
+    },
+    {
+      "epoch": 4.46,
+      "learning_rate": 0.00042908653846153845,
+      "loss": 1.9633,
+      "step": 177
+    },
+    {
+      "epoch": 4.49,
+      "learning_rate": 0.0004286858974358974,
+      "loss": 1.9918,
+      "step": 178
+    },
+    {
+      "epoch": 4.51,
+      "learning_rate": 0.0004282852564102564,
+      "loss": 1.9265,
+      "step": 179
+    },
+    {
+      "epoch": 4.54,
+      "learning_rate": 0.00042788461538461535,
+      "loss": 1.9198,
+      "step": 180
+    },
+    {
+      "epoch": 4.56,
+      "learning_rate": 0.00042748397435897435,
+      "loss": 2.0159,
+      "step": 181
+    },
+    {
+      "epoch": 4.59,
+      "learning_rate": 0.0004270833333333333,
+      "loss": 1.9474,
+      "step": 182
+    },
+    {
+      "epoch": 4.61,
+      "learning_rate": 0.0004266826923076923,
+      "loss": 2.0152,
+      "step": 183
+    },
+    {
+      "epoch": 4.64,
+      "learning_rate": 0.0004262820512820513,
+      "loss": 1.9006,
+      "step": 184
+    },
+    {
+      "epoch": 4.66,
+      "learning_rate": 0.00042588141025641025,
+      "loss": 1.9862,
+      "step": 185
+    },
+    {
+      "epoch": 4.69,
+      "learning_rate": 0.00042548076923076925,
+      "loss": 1.9945,
+      "step": 186
+    },
+    {
+      "epoch": 4.71,
+      "learning_rate": 0.0004250801282051282,
+      "loss": 1.985,
+      "step": 187
+    },
+    {
+      "epoch": 4.74,
+      "learning_rate": 0.0004246794871794872,
+      "loss": 1.9899,
+      "step": 188
+    },
+    {
+      "epoch": 4.76,
+      "learning_rate": 0.00042427884615384615,
+      "loss": 2.0149,
+      "step": 189
+    },
+    {
+      "epoch": 4.79,
+      "learning_rate": 0.00042387820512820515,
+      "loss": 1.9768,
+      "step": 190
+    },
+    {
+      "epoch": 4.81,
+      "learning_rate": 0.0004234775641025641,
+      "loss": 2.0204,
+      "step": 191
+    },
+    {
+      "epoch": 4.84,
+      "learning_rate": 0.0004230769230769231,
+      "loss": 1.9343,
+      "step": 192
+    },
+    {
+      "epoch": 4.86,
+      "learning_rate": 0.00042267628205128205,
+      "loss": 2.0574,
+      "step": 193
+    },
+    {
+      "epoch": 4.89,
+      "learning_rate": 0.00042227564102564105,
+      "loss": 2.0245,
+      "step": 194
+    },
+    {
+      "epoch": 4.91,
+      "learning_rate": 0.000421875,
+      "loss": 2.0356,
+      "step": 195
+    },
+    {
+      "epoch": 4.94,
+      "learning_rate": 0.000421474358974359,
+      "loss": 1.9984,
+      "step": 196
+    },
+    {
+      "epoch": 4.96,
+      "learning_rate": 0.00042107371794871794,
+      "loss": 1.9826,
+      "step": 197
+    },
+    {
+      "epoch": 4.99,
+      "learning_rate": 0.00042067307692307695,
+      "loss": 1.9846,
+      "step": 198
+    },
+    {
+      "epoch": 5.01,
+      "learning_rate": 0.0004202724358974359,
+      "loss": 1.8931,
+      "step": 199
+    },
+    {
+      "epoch": 5.04,
+      "learning_rate": 0.0004198717948717949,
+      "loss": 1.7642,
+      "step": 200
+    },
+    {
+      "epoch": 5.06,
+      "learning_rate": 0.00041947115384615384,
+      "loss": 1.7235,
+      "step": 201
+    },
+    {
+      "epoch": 5.09,
+      "learning_rate": 0.00041907051282051285,
+      "loss": 1.7058,
+      "step": 202
+    },
+    {
+      "epoch": 5.11,
+      "learning_rate": 0.0004186698717948718,
+      "loss": 1.8022,
+      "step": 203
+    },
+    {
+      "epoch": 5.14,
+      "learning_rate": 0.0004182692307692308,
+      "loss": 1.7819,
+      "step": 204
+    },
+    {
+      "epoch": 5.17,
+      "learning_rate": 0.00041786858974358974,
+      "loss": 1.7047,
+      "step": 205
+    },
+    {
+      "epoch": 5.19,
+      "learning_rate": 0.00041746794871794874,
+      "loss": 1.773,
+      "step": 206
+    },
+    {
+      "epoch": 5.22,
+      "learning_rate": 0.0004170673076923077,
+      "loss": 1.7324,
+      "step": 207
+    },
+    {
+      "epoch": 5.24,
+      "learning_rate": 0.0004166666666666667,
+      "loss": 1.7862,
+      "step": 208
+    },
+    {
+      "epoch": 5.27,
+      "learning_rate": 0.0004162660256410257,
+      "loss": 1.764,
+      "step": 209
+    },
+    {
+      "epoch": 5.29,
+      "learning_rate": 0.00041586538461538464,
+      "loss": 1.7156,
+      "step": 210
+    },
+    {
+      "epoch": 5.32,
+      "learning_rate": 0.00041546474358974364,
+      "loss": 1.7948,
+      "step": 211
+    },
+    {
+      "epoch": 5.34,
+      "learning_rate": 0.0004150641025641026,
+      "loss": 1.7608,
+      "step": 212
+    },
+    {
+      "epoch": 5.37,
+      "learning_rate": 0.0004146634615384616,
+      "loss": 1.7204,
+      "step": 213
+    },
+    {
+      "epoch": 5.39,
+      "learning_rate": 0.00041426282051282054,
+      "loss": 1.7691,
+      "step": 214
+    },
+    {
+      "epoch": 5.42,
+      "learning_rate": 0.0004138621794871795,
+      "loss": 1.7969,
+      "step": 215
+    },
+    {
+      "epoch": 5.44,
+      "learning_rate": 0.00041346153846153844,
+      "loss": 1.7194,
+      "step": 216
+    },
+    {
+      "epoch": 5.47,
+      "learning_rate": 0.00041306089743589744,
+      "loss": 1.7022,
+      "step": 217
+    },
+    {
+      "epoch": 5.49,
+      "learning_rate": 0.0004126602564102564,
+      "loss": 1.8287,
+      "step": 218
+    },
+    {
+      "epoch": 5.52,
+      "learning_rate": 0.0004122596153846154,
+      "loss": 1.7074,
+      "step": 219
+    },
+    {
+      "epoch": 5.54,
+      "learning_rate": 0.00041185897435897434,
+      "loss": 1.8354,
+      "step": 220
+    },
+    {
+      "epoch": 5.57,
+      "learning_rate": 0.00041145833333333334,
+      "loss": 1.791,
+      "step": 221
+    },
+    {
+      "epoch": 5.59,
+      "learning_rate": 0.0004110576923076923,
+      "loss": 1.7633,
+      "step": 222
+    },
+    {
+      "epoch": 5.62,
+      "learning_rate": 0.0004106570512820513,
+      "loss": 1.811,
+      "step": 223
+    },
+    {
+      "epoch": 5.64,
+      "learning_rate": 0.00041025641025641023,
+      "loss": 1.7498,
+      "step": 224
+    },
+    {
+      "epoch": 5.67,
+      "learning_rate": 0.00040985576923076924,
+      "loss": 1.7662,
+      "step": 225
+    },
+    {
+      "epoch": 5.69,
+      "learning_rate": 0.0004094551282051282,
+      "loss": 1.7686,
+      "step": 226
+    },
+    {
+      "epoch": 5.72,
+      "learning_rate": 0.0004090544871794872,
+      "loss": 1.8357,
+      "step": 227
+    },
+    {
+      "epoch": 5.74,
+      "learning_rate": 0.00040865384615384613,
+      "loss": 1.7981,
+      "step": 228
+    },
+    {
+      "epoch": 5.77,
+      "learning_rate": 0.00040825320512820513,
+      "loss": 1.7502,
+      "step": 229
+    },
+    {
+      "epoch": 5.8,
+      "learning_rate": 0.0004078525641025641,
+      "loss": 1.7666,
+      "step": 230
+    },
+    {
+      "epoch": 5.82,
+      "learning_rate": 0.0004074519230769231,
+      "loss": 1.798,
+      "step": 231
+    },
+    {
+      "epoch": 5.85,
+      "learning_rate": 0.00040705128205128203,
+      "loss": 1.8072,
+      "step": 232
+    },
+    {
+      "epoch": 5.87,
+      "learning_rate": 0.00040665064102564103,
+      "loss": 1.7611,
+      "step": 233
+    },
+    {
+      "epoch": 5.9,
+      "learning_rate": 0.00040625000000000004,
+      "loss": 1.8126,
+      "step": 234
+    },
+    {
+      "epoch": 5.92,
+      "learning_rate": 0.000405849358974359,
+      "loss": 1.8361,
+      "step": 235
+    },
+    {
+      "epoch": 5.95,
+      "learning_rate": 0.000405448717948718,
+      "loss": 1.7938,
+      "step": 236
+    },
+    {
+      "epoch": 5.97,
+      "learning_rate": 0.00040504807692307693,
+      "loss": 1.7524,
+      "step": 237
+    },
+    {
+      "epoch": 6.0,
+      "learning_rate": 0.00040464743589743593,
+      "loss": 1.7601,
+      "step": 238
+    },
+    {
+      "epoch": 6.02,
+      "learning_rate": 0.0004042467948717949,
+      "loss": 1.5373,
+      "step": 239
+    },
+    {
+      "epoch": 6.05,
+      "learning_rate": 0.0004038461538461539,
+      "loss": 1.5108,
+      "step": 240
+    },
+    {
+      "epoch": 6.07,
+      "learning_rate": 0.00040344551282051283,
+      "loss": 1.568,
+      "step": 241
+    },
+    {
+      "epoch": 6.1,
+      "learning_rate": 0.00040304487179487183,
+      "loss": 1.5175,
+      "step": 242
+    },
+    {
+      "epoch": 6.12,
+      "learning_rate": 0.0004026442307692308,
+      "loss": 1.5748,
+      "step": 243
+    },
+    {
+      "epoch": 6.15,
+      "learning_rate": 0.0004022435897435898,
+      "loss": 1.4772,
+      "step": 244
+    },
+    {
+      "epoch": 6.17,
+      "learning_rate": 0.00040184294871794873,
+      "loss": 1.5035,
+      "step": 245
+    },
+    {
+      "epoch": 6.2,
+      "learning_rate": 0.00040144230769230773,
+      "loss": 1.5318,
+      "step": 246
+    },
+    {
+      "epoch": 6.22,
+      "learning_rate": 0.0004010416666666667,
+      "loss": 1.5139,
+      "step": 247
+    },
+    {
+      "epoch": 6.25,
+      "learning_rate": 0.0004006410256410257,
+      "loss": 1.5599,
+      "step": 248
+    },
+    {
+      "epoch": 6.27,
+      "learning_rate": 0.00040024038461538463,
+      "loss": 1.5475,
+      "step": 249
+    },
+    {
+      "epoch": 6.3,
+      "learning_rate": 0.00039983974358974363,
+      "loss": 1.4819,
+      "step": 250
+    },
+    {
+      "epoch": 6.32,
+      "learning_rate": 0.0003994391025641026,
+      "loss": 1.5419,
+      "step": 251
+    },
+    {
+      "epoch": 6.35,
+      "learning_rate": 0.0003990384615384616,
+      "loss": 1.6027,
+      "step": 252
+    },
+    {
+      "epoch": 6.37,
+      "learning_rate": 0.00039863782051282053,
+      "loss": 1.5683,
+      "step": 253
+    },
+    {
+      "epoch": 6.4,
+      "learning_rate": 0.0003982371794871795,
+      "loss": 1.5963,
+      "step": 254
+    },
+    {
+      "epoch": 6.43,
+      "learning_rate": 0.0003978365384615384,
+      "loss": 1.5874,
+      "step": 255
+    },
+    {
+      "epoch": 6.45,
+      "learning_rate": 0.0003974358974358974,
+      "loss": 1.5161,
+      "step": 256
+    },
+    {
+      "epoch": 6.48,
+      "learning_rate": 0.00039703525641025637,
+      "loss": 1.57,
+      "step": 257
+    },
+    {
+      "epoch": 6.5,
+      "learning_rate": 0.0003966346153846154,
+      "loss": 1.534,
+      "step": 258
+    },
+    {
+      "epoch": 6.53,
+      "learning_rate": 0.0003962339743589743,
+      "loss": 1.5513,
+      "step": 259
+    },
+    {
+      "epoch": 6.55,
+      "learning_rate": 0.0003958333333333333,
+      "loss": 1.5454,
+      "step": 260
+    },
+    {
+      "epoch": 6.58,
+      "learning_rate": 0.0003954326923076923,
+      "loss": 1.5695,
+      "step": 261
+    },
+    {
+      "epoch": 6.6,
+      "learning_rate": 0.0003950320512820513,
+      "loss": 1.5777,
+      "step": 262
+    },
+    {
+      "epoch": 6.63,
+      "learning_rate": 0.0003946314102564103,
+      "loss": 1.5629,
+      "step": 263
+    },
+    {
+      "epoch": 6.65,
+      "learning_rate": 0.0003942307692307692,
+      "loss": 1.5031,
+      "step": 264
+    },
+    {
+      "epoch": 6.68,
+      "learning_rate": 0.0003938301282051282,
+      "loss": 1.5964,
+      "step": 265
+    },
+    {
+      "epoch": 6.7,
+      "learning_rate": 0.00039342948717948717,
+      "loss": 1.5472,
+      "step": 266
+    },
+    {
+      "epoch": 6.73,
+      "learning_rate": 0.0003930288461538462,
+      "loss": 1.5654,
+      "step": 267
+    },
+    {
+      "epoch": 6.75,
+      "learning_rate": 0.0003926282051282051,
+      "loss": 1.5932,
+      "step": 268
+    },
+    {
+      "epoch": 6.78,
+      "learning_rate": 0.0003922275641025641,
+      "loss": 1.6037,
+      "step": 269
+    },
+    {
+      "epoch": 6.8,
+      "learning_rate": 0.00039182692307692307,
+      "loss": 1.6394,
+      "step": 270
+    },
+    {
+      "epoch": 6.83,
+      "learning_rate": 0.00039142628205128207,
+      "loss": 1.5745,
+      "step": 271
+    },
+    {
+      "epoch": 6.85,
+      "learning_rate": 0.000391025641025641,
+      "loss": 1.6512,
+      "step": 272
+    },
+    {
+      "epoch": 6.88,
+      "learning_rate": 0.000390625,
+      "loss": 1.6107,
+      "step": 273
+    },
+    {
+      "epoch": 6.9,
+      "learning_rate": 0.00039022435897435897,
+      "loss": 1.5798,
+      "step": 274
+    },
+    {
+      "epoch": 6.93,
+      "learning_rate": 0.00038982371794871797,
+      "loss": 1.6087,
+      "step": 275
+    },
+    {
+      "epoch": 6.95,
+      "learning_rate": 0.0003894230769230769,
+      "loss": 1.5362,
+      "step": 276
+    },
+    {
+      "epoch": 6.98,
+      "learning_rate": 0.0003890224358974359,
+      "loss": 1.5899,
+      "step": 277
+    },
+    {
+      "epoch": 7.0,
+      "learning_rate": 0.00038862179487179487,
+      "loss": 1.59,
+      "step": 278
+    },
+    {
+      "epoch": 7.03,
+      "learning_rate": 0.00038822115384615387,
+      "loss": 1.3327,
+      "step": 279
+    },
+    {
+      "epoch": 7.06,
+      "learning_rate": 0.0003878205128205128,
+      "loss": 1.3214,
+      "step": 280
+    },
+    {
+      "epoch": 7.08,
+      "learning_rate": 0.0003874198717948718,
+      "loss": 1.2891,
+      "step": 281
+    },
+    {
+      "epoch": 7.11,
+      "learning_rate": 0.00038701923076923077,
+      "loss": 1.3321,
+      "step": 282
+    },
+    {
+      "epoch": 7.13,
+      "learning_rate": 0.00038661858974358977,
+      "loss": 1.284,
+      "step": 283
+    },
+    {
+      "epoch": 7.16,
+      "learning_rate": 0.0003862179487179487,
+      "loss": 1.3346,
+      "step": 284
+    },
+    {
+      "epoch": 7.18,
+      "learning_rate": 0.0003858173076923077,
+      "loss": 1.2874,
+      "step": 285
+    },
+    {
+      "epoch": 7.21,
+      "learning_rate": 0.0003854166666666667,
+      "loss": 1.3104,
+      "step": 286
+    },
+    {
+      "epoch": 7.23,
+      "learning_rate": 0.00038501602564102567,
+      "loss": 1.3138,
+      "step": 287
+    },
+    {
+      "epoch": 7.26,
+      "learning_rate": 0.00038461538461538467,
+      "loss": 1.2871,
+      "step": 288
+    },
+    {
+      "epoch": 7.28,
+      "learning_rate": 0.0003842147435897436,
+      "loss": 1.2885,
+      "step": 289
+    },
+    {
+      "epoch": 7.31,
+      "learning_rate": 0.0003838141025641026,
+      "loss": 1.338,
+      "step": 290
+    },
+    {
+      "epoch": 7.33,
+      "learning_rate": 0.00038341346153846157,
+      "loss": 1.3189,
+      "step": 291
+    },
+    {
+      "epoch": 7.36,
+      "learning_rate": 0.00038301282051282057,
+      "loss": 1.3013,
+      "step": 292
+    },
+    {
+      "epoch": 7.38,
+      "learning_rate": 0.00038261217948717946,
+      "loss": 1.3505,
+      "step": 293
+    },
+    {
+      "epoch": 7.41,
+      "learning_rate": 0.00038221153846153846,
+      "loss": 1.3469,
+      "step": 294
+    },
+    {
+      "epoch": 7.43,
+      "learning_rate": 0.0003818108974358974,
+      "loss": 1.3156,
+      "step": 295
+    },
+    {
+      "epoch": 7.46,
+      "learning_rate": 0.0003814102564102564,
+      "loss": 1.3796,
+      "step": 296
+    },
+    {
+      "epoch": 7.48,
+      "learning_rate": 0.00038100961538461536,
+      "loss": 1.3649,
+      "step": 297
+    },
+    {
+      "epoch": 7.51,
+      "learning_rate": 0.00038060897435897436,
+      "loss": 1.3483,
+      "step": 298
+    },
+    {
+      "epoch": 7.53,
+      "learning_rate": 0.0003802083333333333,
+      "loss": 1.3481,
+      "step": 299
+    },
+    {
+      "epoch": 7.56,
+      "learning_rate": 0.0003798076923076923,
+      "loss": 1.334,
+      "step": 300
+    }
+  ],
+  "logging_steps": 1,
+  "max_steps": 1248,
+  "num_train_epochs": 32,
+  "save_steps": 100,
+  "total_flos": 1.6887156698840064e+17,
+  "trial_name": null,
+  "trial_params": null
+}

checkpoint-300/training_args.bin

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:7015d16f0f7058d5e16036c5259111dacee395cddfb5fc5ae08d917da3d67f65
+size 4155

checkpoint-400/README.md

@@ -0,0 +1,34 @@
+---
+library_name: peft
+---
+## Training procedure
+
+
+The following `bitsandbytes` quantization config was used during training:
+- quant_method: bitsandbytes
+- load_in_8bit: True
+- load_in_4bit: False
+- 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: fp4
+- bnb_4bit_use_double_quant: False
+- bnb_4bit_compute_dtype: float32
+
+The following `bitsandbytes` quantization config was used during training:
+- quant_method: bitsandbytes
+- load_in_8bit: True
+- load_in_4bit: False
+- 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: fp4
+- bnb_4bit_use_double_quant: False
+- bnb_4bit_compute_dtype: float32
+### Framework versions
+
+- PEFT 0.6.0.dev0
+
+- PEFT 0.6.0.dev0

checkpoint-400/adapter_config.json

@@ -0,0 +1,23 @@
+{
+  "auto_mapping": null,
+  "base_model_name_or_path": "bigscience/bloomz-3b",
+  "bias": "none",
+  "fan_in_fan_out": false,
+  "inference_mode": true,
+  "init_lora_weights": true,
+  "layers_pattern": null,
+  "layers_to_transform": null,
+  "lora_alpha": 16,
+  "lora_dropout": 0.0,
+  "modules_to_save": null,
+  "peft_type": "LORA",
+  "r": 8,
+  "revision": null,
+  "target_modules": [
+    "dense_4h_to_h",
+    "dense",
+    "dense_h_to_4h",
+    "query_key_value"
+  ],
+  "task_type": "CAUSAL_LM"
+}

checkpoint-400/adapter_model.bin

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:04ef1cda8fd93cc8b65f531e16b8e1d8b4cdbebaab587c409f886214406138b2
+size 39409357

checkpoint-400/optimizer.pt

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:ad6ede7bca4b1fee9b3db1aa2d5d7d19ee2339a27761954fd1ec8e2552b307db
+size 78844421