llama board: add unsloth

src/llmtuner/webui/chatter.py

@@ -77,8 +77,8 @@ class WebChatModel(ChatModel):
             finetuning_type=get("top.finetuning_type"),
             quantization_bit=int(get("top.quantization_bit")) if get("top.quantization_bit") in ["8", "4"] else None,
             template=get("top.template"),
-            flash_attn=get("top.flash_attn"),
+            flash_attn=(get("top.booster") == "flash_attn"),
-            shift_attn=get("top.shift_attn"),
+            use_unsloth=(get("top.booster") == "unsloth"),
             rope_scaling=get("top.rope_scaling") if get("top.rope_scaling") in ["linear", "dynamic"] else None
         )
         super().__init__(args)

src/llmtuner/webui/components/top.py

@@ -28,10 +28,7 @@ def create_top() -> Dict[str, "Component"]:
             quantization_bit = gr.Dropdown(choices=["none", "8", "4"], value="none")
             template = gr.Dropdown(choices=list(templates.keys()), value="default")
             rope_scaling = gr.Radio(choices=["none", "linear", "dynamic"], value="none")
-
+            booster = gr.Radio(choices=["none", "flash_attn", "unsloth"], value="none")
-            with gr.Column():
-                flash_attn = gr.Checkbox(value=False)
-                shift_attn = gr.Checkbox(value=False)
 
     model_name.change(
         list_adapters, [model_name, finetuning_type], [adapter_path], queue=False
@@ -64,6 +61,5 @@ def create_top() -> Dict[str, "Component"]:
         quantization_bit=quantization_bit,
         template=template,
         rope_scaling=rope_scaling,
-        flash_attn=flash_attn,
+        booster=booster
-        shift_attn=shift_attn
     )

src/llmtuner/webui/locales.py

@@ -85,20 +85,12 @@ LOCALES = {
             "label": "RoPE 插值方法"
         }
     },
-    "flash_attn": {
+    "booster": {
         "en": {
-            "label": "Use FlashAttention-2"
+            "label": "Booster"
         },
         "zh": {
-            "label": "使用 FlashAttention-2"
+            "label": "加速方式"
-        }
-    },
-    "shift_attn": {
-        "en": {
-            "label": "Use shift short attention (S^2-Attn)"
-        },
-        "zh": {
-            "label": "使用 shift short attention (S^2-Attn)"
         }
     },
     "training_stage": {

src/llmtuner/webui/manager.py

@@ -25,9 +25,8 @@ class Manager:
             self.all_elems["top"]["finetuning_type"],
             self.all_elems["top"]["quantization_bit"],
             self.all_elems["top"]["template"],
-            self.all_elems["top"]["flash_attn"],
+            self.all_elems["top"]["rope_scaling"],
-            self.all_elems["top"]["shift_attn"],
+            self.all_elems["top"]["booster"]
-            self.all_elems["top"]["rope_scaling"]
         }
 
     def list_elems(self) -> List["Component"]:

src/llmtuner/webui/runner.py

@@ -102,9 +102,9 @@ class Runner:
             finetuning_type=get("top.finetuning_type"),
             quantization_bit=int(get("top.quantization_bit")) if get("top.quantization_bit") in ["8", "4"] else None,
             template=get("top.template"),
-            flash_attn=get("top.flash_attn"),
-            shift_attn=get("top.shift_attn"),
             rope_scaling=get("top.rope_scaling") if get("top.rope_scaling") in ["linear", "dynamic"] else None,
+            flash_attn=(get("top.booster") == "flash_attn"),
+            use_unsloth=(get("top.booster") == "unsloth"),
             dataset_dir=get("train.dataset_dir"),
             dataset=",".join(get("train.dataset")),
             cutoff_len=get("train.cutoff_len"),
@@ -171,9 +171,9 @@ class Runner:
             finetuning_type=get("top.finetuning_type"),
             quantization_bit=int(get("top.quantization_bit")) if get("top.quantization_bit") in ["8", "4"] else None,
             template=get("top.template"),
-            flash_attn=get("top.flash_attn"),
-            shift_attn=get("top.shift_attn"),
             rope_scaling=get("top.rope_scaling") if get("top.rope_scaling") in ["linear", "dynamic"] else None,
+            flash_attn=(get("top.booster") == "flash_attn"),
+            use_unsloth=(get("top.booster") == "unsloth"),
             dataset_dir=get("eval.dataset_dir"),
             dataset=",".join(get("eval.dataset")),
             cutoff_len=get("eval.cutoff_len"),