Support Qwen2 (#428)

* support Qwen2

* support Qwen2

* Delete README.md

* Revert "Delete README.md"

This reverts commit 026b05f.

* Update README.md

* Qwen2 == Mistral

* Update llama.py

* Update __init__.py

* Update README.md

---------

Co-authored-by: Daniel Han <danielhanchen@gmail.com>

README.md

@@ -36,6 +36,7 @@ All notebooks are **beginner friendly**! Add your dataset, click "Run All", and
 - This [text completion notebook](https://colab.research.google.com/drive/1ef-tab5bhkvWmBOObepl1WgJvfvSzn5Q?usp=sharing) is for continued pretraining / raw text.
 
 ## 🦥 Unsloth.ai News
+- 📣 NEW! Qwen1.5-7B, Qwen1.5-14B, Qwen1.5-32B, Qwen1.5-72B now work, courtesy of Firefly's PR [#428](https://github.com/unslothai/unsloth/pull/428)
 - 📣 NEW! [Llama-3 8b](https://colab.research.google.com/drive/135ced7oHytdxu3N2DNe1Z0kqjyYIkDXp?usp=sharing) now works! Llama-3 70b also works (change the model name in the notebook).
 - 📣 NEW! [ORPO support](https://colab.research.google.com/drive/11t4njE3c4Lxl-07OD8lJSMKkfyJml3Tn?usp=sharing) is here!
 - 📣 NEW! [Phi-3 3.8b support](https://colab.research.google.com/drive/1NvkBmkHfucGO3Ve9s1NKZvMNlw5p83ym?usp=sharing) is here!

unsloth/models/__init__.py

@@ -12,7 +12,8 @@
 # See the License for the specific language governing permissions and
 # limitations under the License.
 
-from .loader import FastLanguageModel
-from .llama import FastLlamaModel
+from .loader  import FastLanguageModel
+from .llama   import FastLlamaModel
 from .mistral import FastMistralModel
-from .dpo import PatchDPOTrainer
+from .qwen2   import FastQwen2Model
+from .dpo     import PatchDPOTrainer

unsloth/models/llama.py

@@ -1605,6 +1605,7 @@ def patch_peft_model(
 
         if   model_type == "llama":   apply_lora_mlp = apply_lora_mlp_swiglu
         elif model_type == "mistral": apply_lora_mlp = apply_lora_mlp_swiglu
+        elif model_type == "qwen2":   apply_lora_mlp = apply_lora_mlp_swiglu
         elif model_type == "gemma":   apply_lora_mlp = apply_lora_mlp_geglu_approx
         else:
             raise NotImplementedError(f"Unsloth: {model_type} is not yet implemented!")

unsloth/models/loader.py

@@ -14,6 +14,7 @@
 
 from .llama import FastLlamaModel, logger
 from .mistral import FastMistralModel
+from .qwen2 import FastQwen2Model
 from transformers import AutoConfig
 from transformers import __version__ as transformers_version
 from peft import PeftConfig, PeftModel
@@ -119,6 +120,8 @@ def from_pretrained(
                     f"to obtain the latest transformers build, then restart this session."\
                 )
             dispatch_model = FastGemmaModel
+        elif model_type == "qwen2":
+            dispatch_model = FastQwen2Model
         else:
             raise NotImplementedError(
                 f"Unsloth: {model_name} not supported yet!\n"\

unsloth/models/mistral.py

@@ -343,7 +343,7 @@ def from_pretrained(
         # Mistral does NOT support RoPE Scaling sadly so we have to error out.
         if max_seq_length > model_max_seq_length:
             raise RuntimeError(
-                "Unsloth: Unfortunately Mistral type models do not support RoPE scaling!\n"\
+                f"Unsloth: Unfortunately {model_patcher.__name__[4:-5]} type models do not support RoPE scaling!\n"\
                 f"The maximum sequence length supported is {model_max_seq_length}.",
             )
         pass

unsloth/models/qwen2.py

@@ -0,0 +1,91 @@
+# Copyright 2023-present Daniel Han-Chen & the Unsloth team. All rights reserved.
+#
+# Licensed under the Apache License, Version 2.0 (the "License");
+# you may not use this file except in compliance with the License.
+# You may obtain a copy of the License at
+#
+#     http://www.apache.org/licenses/LICENSE-2.0
+#
+# Unless required by applicable law or agreed to in writing, software
+# distributed under the License is distributed on an "AS IS" BASIS,
+# WITHOUT WARRANTIES OR CONDITIONS OF ANY KIND, either express or implied.
+# See the License for the specific language governing permissions and
+# limitations under the License.
+
+from .llama import *
+from .mistral import FastMistralModel
+import os
+from ._utils import __version__
+
+from transformers.models.qwen2.modeling_qwen2 import (
+    Qwen2Attention,
+    Qwen2DecoderLayer,
+    Qwen2Model,
+    Qwen2ForCausalLM,
+)
+# For Pytorch 2.1.1
+try:
+    from transformers.models.qwen2.modeling_qwen2 import (
+        Qwen2SdpaAttention,
+        Qwen2FlashAttention2,
+    )
+except:
+    Qwen2SdpaAttention   = Qwen2Attention
+    Qwen2FlashAttention2 = Qwen2Attention
+pass
+
+
+class FastQwen2Model(FastLlamaModel):
+
+    @staticmethod
+    def pre_patch():
+        Qwen2Attention      .forward = LlamaAttention_fast_forward
+        Qwen2SdpaAttention  .forward = LlamaAttention_fast_forward
+        Qwen2FlashAttention2.forward = LlamaAttention_fast_forward
+        Qwen2DecoderLayer   .forward = LlamaDecoderLayer_fast_forward
+        Qwen2Model          .forward = LlamaModel_fast_forward
+        Qwen2ForCausalLM    .forward = CausalLM_fast_forward(LlamaModel_fast_forward_inference)
+        PeftModelForCausalLM.forward = PeftModelForCausalLM_fast_forward
+
+        # Solves https://github.com/unslothai/unsloth/issues/168
+        # Static KV Cache was introduced in 4.38.0, causing training to be much slower.
+        # Inferene can now be CUDAGraphed, but we shall retain the old rotary embeddings.
+        # https://github.com/huggingface/transformers/pull/27931
+        # https://github.com/huggingface/transformers/blob/v4.37.2/src/transformers/models/llama/modeling_llama.py
+        import transformers.models.qwen2.modeling_qwen2
+        transformers.models.qwen2.modeling_qwen2.Qwen2RotaryEmbedding = LlamaRotaryEmbedding
+        return
+    pass
+
+
+    @staticmethod
+    def from_pretrained(
+        model_name     = "Qwen/Qwen1.5-7B",
+        max_seq_length = 4096,
+        dtype          = None,
+        load_in_4bit   = True,
+        token          = None,
+        device_map     = "sequential",
+        rope_scaling   = None, # Qwen2 does not support RoPE scaling
+        fix_tokenizer  = True,
+        model_patcher  = None,
+        tokenizer_name = None,
+        trust_remote_code = False,
+        **kwargs,
+    ):
+        return FastMistralModel.from_pretrained(
+            model_name     = model_name,
+            max_seq_length = max_seq_length,
+            dtype          = dtype,
+            load_in_4bit   = load_in_4bit,
+            token          = token,
+            device_map     = device_map,
+            rope_scaling   = rope_scaling,
+            fix_tokenizer  = fix_tokenizer,
+            model_patcher  = FastQwen2Model,
+            tokenizer_name = tokenizer_name,
+            trust_remote_code = trust_remote_code,
+            **kwargs,
+        )
+    pass
+pass