[Model] DeepseekV2 Support

README.md

@@ -255,6 +255,8 @@ loss.backward()
 | Qwen2, Qwen2.5, & QwQ      | `liger_kernel.transformers.apply_liger_kernel_to_qwen2`    | RoPE, RMSNorm, SwiGLU, CrossEntropyLoss, FusedLinearCrossEntropy        |
 | Qwen2-VL, & QVQ       | `liger_kernel.transformers.apply_liger_kernel_to_qwen2_vl`    | RMSNorm, LayerNorm, SwiGLU, CrossEntropyLoss, FusedLinearCrossEntropy        |
 | Phi3 & Phi3.5       | `liger_kernel.transformers.apply_liger_kernel_to_phi3`     | RoPE, RMSNorm, SwiGLU, CrossEntropyLoss, FusedLinearCrossEntropy         |
+| DeepSeekv2  | `liger_kernel.transformers.apply_liger_kernel_to_deepseek_v2` | RoPE, RMSNorm, SwiGLU, CrossEntropyLoss, FusedLinearCrossEntropy        |
+
 
 
 ## Low-level APIs

src/liger_kernel/transformers/__init__.py

@@ -7,6 +7,7 @@
 from liger_kernel.transformers.layer_norm import LigerLayerNorm  # noqa: F401
 from liger_kernel.transformers.monkey_patch import _apply_liger_kernel  # noqa: F401
 from liger_kernel.transformers.monkey_patch import _apply_liger_kernel_to_instance  # noqa: F401
+from liger_kernel.transformers.monkey_patch import apply_liger_kernel_to_deepseek_v2  # noqa: F401
 from liger_kernel.transformers.monkey_patch import apply_liger_kernel_to_gemma  # noqa: F401
 from liger_kernel.transformers.monkey_patch import apply_liger_kernel_to_gemma2  # noqa: F401
 from liger_kernel.transformers.monkey_patch import apply_liger_kernel_to_llama  # noqa: F401

src/liger_kernel/transformers/model/deepseekv2.py

@@ -0,0 +1,176 @@
+from typing import List
+from typing import Optional
+from typing import Tuple
+from typing import Union
+
+import torch
+
+from torch.nn import CrossEntropyLoss
+from transformers.modeling_outputs import CausalLMOutputWithPast
+from transformers.utils import add_start_docstrings_to_model_forward
+from transformers.utils import replace_return_docstrings
+
+from liger_kernel.transformers.fused_linear_cross_entropy import LigerFusedLinearCrossEntropyLoss
+
+# This docstring is ported from the DeepSeek V2 model source code.
+# Source: https://huggingface.co/deepseek-ai/DeepSeek-V2/blob/main/modeling_deepseek.py
+DeepseekV2_INPUTS_DOCSTRING = r"""
+    Args:
+        input_ids (`torch.LongTensor` of shape `(batch_size, sequence_length)`):
+            Indices of input sequence tokens in the vocabulary. Padding will be ignored by default should you provide
+            it.
+            Indices can be obtained using [`AutoTokenizer`]. See [`PreTrainedTokenizer.encode`] and
+            [`PreTrainedTokenizer.__call__`] for details.
+            [What are input IDs?](../glossary#input-ids)
+        attention_mask (`torch.Tensor` of shape `(batch_size, sequence_length)`, *optional*):
+            Mask to avoid performing attention on padding token indices. Mask values selected in `[0, 1]`:
+            - 1 for tokens that are **not masked**,
+            - 0 for tokens that are **masked**.
+            [What are attention masks?](../glossary#attention-mask)
+            Indices can be obtained using [`AutoTokenizer`]. See [`PreTrainedTokenizer.encode`] and
+            [`PreTrainedTokenizer.__call__`] for details.
+            If `past_key_values` is used, optionally only the last `input_ids` have to be input (see
+            `past_key_values`).
+            If you want to change padding behavior, you should read [`modeling_opt._prepare_decoder_attention_mask`]
+            and modify to your needs. See diagram 1 in [the paper](https://arxiv.org/abs/1910.13461) for more
+            information on the default strategy.
+            - 1 indicates the head is **not masked**,
+            - 0 indicates the head is **masked**.
+        position_ids (`torch.LongTensor` of shape `(batch_size, sequence_length)`, *optional*):
+            Indices of positions of each input sequence tokens in the position embeddings. Selected in the range `[0,
+            config.n_positions - 1]`.
+            [What are position IDs?](../glossary#position-ids)
+        past_key_values (`Cache` or `tuple(tuple(torch.FloatTensor))`, *optional*):
+            Pre-computed hidden-states (key and values in the self-attention blocks and in the cross-attention
+            blocks) that can be used to speed up sequential decoding. This typically consists in the `past_key_values`
+            returned by the model at a previous stage of decoding, when `use_cache=True` or `config.use_cache=True`.
+            Two formats are allowed:
+            - a [`~cache_utils.Cache`] instance;
+            - Tuple of `tuple(torch.FloatTensor)` of length `config.n_layers`, with each tuple having 2 tensors of
+            shape `(batch_size, num_heads, sequence_length, embed_size_per_head)`). This is also known as the legacy
+            cache format.
+            The model will output the same cache format that is fed as input. If no `past_key_values` are passed, the
+            legacy cache format will be returned.
+            If `past_key_values` are used, the user can optionally input only the last `input_ids` (those that don't
+            have their past key value states given to this model) of shape `(batch_size, 1)` instead of all `input_ids`
+            of shape `(batch_size, sequence_length)`.
+        inputs_embeds (`torch.FloatTensor` of shape `(batch_size, sequence_length, hidden_size)`, *optional*):
+            Optionally, instead of passing `input_ids` you can choose to directly pass an embedded representation. This
+            is useful if you want more control over how to convert `input_ids` indices into associated vectors than the
+            model's internal embedding lookup matrix.
+        use_cache (`bool`, *optional*):
+            If set to `True`, `past_key_values` key value states are returned and can be used to speed up decoding (see
+            `past_key_values`).
+        output_attentions (`bool`, *optional*):
+            Whether or not to return the attentions tensors of all attention layers. See `attentions` under returned
+            tensors for more detail.
+        output_hidden_states (`bool`, *optional*):
+            Whether or not to return the hidden states of all layers. See `hidden_states` under returned tensors for
+            more detail.
+        return_dict (`bool`, *optional*):
+            Whether or not to return a [`~utils.ModelOutput`] instead of a plain tuple.
+"""
+
+_CONFIG_FOR_DOC = "DeepseekV2Config"
+
+
+@add_start_docstrings_to_model_forward(DeepseekV2_INPUTS_DOCSTRING)
+@replace_return_docstrings(output_type=CausalLMOutputWithPast, config_class=_CONFIG_FOR_DOC)
+def lce_forward(
+    self,
+    input_ids: torch.LongTensor = None,
+    attention_mask: Optional[torch.Tensor] = None,
+    position_ids: Optional[torch.LongTensor] = None,
+    past_key_values: Optional[List[torch.FloatTensor]] = None,
+    inputs_embeds: Optional[torch.FloatTensor] = None,
+    labels: Optional[torch.LongTensor] = None,
+    use_cache: Optional[bool] = None,
+    output_attentions: Optional[bool] = None,
+    output_hidden_states: Optional[bool] = None,
+    return_dict: Optional[bool] = None,
+) -> Union[Tuple, CausalLMOutputWithPast]:
+    r"""
+    Copy paste deepseekv2 forward but replace torch cross entropy with liger fused linear cross entropy
+
+    Args:
+        labels (`torch.LongTensor` of shape `(batch_size, sequence_length)`, *optional*):
+            Labels for computing the masked language modeling loss. Indices should either be in `[0, transformers.,
+            config.vocab_size]` or -100 (see `input_ids` docstring). Tokens with indices set to `-100` are ignored
+            (masked), the loss is only computed for the tokens with labels in `[0, transformers., config.vocab_size]`.
+    Returns:
+    Example:
+    ```python
+    >>> from transformers import AutoTokenizer, DeepseekV2ForCausalLM
+    >>> model = DeepseekV2ForCausalLM.from_pretrained(PATH_TO_CONVERTED_WEIGHTS)
+    >>> tokenizer = AutoTokenizer.from_pretrained(PATH_TO_CONVERTED_TOKENIZER)
+    >>> prompt = "Hey, are you conscious? Can you talk to me?"
+    >>> inputs = tokenizer(prompt, return_tensors="pt")
+    >>> # Generate
+    >>> generate_ids = model.generate(inputs.input_ids, max_length=30)
+    >>> tokenizer.batch_decode(generate_ids, skip_special_tokens=True, clean_up_tokenization_spaces=False)[0]
+    "Hey, are you conscious? Can you talk to me?\nI'm not conscious, but I can talk to you."
+    ```"""
+
+    output_attentions = output_attentions if output_attentions is not None else self.config.output_attentions
+    output_hidden_states = (
+        output_hidden_states if output_hidden_states is not None else self.config.output_hidden_states
+    )
+    return_dict = return_dict if return_dict is not None else self.config.use_return_dict
+
+    # decoder outputs consists of (dec_features, layer_state, dec_hidden, dec_attn)
+    outputs = self.model(
+        input_ids=input_ids,
+        attention_mask=attention_mask,
+        position_ids=position_ids,
+        past_key_values=past_key_values,
+        inputs_embeds=inputs_embeds,
+        use_cache=use_cache,
+        output_attentions=output_attentions,
+        output_hidden_states=output_hidden_states,
+        return_dict=return_dict,
+    )
+
+    hidden_states = outputs[0]
+
+    loss = None
+    logits = None
+
+    if self.training and labels is not None:
+        shift_hidden_states = hidden_states[..., :-1, :].contiguous()
+        shift_labels = labels[..., 1:].contiguous()
+
+        # flatten tokens
+        shift_hidden_states = shift_hidden_states.view(-1, self.config.hidden_size)
+        shift_labels = shift_labels.view(-1)
+
+        lce = LigerFusedLinearCrossEntropyLoss()
+        loss = lce(self.lm_head.weight, shift_hidden_states, shift_labels)
+    else:
+        logits = self.lm_head(hidden_states)
+
+        loss = None
+        if labels is not None:
+            # Upcast to float if we need to compute the loss to avoid potential precision issues
+            logits = logits.float()
+            # Shift so that tokens < n predict n
+            shift_logits = logits[..., :-1, :].contiguous()
+            shift_labels = labels[..., 1:].contiguous()
+            # Flatten the tokens
+            loss_fct = CrossEntropyLoss()
+            shift_logits = shift_logits.view(-1, self.config.vocab_size)
+            shift_labels = shift_labels.view(-1)
+            # Enable model parallelism
+            shift_labels = shift_labels.to(shift_logits.device)
+            loss = loss_fct(shift_logits, shift_labels)
+
+    if not return_dict:
+        output = (logits,) + outputs[1:]
+        return (loss,) + output if loss is not None else output
+
+    return CausalLMOutputWithPast(
+        loss=loss,
+        logits=logits,
+        past_key_values=outputs.past_key_values,
+        hidden_states=outputs.hidden_states,
+        attentions=outputs.attentions,
+    )

src/liger_kernel/transformers/monkey_patch.py

@@ -13,6 +13,7 @@
 from liger_kernel.transformers.functional import liger_cross_entropy
 from liger_kernel.transformers.geglu import LigerGEGLUMLP
 from liger_kernel.transformers.layer_norm import LigerLayerNorm
+from liger_kernel.transformers.model.deepseekv2 import lce_forward as deepseek_v2_lce_forward
 from liger_kernel.transformers.model.gemma import lce_forward as gemma_lce_forward
 from liger_kernel.transformers.model.gemma import lce_forward_deprecated as gemma_lce_forward_deprecated
 from liger_kernel.transformers.model.gemma2 import lce_forward as gemma2_lce_forward
@@ -735,6 +736,80 @@ def apply_liger_kernel_to_phi3(
                 _patch_rms_norm_module(decoder_layer.post_attention_layernorm)
 
 
+def apply_liger_kernel_to_deepseek_v2(
+    rope: bool = True,
+    cross_entropy: bool = False,
+    fused_linear_cross_entropy: bool = True,
+    rms_norm: bool = True,
+    swiglu: bool = True,
+    model: PreTrainedModel = None,
+) -> None:
+    """
+    Apply Liger kernels to replace original implementations in DeepSeekv2 models.
+
+    Args:
+        rope (bool): Whether to apply Liger's rotary position embedding. Default is True.
+        cross_entropy (bool): Whether to apply Liger's cross entropy loss. Default is False.
+        fused_linear_cross_entropy (bool):
+            Whether to apply Liger's fused linear cross entropy loss. Default is True.
+            `cross_entropy` and `fused_linear_cross_entropy` cannot both be True.
+            If `fused_linear_cross_entropy` is True, the logits will not be materialized, enhancing memory efficiency.
+        rms_norm (bool): Whether to apply Liger's RMSNorm. Default is True.
+        swiglu (bool): Whether to apply Liger's SwiGLU MLP. Default is True.
+        model (PreTrainedModel): The model instance to apply Liger kernels to, if already loaded. Default is None.
+    """
+    assert not (
+        cross_entropy and fused_linear_cross_entropy
+    ), "cross_entropy and fused_linear_cross_entropy cannot both be True."
+
+    import sys
+
+    # Ensure the model is a DeepSeek model
+    if "deepseek" not in model.__class__.__module__:
+        raise ValueError("The provided model is not a DeepSeek model")
+
+    modeling_mod = sys.modules[model.__class__.__module__]
+
+    if rope:
+        pass
+        # modeling_mod.apply_rotary_pos_emb = liger_rotary_pos_emb
+    if rms_norm:
+        modeling_mod.DeepseekV2RMSNorm = LigerRMSNorm
+    if swiglu:
+        modeling_mod.DeepseekV2MLP.forward = LigerSwiGLUMLP.forward
+    if cross_entropy:
+        if transformer_version >= version.parse(SUPPORTED_TRANSFORMER_VERSION):
+            from transformers.loss.loss_utils import nn
+
+            nn.functional.cross_entropy = liger_cross_entropy
+        else:
+            logger.warning(TRANSFORMER_DEPRECATION_WARNING)
+            modeling_mod.CrossEntropyLoss = LigerCrossEntropyLoss
+    if fused_linear_cross_entropy:
+        modeling_mod.DeepseekV2ForCausalLM.forward = deepseek_v2_lce_forward
+
+    if model is not None:
+        # The model instance already exists, so we need to additionally patch the
+        # instance variables that reference already-instantiated modules
+
+        # get the base model from the model instance
+        base_model = getattr(model, model.base_model_prefix, model)
+
+        if rms_norm:
+            _patch_rms_norm_module(base_model.norm)
+
+        for i, decoder_layer in enumerate(base_model.layers):
+            if swiglu:
+                if i == 0:
+                    _bind_method_to_module(decoder_layer.mlp, "forward", LigerSwiGLUMLP.forward)
+                else:
+                    for expert in decoder_layer.mlp.experts:
+                        _bind_method_to_module(expert, "forward", LigerSwiGLUMLP.forward)
+            if rms_norm:
+                _patch_rms_norm_module(decoder_layer.self_attn.kv_a_layernorm)
+                _patch_rms_norm_module(decoder_layer.post_attention_layernorm)
+
+
 # Model type corresponds to the keys defined in transformers/models/auto/modeling_auto.py
 MODEL_TYPE_TO_APPLY_LIGER_FN = {
     "gemma": apply_liger_kernel_to_gemma,
@@ -747,6 +822,7 @@ def apply_liger_kernel_to_phi3(
     "qwen2": apply_liger_kernel_to_qwen2,
     "qwen2_vl": apply_liger_kernel_to_qwen2_vl,
     "phi3": apply_liger_kernel_to_phi3,
+    "deepseek_v2": apply_liger_kernel_to_deepseek_v2,
 }