optimum/habana/transformers/models/qwen2_moe/modeling_qwen2_moe.py

# coding=utf-8
# Copyright 2023 Mistral AI and the HuggingFace Inc. team. All rights reserved.
#
# This code is based on EleutherAI's GPT-NeoX library and the GPT-NeoX
# and OPT implementations in this library. It has been modified from its
# original forms to accommodate minor architectural differences compared
# to GPT-NeoX and OPT used by the Meta AI team that trained the model.
#
# 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.
"""PyTorch Qwen2MoE model."""

import math
import warnings
from typing import List, Optional, Tuple, Union

import habana_frameworks.torch.core as htcore
import torch
import torch.nn.functional as F
from transformers.cache_utils import Cache, DynamicCache, StaticCache
from transformers.integrations.deepspeed import is_deepspeed_available
from transformers.modeling_outputs import MoeCausalLMOutputWithPast, MoeModelOutputWithPast
from transformers.models.qwen2_moe.configuration_qwen2_moe import Qwen2MoeConfig
from transformers.models.qwen2_moe.modeling_qwen2_moe import (
    Qwen2MoeAttention,
    Qwen2MoeDecoderLayer,
    Qwen2MoeForCausalLM,
    Qwen2MoeMLP,
    Qwen2MoeModel,
    Qwen2MoeRMSNorm,
    Qwen2MoeSparseMoeBlock,
    apply_rotary_pos_emb,
    load_balancing_loss_func,
)
from transformers.utils import logging

from ...modeling_attn_mask_utils import (
    _gaudi_prepare_4d_causal_attention_mask,
)
from ...modeling_rope_utils import GaudiRotaryEmbedding


try:
    from habana_frameworks.torch.hpex.kernels import RotaryPosEmbeddingHelperV2 as FusedRoPE

    has_fused_rope = True
except ImportError:
    has_fused_rope = False
    print("Not using HPU fused kernel for apply_rotary_pos_emb")

try:
    from habana_frameworks.torch.hpex.normalization import FusedRMSNorm as FusedRMSNorm

    has_fused_rms_norm = True
except ImportError:
    has_fused_rms_norm = False
    print("Not using HPU fused kernel for RMSNorm")

try:
    from habana_frameworks.torch.hpex.kernels import FusedSDPA
except ImportError:
    print("Not using HPU fused scaled dot-product attention kernel.")
    FusedSDPA = None

logger = logging.get_logger(__name__)


def apply_customized_rope(q, k, cos, sin, position_ids):
    if q.device.type == "hpu" and has_fused_rope:
        # TODO: remove `.clone()` when it is fixed in SynapseAI
        if k.dtype == torch.bfloat16:
            return FusedRoPE.apply(
                q, cos.unsqueeze(0).unsqueeze(0).clone(), sin.unsqueeze(0).unsqueeze(0).clone(), position_ids
            ), FusedRoPE.apply(
                k,
                cos.unsqueeze(0).unsqueeze(0).clone().to(torch.bfloat16),
                sin.unsqueeze(0).unsqueeze(0).clone().to(torch.bfloat16),
                position_ids,
            )
        return FusedRoPE.apply(
            q, cos.unsqueeze(0).unsqueeze(0).clone(), sin.unsqueeze(0).unsqueeze(0).clone(), position_ids
        ), FusedRoPE.apply(
            k, cos.unsqueeze(0).unsqueeze(0).clone(), sin.unsqueeze(0).unsqueeze(0).clone(), position_ids
        )
    else:
        # keep the same implementation as Transformers v4.37.2
        return apply_rotary_pos_emb(q, k, cos[position_ids], sin[position_ids])


def gaudi_qwen2moe_rmsnorm_forward(self, hidden_states):
    """
    Copied from MixtralRMSNorm.forward: https://github.com/huggingface/transformers/blob/v4.37.0/src/transformers/models/mixtral/modeling_mixtral.py
    The only differences are:
        - override RMSNorm with Habana fused RMSNorm
    """
    if hidden_states.device.type == "hpu" and has_fused_rms_norm:
        # mixed dtypes are not good for FusedRMSNorm, both inputs need to have same dtype
        if hidden_states.dtype != self.weight.dtype:
            orig_dtype = hidden_states.dtype
            hidden_states = FusedRMSNorm.apply(hidden_states.to(self.weight.dtype), self.weight, self.variance_epsilon)
            return hidden_states.to(orig_dtype)
        else:
            hidden_states = FusedRMSNorm.apply(hidden_states, self.weight, self.variance_epsilon)
            return hidden_states
    else:
        input_dtype = hidden_states.dtype
        hidden_states = hidden_states.to(torch.float32)
        variance = hidden_states.pow(2).mean(-1, keepdim=True)
        hidden_states = hidden_states * torch.rsqrt(variance + self.variance_epsilon)
        return self.weight * hidden_states.to(input_dtype)


class GaudiQwen2MoeMLP(Qwen2MoeMLP):
    def pre_mlp_forward(self, x):
        input = self.act_fn(self.gate_proj(x)) * self.up_proj(x)
        output = self.down_proj(input)
        return output

    def mlp_all_reduce(self, x):
        if hasattr(self.down_proj, "all_reduce"):
            self.down_proj.all_reduce(x)

    def post_mlp_forward(self, x):
        if hasattr(self.down_proj, "post_all_reduce"):
            return self.down_proj.post_all_reduce(x)
        return x


def gaudi_qwen2moe_repeat_kv(
    query_states: torch.Tensor,
    key_states: torch.Tensor,
    value_states: torch.Tensor,
    attention_mask: torch.Tensor,
    n_rep: int,
):
    """
    Copied from repeat_kv: https://github.com/huggingface/transformers/blob/v4.37.0/src/transformers/models/mixtral/modeling_mixtral.py
    The only differences are:
    - Append num_key_value_heads == 1 check as kv states can be broadcasted during matmuls so need to expand and reshape them.
    - Add new args query_states, key_states, value_states and attention_mask and update the logic for expansion.
    The query states go from (batch, num_heads, seqlen, head_dim) to (batch, num_key_value_heads, n_rep, seqlen, head_dim)
    The key/value states go from (batch, num_key_value_heads, seqlen, head_dim) to (batch, num_key_value_heads, 1, seqlen, head_dim)
    """
    batch, num_key_value_heads, kv_len, head_dim = key_states.shape
    if n_rep == 1 or num_key_value_heads == 1:
        return query_states, key_states, value_states, attention_mask

    new_kv_shape = (batch, num_key_value_heads, 1, kv_len, head_dim)
    key_states = key_states.reshape(new_kv_shape)
    value_states = value_states.reshape(new_kv_shape)

    batch, q_heads, q_len, head_dim = query_states.shape
    new_q_shape = (batch, num_key_value_heads, n_rep, q_len, head_dim)
    query_states = query_states.reshape(new_q_shape)

    if attention_mask is not None:
        # Add groups dim and set to 1
        attention_mask = attention_mask.unsqueeze(1)

    return query_states, key_states, value_states, attention_mask


class Matmul(torch.nn.Module):
    def __init__(self):
        super().__init__()

    def forward(self, x, y):
        return torch.matmul(x, y)


class KVCache(torch.nn.Module):
    def __init__(self):
        super(KVCache, self).__init__()
        self.cache = None
        self.inp_seq_len = -1

    def allocate(self, inp_seq_len, dtype, device, shape):
        if self.cache is None or self.cache.shape != shape:
            self.inp_seq_len = inp_seq_len
            self.cache = torch.zeros(shape, dtype=dtype, device=device)
        else:
            assert self.inp_seq_len == inp_seq_len, (
                f"inp_seq_len must be the same. self.inp_seq_len:{self.inp_seq_len} inp_seq_len:{inp_seq_len}"
            )
            self.cache.fill_(0)

    @staticmethod
    def update(prev, cur, dim, idx, inp_seq_len):
        orig_cur = cur
        if prev.shape == cur.shape:
            prev.copy_(cur)
            return orig_cur
        if idx is not None and cur.shape[2] > 1 and cur.shape[2] <= prev.shape[2]:
            # Initialize
            prev[:, :, :inp_seq_len, :].copy_(cur)
            return orig_cur
        if idx is not None:
            prev.index_copy_(dim, idx - 1, cur)
            return prev
        else:
            return torch.cat((prev, cur), dim=dim)

    def get_shape(self):
        if self.cache is None:
            return None
        return self.cache.shape

    def forward(self, cur, dim, idx):
        return self.update(self.cache, cur, dim, idx, self.inp_seq_len)


#  FusedScaledDotProductAttention
class ModuleFusedSDPA(torch.nn.Module):
    def __init__(self, fusedSDPA, scale, attention_dropout, enable_recompute, flash_attention_fp8):
        super().__init__()
        self._hpu_kernel_fsdpa = fusedSDPA
        self.scale = scale
        self.attention_dropout = attention_dropout
        self.enable_recompute = enable_recompute
        self.flash_attention_fp8 = flash_attention_fp8

    def forward(
        self,
        query,
        key,
        value,
        attn_mask,
        dropout_p,
        is_casual,
        scale,
        softmax_mode,
        recompute_mode,
        valid_sequence_lengths,
        padding_side="left",
    ):
        return self._hpu_kernel_fsdpa.apply(
            query,
            key,
            value,
            attn_mask,
            dropout_p,
            is_casual,
            scale,
            softmax_mode,
            recompute_mode,
            valid_sequence_lengths,
            padding_side,
        )


class GaudiQwen2MoeAttention(Qwen2MoeAttention):
    def __init__(self, config: Qwen2MoeConfig, layer_idx: Optional[int] = None):
        super().__init__(config, layer_idx)

        self.matmul_qk = Matmul()
        self.matmul_av = Matmul()
        self.k_cache = KVCache()
        self.v_cache = KVCache()

        self.inp_seq_len = -1
        self.norm_factor = 1.0 / math.sqrt(self.head_dim)

        self.rotary_emb = GaudiRotaryEmbedding(config=self.config)

        self.fused_scaled_dot_product_attention = (
            ModuleFusedSDPA(
                FusedSDPA,
                scale=self.norm_factor,
                attention_dropout=self.attention_dropout,
                enable_recompute=False,
                flash_attention_fp8=getattr(config, "flash_attention_fp8", False),
            )
            if FusedSDPA
            else None
        )

    def allocate_kv_cache(self, batch_size, max_seq_len, inp_seq_len):
        cache_shape = (batch_size, self.num_key_value_heads, max_seq_len, self.head_dim)
        device = self.k_proj.weight.device
        dtype = self.config.torch_dtype
        self.k_cache.allocate(inp_seq_len, dtype, device, cache_shape)
        self.v_cache.allocate(inp_seq_len, dtype, device, cache_shape)

    def update_sincos_cache(self, seq_len):
        # Call rotary emb forward() to update cos/sin cache when infering more than self.max_position_embeddings
        # This helps in avoiding creation of these caches during actual model forward pass and
        # reduce memory consumption and improve performance.
        if seq_len > self.max_position_embeddings:
            self.max_position_embeddings = seq_len
            _, _ = self.rotary_emb(self.k_proj.weight, seq_len=seq_len)

    def reorder(self, tensor, beam_idx, dim_a, dim_b):
        updated = tensor.index_select(0, beam_idx)
        tensor.copy_(updated)

    def reorder_kv_cache(self, beam_idx: torch.LongTensor):
        if self.k_cache.cache is None:
            return (None, None)

        head_dim = self.k_cache.cache.size(-1)
        seq_length = self.k_cache.cache.size(-2)
        self.reorder(self.k_cache.cache, beam_idx, seq_length, head_dim)
        self.reorder(self.v_cache.cache, beam_idx, seq_length, head_dim)
        return (self.k_cache.cache.shape, self.v_cache.cache.shape)

    def pre_attn_forward(
        self,
        hidden_states: torch.Tensor,
        attention_mask: Optional[torch.Tensor] = None,
        position_ids: Optional[torch.LongTensor] = None,
        past_key_value: Optional[Cache] = None,
        output_attentions: bool = False,
        use_cache: bool = False,
        cache_position: Optional[torch.LongTensor] = None,
        position_embeddings: Optional[Tuple[torch.Tensor, torch.Tensor]] = None,  # necessary, but kept here for BC
        token_idx: Optional[torch.Tensor] = None,
        attn_softmax_bf16: Optional[bool] = False,
        reuse_cache: Optional[bool] = False,
        use_flash_attention: Optional[bool] = False,
        flash_attention_recompute: Optional[bool] = False,
        flash_attention_causal_mask: Optional[bool] = False,
        flash_attention_fast_softmax: Optional[bool] = False,
        valid_sequence_lengths: Optional[torch.Tensor] = None,
        cache_idx: int = None,
        num_virtual_tokens: int = None,
        **kwargs,
    ) -> Tuple[torch.Tensor, Optional[torch.Tensor], Optional[Tuple[torch.Tensor]]]:
        """
        Copied from LlamaAttention.forward: https://github.com/huggingface/transformers/blob/main/src/transformers/models/llama/modeling_llama.py
        The only differences are:
        - add new args token_idx
        - optimize KV cache
        - add new args attn_softmax_bf16
        - add new args reuse_cache
        - add new args use_flash_attention
        - add new arg flash_attention_recompute
        - add new arg flash_attention_causal_mask
        - add new arg flash_attention_fast_softmax
        - add new arg num_virtual_tokens
        """
        bsz, q_len, _ = hidden_states.size()

        query_states = self.q_proj(hidden_states)
        key_states = self.k_proj(hidden_states)
        value_states = self.v_proj(hidden_states)

        query_states = query_states.view(bsz, q_len, -1, self.head_dim).transpose(1, 2)
        key_states = key_states.view(bsz, q_len, -1, self.head_dim).transpose(1, 2)
        value_states = value_states.view(bsz, q_len, -1, self.head_dim).transpose(1, 2)

        kv_seq_len = key_states.shape[-2]
        if past_key_value is not None:
            if token_idx is None:
                if hasattr(past_key_value, "get_usable_length"):
                    kv_seq_len += past_key_value.get_usable_length(kv_seq_len, self.layer_idx)
                else:
                    kv_seq_len += past_key_value[0].shape[-2]
            else:
                if reuse_cache and not isinstance(past_key_value[0], torch.Tensor):
                    kv_seq_len = past_key_value[0][-2]
                else:
                    if num_virtual_tokens is not None and num_virtual_tokens == past_key_value[0].shape[-2]:
                        kv_seq_len = past_key_value[0].shape[-2] + kv_seq_len
                    else:
                        kv_seq_len = past_key_value[0].shape[-2]

        cos, sin = self.rotary_emb(value_states, seq_len=kv_seq_len)
        query_states, key_states = apply_customized_rope(query_states, key_states, cos, sin, position_ids)

        if use_cache:
            # reuse k, v, self_attention
            if reuse_cache:
                if past_key_value is not None and isinstance(past_key_value[0], torch.Tensor):
                    # prefix tuning case. attach past_key_value to generate first token.
                    key_states = torch.cat((past_key_value[0], key_states), -2)
                    value_states = torch.cat((past_key_value[1], value_states), -2)
                key_states = self.k_cache(key_states, 2, token_idx)
                value_states = self.v_cache(value_states, 2, token_idx)
                past_key_value = (self.k_cache.get_shape(), self.v_cache.get_shape())
            else:
                if past_key_value is None:
                    past_key = torch.zeros(key_states.shape, dtype=self.k_proj.weight.dtype, device=key_states.device)
                    past_value = torch.zeros(
                        key_states.shape, dtype=self.k_proj.weight.dtype, device=key_states.device
                    )
                    # Return list instead of tuple
                    past_key_value = [past_key, past_value]
                if (
                    token_idx is not None
                    and num_virtual_tokens is not None
                    and num_virtual_tokens == past_key_value[0].shape[-2]
                ):
                    # prefix tuning case. attach past_key_value to generate first token.
                    key_states = torch.cat((past_key_value[0], key_states), -2)
                    value_states = torch.cat((past_key_value[1], value_states), -2)
                    past_key_value = (key_states, value_states)
                else:
                    key_states = self.k_cache.update(past_key_value[0], key_states, 2, token_idx, self.inp_seq_len)
                    value_states = self.v_cache.update(past_key_value[1], value_states, 2, token_idx, self.inp_seq_len)

                if token_idx is None:
                    past_key_value = (key_states, value_states)

            if cache_idx is not None and q_len == 1:
                key_states = key_states[:, :, :cache_idx, :]
                value_states = value_states[:, :, :cache_idx, :]
                if attention_mask is not None:
                    attention_mask = attention_mask[:, :, :, :cache_idx]
                kv_seq_len = key_states.shape[-2]
        else:
            past_key_value = None

        if use_flash_attention and FusedSDPA is not None:
            if q_len == 1:
                # next token
                attn_output = self.fused_scaled_dot_product_attention(
                    query_states,
                    key_states,
                    value_states,
                    attention_mask,
                    0.0,
                    False,
                    None,
                    "None",
                    False,
                    None,
                    "None",
                )
            else:
                # first token
                softmax_mode = "fast" if flash_attention_fast_softmax else "None"
                if flash_attention_causal_mask:
                    attn_output = self.fused_scaled_dot_product_attention(
                        query_states,
                        key_states,
                        value_states,
                        None,
                        0.0,
                        True,
                        None,
                        softmax_mode,
                        flash_attention_recompute,
                        valid_sequence_lengths,
                        "left",
                    )
                else:
                    attn_output = self.fused_scaled_dot_product_attention(
                        query_states,
                        key_states,
                        value_states,
                        attention_mask,
                        0.0,
                        False,
                        None,
                        softmax_mode,
                        flash_attention_recompute,
                        None,
                        "None",
                    )

        else:
            query_states, key_states, value_states, attention_mask = gaudi_qwen2moe_repeat_kv(
                query_states, key_states, value_states, attention_mask, self.num_key_value_groups
            )

            query_states = query_states * self.norm_factor
            attn_weights = self.matmul_qk(query_states, key_states.transpose(-2, -1)).float()
            htcore.mark_step()

            if attention_mask is not None:  # no matter the length, we just slice it
                causal_mask = attention_mask
                if cache_position is not None:
                    causal_mask = attention_mask[:, :, cache_position, : key_states.shape[-2]]
                attn_weights = attn_weights + causal_mask.float()

            if attn_softmax_bf16:
                attn_weights = torch.nn.functional.softmax(attn_weights, dim=-1, dtype=query_states.dtype)
            else:
                # upcast attention to fp32
                attn_weights = torch.nn.functional.softmax(attn_weights, dim=-1, dtype=torch.float32).to(
                    query_states.dtype
                )
            attn_weights = torch.nn.functional.dropout(attn_weights, p=self.attention_dropout, training=self.training)
            attn_output = self.matmul_av(attn_weights, value_states)
            attn_output = attn_output.reshape(bsz, -1, q_len, self.head_dim)

        if attn_output.size() != (bsz, self.num_heads, q_len, self.head_dim):
            raise ValueError(
                f"`attn_output` should be of size {(bsz, self.num_heads, q_len, self.head_dim)}, but is"
                f" {attn_output.size()}"
            )

        attn_output = attn_output.transpose(1, 2).contiguous()

        attn_output = attn_output.reshape(bsz, q_len, -1)

        attn_output = self.o_proj(attn_output)

        if not output_attentions:
            attn_weights = None

        if not reuse_cache and token_idx is not None and cache_idx is not None and q_len == 1:
            # Return only past key value shapes and not the tensors during decode phase (q len is 1)
            # to avoid making past key values as persistent output tensors of HPU graphs.
            past_key_value = (past_key_value[0].shape, past_key_value[1].shape)

        return attn_output, attn_weights, past_key_value

    def attention_all_reduce(self, attn_output):
        if hasattr(self.o_proj, "all_reduce"):
            self.o_proj.all_reduce(attn_output)

    def post_attn_forward(self, attn_output):
        if hasattr(self.o_proj, "post_all_reduce"):
            return self.o_proj.post_all_reduce(attn_output)
        return attn_output


def gaudi_qwen2moe_block_sparse_moe_forward(self, hidden_states: torch.Tensor) -> Tuple[torch.Tensor, torch.Tensor]:
    """
    - optimize expert forward, remove dynamic control and dynamic shape
    """
    batch_size, sequence_length, hidden_dim = hidden_states.shape
    hidden_states = hidden_states.view(-1, hidden_dim)

    # router_logits: (batch * sequence_length, n_experts)
    router_logits = self.gate(hidden_states)

    routing_weights = F.softmax(router_logits, dim=1, dtype=torch.float)
    routing_weights, selected_experts = torch.topk(routing_weights, self.top_k, dim=-1)

    if self.norm_topk_prob:
        routing_weights /= routing_weights.sum(dim=-1, keepdim=True)
    # we cast back to the input dtype
    routing_weights = routing_weights.to(hidden_states.dtype)

    if self.training:
        final_hidden_states = torch.zeros(
            (batch_size, sequence_length, hidden_dim), dtype=hidden_states.dtype, device=hidden_states.device
        )

        padded_weights = torch.zeros(
            (batch_size * sequence_length, self.num_experts), dtype=hidden_states.dtype, device=hidden_states.device
        )
        padded_weights.scatter_(-1, selected_experts, routing_weights)
        padded_weights = padded_weights.reshape(-1, sequence_length, self.num_experts)
        padded_weights = padded_weights.permute(2, 0, 1).unsqueeze(-1)

        # Loop over all available experts in the model and perform the computation on each expert
        for expert_idx in range(self.num_experts):
            expert_layer = self.experts[expert_idx]
            padded_weight = padded_weights[expert_idx]
            current_state_static = hidden_states.reshape(-1, hidden_dim)
            current_hidden_states_static = (
                expert_layer.pre_mlp_forward(current_state_static).reshape(-1, sequence_length, hidden_dim)
                * padded_weight
            )
            final_hidden_states = final_hidden_states + current_hidden_states_static
    else:
        experts_range = range(self.num_experts)
        w1_list = [self.experts[i].gate_proj.weight.squeeze() for i in experts_range]
        w2_list = [self.experts[i].down_proj.weight.squeeze() for i in experts_range]
        w3_list = [self.experts[i].up_proj.weight.squeeze() for i in experts_range]

        final_hidden_states = torch.ops.hpu.mixture_of_experts(
            hidden_states=hidden_states,
            expert_routing_table=selected_experts,
            router_weights=routing_weights,
            w1=w1_list,
            w2=w3_list,  # Note that there is a different naming convention of w1, w2, and w3 between optimum habana's mixtral model and dynamic MoE kernel.
            w3=w2_list,
            permuted_weights=True,
            activation="silu",
            experts_min=0,
            experts_max=(self.num_experts - 1),
        )
        final_hidden_states = final_hidden_states.reshape(-1, sequence_length, hidden_dim)

        if is_deepspeed_available():
            from deepspeed import comm as dist

            if dist.is_initialized():
                dist.all_reduce(final_hidden_states, op=dist.ReduceOp.SUM)

    shared_expert_output = self.shared_expert(hidden_states)

    shared_expert_output = F.sigmoid(self.shared_expert_gate(hidden_states)) * shared_expert_output

    shared_expert_output = shared_expert_output.reshape(-1, sequence_length, hidden_dim)

    final_hidden_states = final_hidden_states + shared_expert_output

    return final_hidden_states, router_logits


class GaudiQwen2MoeDecoderLayer(Qwen2MoeDecoderLayer):
    def __init__(self, config: Qwen2MoeConfig, layer_idx: int):
        super(Qwen2MoeDecoderLayer, self).__init__()
        self.hidden_size = config.hidden_size

        self.self_attn = GaudiQwen2MoeAttention(config=config, layer_idx=layer_idx)

        if config.num_experts > 0 and (layer_idx + 1) % config.decoder_sparse_step == 0:
            self.mlp = Qwen2MoeSparseMoeBlock(config)
        else:
            self.mlp = GaudiQwen2MoeMLP(config, intermediate_size=config.intermediate_size)

        self.input_layernorm = Qwen2MoeRMSNorm(config.hidden_size, eps=config.rms_norm_eps)
        self.post_attention_layernorm = Qwen2MoeRMSNorm(config.hidden_size, eps=config.rms_norm_eps)

    def allocate_kv_cache(self, batch_size, max_seq_len, inp_seq_len):
        self.self_attn.allocate_kv_cache(batch_size, max_seq_len, inp_seq_len)

    def reorder_kv_cache(self, beam_idx: torch.LongTensor):
        return self.self_attn.reorder_kv_cache(beam_idx)

    def update_sincos_cache(self, seq_len):
        self.self_attn.update_sincos_cache(seq_len)

    def forward(
        self,
        hidden_states: torch.Tensor,
        attention_mask: Optional[torch.Tensor] = None,
        position_ids: Optional[torch.LongTensor] = None,
        past_key_value: Optional[Tuple[torch.Tensor]] = None,
        output_attentions: Optional[bool] = False,
        output_router_logits: Optional[bool] = False,
        use_cache: Optional[bool] = False,
        cache_position: Optional[torch.LongTensor] = None,
        position_embeddings: Optional[Tuple[torch.Tensor, torch.Tensor]] = None,  # necessary, but kept here for BC
        token_idx: Optional[torch.Tensor] = None,
        attn_softmax_bf16: Optional[bool] = False,
        reuse_cache: Optional[bool] = False,
        use_flash_attention: Optional[bool] = False,
        flash_attention_recompute: Optional[bool] = False,
        flash_attention_causal_mask: Optional[bool] = False,
        flash_attention_fast_softmax: Optional[bool] = False,
        valid_sequence_lengths: Optional[torch.Tensor] = None,
        cache_idx: int = None,
        num_virtual_tokens: int = None,
        **kwargs,
    ) -> Tuple[torch.FloatTensor, Optional[Tuple[torch.FloatTensor, torch.FloatTensor]]]:
        """
        Copied from LlamaDecoderLayer.forward: https://github.com/huggingface/transformers/blob/main/src/transformers/models/llama/modeling_llama.py
        The only differences are:
        - add new args token_idx
        - add new args attn_softmax_bf16
        - add new args reuse_cache
        - add new args use_flash_attention
        - add new arg flash_attention_recompute
        - add new arg flash_attention_causal_mask
        - add new arg flash_attention_fast_softmax
        """
        if "padding_mask" in kwargs:
            warnings.warn(
                "Passing `padding_mask` is deprecated and will be removed in v4.37. Please make sure use `attention_mask` instead.`"
            )
        residual = hidden_states
        hidden_states, self_attn_weights, present_key_value = self.pre_attn(
            hidden_states=hidden_states,
            attention_mask=attention_mask,
            position_ids=position_ids,
            past_key_value=past_key_value,
            output_attentions=output_attentions,
            use_cache=use_cache,
            cache_position=cache_position,
            token_idx=token_idx,
            attn_softmax_bf16=attn_softmax_bf16,
            reuse_cache=reuse_cache,
            use_flash_attention=use_flash_attention,
            flash_attention_recompute=flash_attention_recompute,
            flash_attention_causal_mask=flash_attention_causal_mask,
            flash_attention_fast_softmax=flash_attention_fast_softmax,
            valid_sequence_lengths=valid_sequence_lengths,
            cache_idx=cache_idx,
            num_virtual_tokens=num_virtual_tokens,
            **kwargs,
        )

        self.self_attn.attention_all_reduce(hidden_states)
        hidden_states, residual, router_logits = self.post_attn_pre_mlp(hidden_states, residual)

        hidden_states = self.post_mlp(hidden_states, residual)
        outputs = (hidden_states,)
        if output_attentions:
            outputs += (self_attn_weights,)
        if use_cache:
            outputs += (present_key_value,)
        if output_router_logits:
            outputs += (router_logits,)

        return outputs

    def pre_attn(
        self,
        hidden_states: torch.Tensor,
        attention_mask: Optional[torch.Tensor] = None,
        position_ids: Optional[torch.LongTensor] = None,
        past_key_value: Optional[Tuple[torch.Tensor]] = None,
        output_attentions: Optional[bool] = False,
        use_cache: Optional[bool] = False,
        cache_position: Optional[torch.LongTensor] = None,
        position_embeddings: Optional[Tuple[torch.Tensor, torch.Tensor]] = None,  # necessary, but kept here for BC
        token_idx: Optional[torch.Tensor] = None,
        attn_softmax_bf16: Optional[bool] = False,
        reuse_cache: Optional[bool] = False,
        use_flash_attention: Optional[bool] = False,
        flash_attention_recompute: Optional[bool] = False,
        flash_attention_causal_mask: Optional[bool] = False,
        flash_attention_fast_softmax: Optional[bool] = False,
        valid_sequence_lengths: Optional[torch.Tensor] = None,
        cache_idx: int = None,
        num_virtual_tokens: int = None,
        **kwargs,
    ) -> Tuple[torch.FloatTensor, Optional[Tuple[torch.FloatTensor, torch.FloatTensor]]]:
        hidden_states = self.input_layernorm(hidden_states)
        hidden_states, attn_weights, present_key_value = self.self_attn.pre_attn_forward(
            hidden_states=hidden_states,
            attention_mask=attention_mask,
            position_ids=position_ids,
            past_key_value=past_key_value,
            output_attentions=output_attentions,
            use_cache=use_cache,
            cache_position=cache_position,
            position_embeddings=position_embeddings,
            token_idx=token_idx,
            attn_softmax_bf16=attn_softmax_bf16,
            reuse_cache=reuse_cache,
            use_flash_attention=use_flash_attention,
            flash_attention_recompute=flash_attention_recompute,
            flash_attention_causal_mask=flash_attention_causal_mask,
            flash_attention_fast_softmax=flash_attention_fast_softmax,
            valid_sequence_lengths=valid_sequence_lengths,
            cache_idx=cache_idx,
            num_virtual_tokens=num_virtual_tokens,
            **kwargs,
        )
        return hidden_states, attn_weights, present_key_value

    def post_attn_pre_mlp(self, hidden_states, residual):
        hidden_states = self.self_attn.post_attn_forward(hidden_states)

        if self.training:
            hidden_states = hidden_states + residual
            residual = hidden_states
        else:
            residual.add_(hidden_states)
            hidden_states = residual

        hidden_states = self.post_attention_layernorm(hidden_states)

        hidden_states = self.mlp(hidden_states)

        if isinstance(hidden_states, tuple):
            hidden_states, router_logits = hidden_states
        else:
            router_logits = None

        return hidden_states, residual, router_logits

    def post_mlp(self, hidden_states, residual):
        if self.training:
            hidden_states = hidden_states + residual
        else:
            residual.add_(hidden_states)
            hidden_states = residual

        return hidden_states


class GaudiQwen2MoeModel(Qwen2MoeModel):
    def __init__(self, config: Qwen2MoeConfig):
        super(Qwen2MoeModel, self).__init__(config)
        self.padding_idx = config.pad_token_id
        self.vocab_size = config.vocab_size

        self.embed_tokens = torch.nn.Embedding(config.vocab_size, config.hidden_size, self.padding_idx)
        self.layers = torch.nn.ModuleList(
            [GaudiQwen2MoeDecoderLayer(config, layer_idx) for layer_idx in range(config.num_hidden_layers)]
        )
        self.norm = Qwen2MoeRMSNorm(config.hidden_size, eps=config.rms_norm_eps)
        self.gradient_checkpointing = False

        # Initialize weights and apply final processing
        self.post_init()

    def allocate_kv_cache(self, batch_size, max_seq_len, inp_seq_len):
        for layer in self.layers:
            layer.allocate_kv_cache(batch_size, max_seq_len, inp_seq_len)

    def reorder_kv_cache(self, beam_idx: torch.LongTensor):
        return tuple(layer.reorder_kv_cache(beam_idx) for layer in self.layers)

    def update_sincos_cache(self, seq_len):
        for layer in self.layers:
            layer.update_sincos_cache(seq_len)

    def forward(
        self,
        input_ids: Optional[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,
        use_cache: Optional[bool] = None,
        output_attentions: Optional[bool] = None,
        output_hidden_states: Optional[bool] = None,
        output_router_logits: Optional[bool] = None,
        cache_position: Optional[torch.LongTensor] = None,
        token_idx: Optional[torch.Tensor] = None,
        attn_softmax_bf16: Optional[bool] = False,
        reuse_cache: Optional[bool] = False,
        use_flash_attention: Optional[bool] = False,
        flash_attention_recompute: Optional[bool] = False,
        flash_attention_causal_mask: Optional[bool] = False,
        flash_attention_fast_softmax: Optional[bool] = False,
        valid_sequence_lengths: torch.Tensor = None,
        cache_idx: int = None,
        lazy_mode: Optional[bool] = True,
        num_virtual_tokens: int = None,
    ) -> MoeModelOutputWithPast:
        """
        Copied from LlamaModel.forward: https://github.com/huggingface/transformers/blob/main/src/transformers/models/llama/modeling_llama.py
        The only differences are:
        - add new args token_idx
        - add new args attn_softmax_bf16
        - add new args reuse_cache
        - add new args use_flash_attention
        - add new arg flash_attention_recompute
        - add new arg flash_attention_causal_mask
        - add new arg flash_attention_fast_softmax
        - add new arg lazy_mode
        """
        output_attentions = output_attentions if output_attentions is not None else self.config.output_attentions
        output_router_logits = (
            output_router_logits if output_router_logits is not None else self.config.output_router_logits
        )
        output_hidden_states = (
            output_hidden_states if output_hidden_states is not None else self.config.output_hidden_states
        )
        use_cache = use_cache if use_cache is not None else self.config.use_cache

        if (input_ids is None) ^ (inputs_embeds is not None):
            raise ValueError("You must specify exactly one of input_ids or inputs_embeds")
        elif input_ids is not None:
            batch_size, seq_length = input_ids.shape[:2]
        elif inputs_embeds is not None:
            batch_size, seq_length = inputs_embeds.shape[:2]
        else:
            raise ValueError("You have to specify either input_ids or inputs_embeds")

        if self.gradient_checkpointing and self.training and use_cache:
            logger.warning_once(
                "`use_cache=True` is incompatible with gradient checkpointing. Setting `use_cache=False`."
            )
            use_cache = False

        if inputs_embeds is None:
            inputs_embeds = self.embed_tokens(input_ids)

        ignore_cache_position = True  # Ignoring cache position for HPU
        use_new_cache = False  # Ignoring new Cache path for HPU

        past_seen_tokens = 0

        if past_key_values is not None and use_cache:  # kept for BC (cache positions)
            if reuse_cache:
                if isinstance(past_key_values[0][0], torch.Tensor):
                    past_seen_tokens = past_key_values[0][0].shape[2]
                else:
                    past_seen_tokens = past_key_values[0][0][2]
            else:
                if use_new_cache:
                    if not isinstance(past_key_values, StaticCache):
                        past_key_values = DynamicCache.from_legacy_cache(past_key_values)
                    past_seen_tokens = past_key_values.get_seq_length()
                else:
                    if past_key_values[0] is not None:  ##added for (None, None)
                        past_seen_tokens = past_key_values[0][0].shape[2]

        if ignore_cache_position is False:
            if cache_position is None:
                if isinstance(past_key_values, StaticCache):
                    raise ValueError("cache_position is a required argument when using StaticCache.")
                cache_position = torch.arange(
                    past_seen_tokens, past_seen_tokens + inputs_embeds.shape[1], device=inputs_embeds.device
                )
            if position_ids is None and cache_position:
                position_ids = cache_position.unsqueeze(0)

        else:
            if position_ids is None:
                position_ids = torch.arange(
                    past_seen_tokens, seq_length + past_seen_tokens, dtype=torch.long, device=inputs_embeds.device
                )
                position_ids = position_ids.unsqueeze(0)
            cache_position = None

        # HPU specific mask generation
        if ignore_cache_position:
            causal_mask = _gaudi_prepare_4d_causal_attention_mask(
                attention_mask,
                input_ids.shape if input_ids is not None else (batch_size, seq_length),
                inputs_embeds,
                past_seen_tokens,
            )
        else:
            causal_mask = self._update_causal_mask(attention_mask, inputs_embeds, cache_position, past_seen_tokens)

        # embed positions
        hidden_states = inputs_embeds

        # decoder layers
        all_hidden_states = () if output_hidden_states else None
        all_self_attns = () if output_attentions else None
        all_router_logits = () if output_router_logits else None
        next_decoder_cache = () if not use_new_cache else None

        if lazy_mode:
            htcore.mark_step()

        for layer_idx, decoder_layer in enumerate(self.layers):
            if output_hidden_states:
                all_hidden_states += (hidden_states,)

            if self.gradient_checkpointing and self.training:
                layer_outputs = self._gradient_checkpointing_func(
                    decoder_layer.__call__,
                    hidden_states,
                    causal_mask,
                    position_ids,
                    past_key_values,
                    output_attentions,
                    output_router_logits,
                    use_cache,
                    cache_position,
                    None,
                    attn_softmax_bf16,
                    False,
                    use_flash_attention,
                    flash_attention_recompute,
                    flash_attention_causal_mask,
                    flash_attention_fast_softmax,
                    valid_sequence_lengths,
                    None,
                )
            else:
                layer_outputs = decoder_layer(
                    hidden_states,
                    attention_mask=causal_mask,
                    position_ids=position_ids,
                    past_key_value=None if past_key_values is None else past_key_values[layer_idx],
                    output_attentions=output_attentions,
                    output_router_logits=output_router_logits,
                    use_cache=use_cache,
                    cache_position=cache_position,
                    token_idx=token_idx,
                    attn_softmax_bf16=attn_softmax_bf16,
                    reuse_cache=reuse_cache,
                    use_flash_attention=use_flash_attention,
                    flash_attention_recompute=flash_attention_recompute,
                    flash_attention_causal_mask=flash_attention_causal_mask,
                    flash_attention_fast_softmax=flash_attention_fast_softmax,
                    valid_sequence_lengths=valid_sequence_lengths,
                    cache_idx=cache_idx,
                    num_virtual_tokens=num_virtual_tokens,
                )

            hidden_states = layer_outputs[0]

            if use_cache:
                next_decoder_cache += (layer_outputs[2 if output_attentions else 1],)
            if output_attentions:
                all_self_attns += (layer_outputs[1],)

            if output_router_logits:
                all_router_logits += (layer_outputs[-1],)

        hidden_states = self.norm(hidden_states)

        # add hidden states from the last decoder layer
        if output_hidden_states:
            all_hidden_states += (hidden_states,)

        next_cache = None
        if use_cache:
            next_cache = (
                next_decoder_cache.to_legacy_cache() if isinstance(next_decoder_cache, Cache) else next_decoder_cache
            )

        return MoeModelOutputWithPast(
            last_hidden_state=hidden_states,
            past_key_values=next_cache,
            hidden_states=all_hidden_states,
            attentions=all_self_attns,
            router_logits=all_router_logits,
        )


class GaudiQwen2MoeForCausalLM(Qwen2MoeForCausalLM):
    """
    Inherits from Qwen2MoeForCausalLM: https://github.com/huggingface/transformers/blob/v4.37.0/src/transformers/models/mixtral/modeling_mixtral.py#L1231
    The only differences are:
    - add new args token_idx
    - add token_idx into model_inputs
    - from step2 when enable KV cache, slice next_input_ids from input_ids base on the token_idx
    - from step2 when enable KV cache, slice next_position_ids from position_ids base on the token_idx
    """

    def allocate_kv_cache(self, batch_size, max_seq_len, inp_seq_len):
        self.model.allocate_kv_cache(batch_size, max_seq_len, inp_seq_len)
        self.kv_cache_len = max_seq_len

    def reorder_kv_cache(self, beam_idx: torch.LongTensor):
        return self.model.reorder_kv_cache(beam_idx)

    def update_sincos_cache(self, seq_len):
        self.model.update_sincos_cache(seq_len)

    def forward(
        self,
        input_ids: Optional[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,
        output_router_logits: Optional[bool] = None,
        cache_position: Optional[torch.LongTensor] = None,
        logits_to_keep: Union[int, torch.Tensor] = 0,
        token_idx: Optional[torch.Tensor] = None,
        trim_logits: Optional[bool] = False,
        reuse_cache: Optional[bool] = None,
        attn_softmax_bf16: Optional[bool] = False,
        use_flash_attention: Optional[bool] = False,
        flash_attention_recompute: Optional[bool] = False,
        flash_attention_causal_mask: Optional[bool] = False,
        flash_attention_fast_softmax: Optional[bool] = False,
        valid_sequence_lengths: torch.Tensor = None,
        cache_idx: int = None,
        lazy_mode: Optional[bool] = True,
        num_virtual_tokens: int = None,
        **loss_kwargs,
    ) -> MoeCausalLMOutputWithPast:
        output_attentions = output_attentions if output_attentions is not None else self.config.output_attentions
        output_router_logits = (
            output_router_logits if output_router_logits is not None else self.config.output_router_logits
        )

        output_hidden_states = (
            output_hidden_states if output_hidden_states is not None else self.config.output_hidden_states
        )
        if self.generation_config.use_fused_rope is False:
            global has_fused_rope
            has_fused_rope = False

        # decoder outputs consists of (dec_features, layer_state, dec_hidden, dec_attn)
        outputs: MoeModelOutputWithPast = 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,
            output_router_logits=output_router_logits,
            cache_position=cache_position,
            token_idx=token_idx,
            attn_softmax_bf16=attn_softmax_bf16,
            reuse_cache=reuse_cache,
            use_flash_attention=use_flash_attention,
            flash_attention_recompute=flash_attention_recompute,
            flash_attention_causal_mask=flash_attention_causal_mask,
            flash_attention_fast_softmax=flash_attention_fast_softmax,
            valid_sequence_lengths=valid_sequence_lengths,
            cache_idx=cache_idx,
            lazy_mode=lazy_mode,
            num_virtual_tokens=num_virtual_tokens,
        )

        hidden_states = outputs.last_hidden_state
        _, seq_len, _ = hidden_states.shape
        if seq_len > 1 and trim_logits and not self.training:
            if token_idx is not None:
                hidden_states = hidden_states.index_select(1, token_idx - 1)
            else:
                hidden_states = hidden_states[:, -1, :]

        # Only compute necessary logits, and do not upcast them to float if we are not computing the loss
        slice_indices = slice(-logits_to_keep, None) if isinstance(logits_to_keep, int) else logits_to_keep
        logits = self.lm_head(hidden_states[:, slice_indices, :])

        loss = None
        if labels is not None:
            loss = self.loss_function(logits, labels, self.vocab_size, **loss_kwargs)

        aux_loss = None
        if output_router_logits:
            aux_loss = load_balancing_loss_func(
                outputs.router_logits,
                self.num_experts,
                self.num_experts_per_tok,
                attention_mask,
            )
            if labels is not None:
                loss += self.router_aux_loss_coef * aux_loss.to(loss.device)  # make sure to reside in the same device

        return MoeCausalLMOutputWithPast(
            loss=loss,
            aux_loss=aux_loss,
            logits=logits,
            past_key_values=outputs.past_key_values,
            hidden_states=outputs.hidden_states,
            attentions=outputs.attentions,
            router_logits=outputs.router_logits,
        )

    @staticmethod
    def _reorder_cache(
        past: Tuple[Tuple[torch.Tensor, torch.Tensor], ...], beam_idx: torch.LongTensor
    ) -> Tuple[Tuple[torch.Tensor, torch.Tensor], ...]:
        """
        This function is used to re-order the `past_key_values` cache if [`~PreTrainedModel.beam_search`] or
        [`~PreTrainedModel.beam_sample`] is called. This is required to match `past_key_values` with the correct
        beam_idx at every generation step.

        Output shares the same memory storage as `past`.
        """
        return tuple(
            (
                layer_past[0].index_select(0, beam_idx.to(layer_past[0].device)),
                layer_past[1].index_select(0, beam_idx.to(layer_past[1].device)),
            )
            for layer_past in past
        )

    def prepare_inputs_for_generation(
        self,
        input_ids,
        past_key_values=None,
        attention_mask=None,
        inputs_embeds=None,
        cache_position=None,
        position_ids=None,
        use_cache=True,
        num_logits_to_keep=None,
        token_idx=None,
        **kwargs,
    ):
        reuse_cache = kwargs.get("reuse_cache")
        bucket_internal = kwargs.get("bucket_internal")

        if past_key_values is not None:
            if token_idx is not None:
                idx = token_idx + kwargs.get("inputs_embeds_offset", 0) - 1
                input_ids = torch.index_select(input_ids, 1, idx)
            else:
                if inputs_embeds is not None:  # Exception 1
                    input_ids = input_ids[:, -cache_position.shape[0] :]
                elif (
                    input_ids.shape[1] != cache_position.shape[0]
                ):  # Default case (the "else", a no op, is Exception 2)
                    input_ids = input_ids[:, cache_position]
        elif (reuse_cache or bucket_internal) and token_idx is not None:
            # KV cache is pre allocated with reuse cache or will be padded with bucket internal
            # hence for the 1st token we can slice the inputs till token idx for the fwd pass.
            input_ids = input_ids[:, :token_idx]
            attention_mask = attention_mask[:, :token_idx]

        if attention_mask is not None and position_ids is None:
            # create position_ids on the fly for batch generation
            position_ids = attention_mask.long().cumsum(-1) - 1
            position_ids.masked_fill_(attention_mask == 0, 1)
            if past_key_values:
                if token_idx is not None:
                    position_ids = torch.index_select(position_ids, 1, token_idx - 1)
                else:
                    position_ids = position_ids[:, -input_ids.shape[1] :]

        # keep cache_position implementation as None for HPU
        cache_position = None

        # if `inputs_embeds` are passed, we only want to use them in the 1st generation step
        if inputs_embeds is not None and past_key_values is None:
            model_inputs = {"inputs_embeds": inputs_embeds}
        else:
            model_inputs = {"input_ids": input_ids.contiguous()}

        if num_logits_to_keep is not None:
            model_inputs["num_logits_to_keep"] = num_logits_to_keep

        model_inputs.update(
            {
                "position_ids": position_ids,
                "cache_position": cache_position,
                "past_key_values": past_key_values,
                "use_cache": use_cache,
                "attention_mask": attention_mask,
                "token_idx": token_idx,
                "trim_logits": kwargs.get("trim_logits"),
                "attn_softmax_bf16": kwargs.get("attn_softmax_bf16"),
                "reuse_cache": reuse_cache,
                "use_flash_attention": kwargs.get("use_flash_attention"),
                "flash_attention_recompute": kwargs.get("flash_attention_recompute"),
                "flash_attention_causal_mask": kwargs.get("flash_attention_causal_mask"),
                "flash_attention_fast_softmax": kwargs.get("flash_attention_fast_softmax"),
                "valid_sequence_lengths": kwargs.get("valid_sequence_lengths"),
                "cache_idx": kwargs.get("cache_idx"),
                "lazy_mode": kwargs.get("lazy_mode"),
                "num_virtual_tokens": kwargs.get("num_virtual_tokens"),
            }
        )
        return model_inputs