optimum/habana/transformers/models/stablelm/modeling_stablelm.py

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

import torch
from torch import nn
from transformers.cache_utils import Cache, DynamicCache
from transformers.modeling_outputs import BaseModelOutputWithPast, CausalLMOutputWithPast
from transformers.models.stablelm.configuration_stablelm import StableLmConfig
from transformers.models.stablelm.modeling_stablelm import (
    StableLmAttention,
    StableLmForCausalLM,
    StableLmMLP,
    apply_rotary_pos_emb,
    repeat_kv,
)
from transformers.utils import logging

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


logger = logging.get_logger(__name__)


class GaudiStableLmAttention(StableLmAttention):
    """Multi-headed attention from 'Attention Is All You Need' paper"""

    def __init__(self, config: StableLmConfig, layer_idx: Optional[int] = None):
        super().__init__(config, layer_idx)
        self.rotary_emb = GaudiRotaryEmbedding(config=self.config)

    def 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,
    ) -> Tuple[torch.Tensor, Optional[torch.Tensor], Optional[Tuple[torch.Tensor]]]:
        """
        Copied from StableLmAttention.forward: https://github.com/huggingface/transformers/blob/v4.38.2/src/transformers/models/stablelm/modeling_stablelm.py
        The only differences are:
        - add new args token_idx
        - optimize KV cache
        """
        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, self.num_heads, self.head_dim).transpose(1, 2)
        key_states = key_states.view(bsz, q_len, self.num_key_value_heads, self.head_dim).transpose(1, 2)
        value_states = value_states.view(bsz, q_len, self.num_key_value_heads, self.head_dim).transpose(1, 2)

        if self.qk_layernorm:
            query_states = self.q_layernorm(query_states)
            key_states = self.k_layernorm(key_states)

        kv_seq_len = key_states.shape[-2]
        if past_key_value is not None:
            if self.layer_idx is None:
                raise ValueError(
                    f"The cache structure has changed since version v4.36. If you are using {self.__class__.__name__} "
                    "for auto-regressive decoding with k/v caching, please make sure to initialize the attention class "
                    "with a layer index."
                )
            if token_idx is not None and past_key_value.get_usable_length(kv_seq_len, self.layer_idx) > 0:
                # When token_idx is used, static seq len = (input token len + max output token len)
                kv_seq_len = past_key_value.get_usable_length(kv_seq_len, self.layer_idx)
            else:
                kv_seq_len += past_key_value.get_usable_length(kv_seq_len, self.layer_idx)
        cos, sin = self.rotary_emb(value_states, seq_len=kv_seq_len)

        # Partial rotary embedding
        query_rot, query_pass = (
            query_states[..., : self.rotary_ndims],
            query_states[..., self.rotary_ndims :],
        )
        key_rot, key_pass = (
            key_states[..., : self.rotary_ndims],
            key_states[..., self.rotary_ndims :],
        )
        # [batch_size, seq_length, num_heads, head_dim // config.partial_rotary_factor]
        query_rot, key_rot = apply_rotary_pos_emb(query_rot, key_rot, cos[position_ids], sin[position_ids])

        # [batch_size, seq_length, num_heads, head_dim]
        query_states = torch.cat((query_rot, query_pass), dim=-1)
        key_states = torch.cat((key_rot, key_pass), dim=-1)

        if past_key_value is not None:
            if token_idx is not None:
                if 0 <= self.layer_idx < len(past_key_value.key_cache):
                    past_key_value.key_cache[self.layer_idx].index_copy_(2, token_idx - 1, key_states)
                    past_key_value.value_cache[self.layer_idx].index_copy_(2, token_idx - 1, value_states)
                    key_states = past_key_value.key_cache[self.layer_idx]
                    value_states = past_key_value.value_cache[self.layer_idx]
                else:
                    past_key_value.key_cache.append(key_states)
                    past_key_value.value_cache.append(value_states)
            else:
                # Specific to RoPE models with partial rotation
                cache_kwargs = {
                    "sin": sin,
                    "cos": cos,
                    "partial_rotation_size": self.rotary_ndims,
                    "cache_position": cache_position,
                }
                key_states, value_states = past_key_value.update(
                    key_states, value_states, self.layer_idx, cache_kwargs
                )

        # Repeat k/v heads if n_kv_heads < n_heads
        key_states = repeat_kv(key_states, self.num_key_value_groups)
        value_states = repeat_kv(value_states, self.num_key_value_groups)

        attn_weights = torch.matmul(query_states, key_states.transpose(2, 3)) / math.sqrt(self.head_dim)

        if attn_weights.size() != (bsz, self.num_heads, q_len, kv_seq_len):
            raise ValueError(
                f"Attention weights should be of size {(bsz, self.num_heads, q_len, kv_seq_len)}, but is"
                f" {attn_weights.size()}"
            )

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

        # upcast attention to fp32
        attn_weights = nn.functional.softmax(attn_weights, dtype=torch.float32, dim=-1).to(query_states.dtype)
        attn_weights = self.attention_dropout(attn_weights)

        attn_output = torch.matmul(attn_weights, value_states)

        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, self.hidden_size)

        attn_output = self.o_proj(attn_output)

        if not output_attentions:
            attn_weights = None

        return attn_output, attn_weights, past_key_value


class GaudiStableLmDecoderLayer(torch.nn.Module):
    def __init__(self, config: StableLmConfig, layer_idx: int):
        super().__init__()
        self.use_parallel_residual = config.use_parallel_residual
        self.hidden_size = config.hidden_size
        self.self_attn = GaudiStableLmAttention(config, layer_idx=layer_idx)
        self.mlp = StableLmMLP(config)
        self.input_layernorm = torch.nn.LayerNorm(config.hidden_size, eps=config.layer_norm_eps)
        self.post_attention_layernorm = None
        if not self.use_parallel_residual:
            self.post_attention_layernorm = torch.nn.LayerNorm(config.hidden_size, eps=config.layer_norm_eps)
        self.dropout = torch.nn.Dropout(config.hidden_dropout)

    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,
        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,
    ) -> Tuple[torch.FloatTensor, Optional[Tuple[torch.FloatTensor, torch.FloatTensor]]]:
        """
        Copied from StableLmDecoderLayer.forward: https://github.com/huggingface/transformers/blob/v4.38.2/src/transformers/models/stablelm/modeling_stablelm.py
        The only differences are:
        - add new args token_idx
        """
        residual = hidden_states

        hidden_states = self.input_layernorm(hidden_states)

        # Self Attention
        self_attn_output, self_attn_weights, present_key_value = self.self_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,
        )

        # copied from transformers.models.gpt_neox.modeling_gpt_neox.GPTNeoXLayer.forward
        if self.use_parallel_residual:
            # x = x + attn(ln1(x)) + mlp(ln1(x))
            # Fully Connected
            mlp_output = self.mlp(hidden_states)
            mlp_output = self.dropout(mlp_output)
            hidden_states = residual + self_attn_output + mlp_output
        else:
            # x = x + attn(ln1(x))
            # x = x + mlp(ln2(x))
            residual = residual + self_attn_output
            # Fully Connected
            mlp_output = self.mlp(self.post_attention_layernorm(residual))
            mlp_output = self.dropout(mlp_output)
            hidden_states = residual + mlp_output

        outputs = (hidden_states,)

        if output_attentions:
            outputs += (self_attn_weights,)

        if use_cache:
            outputs += (present_key_value,)

        return outputs


def gaudi_stablelm_model_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,
    cache_position: Optional[torch.LongTensor] = None,
    token_idx: Optional[torch.Tensor] = None,
) -> BaseModelOutputWithPast:
    """
    Copied from StableLmModel.forward: https://github.com/huggingface/transformers/blob/v4.38.2/src/transformers/models/stablelm/modeling_stablelm.py
    The only differences are:
    - add new args token_idx
    - replace _prepare_4d_causal_attention_mask with _gaudi_prepare_4d_causal_attention_mask
    """
    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
    )
    use_cache = use_cache if use_cache is not None else self.config.use_cache

    # retrieve input_ids and inputs_embeds
    if input_ids is not None and 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
    elif inputs_embeds is not None:
        batch_size, seq_length, _ = inputs_embeds.shape
    else:
        raise ValueError("You have to specify either decoder_input_ids or decoder_inputs_embeds")

    seq_length_with_past = seq_length
    past_key_values_length = 0

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

    if use_cache:
        use_legacy_cache = not isinstance(past_key_values, Cache)
        if use_legacy_cache:
            past_key_values = DynamicCache.from_legacy_cache(past_key_values)
        if token_idx is None:
            past_key_values_length = past_key_values.get_usable_length(seq_length)
            seq_length_with_past = seq_length_with_past + past_key_values_length

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

    if cache_position is None:
        past_seen_tokens = past_key_values.get_seq_length() if past_key_values is not None else 0
        cache_position = torch.arange(
            past_seen_tokens, past_seen_tokens + inputs_embeds.shape[1], device=inputs_embeds.device
        )
    if position_ids is None:
        position_ids = cache_position.unsqueeze(0)

    if self._attn_implementation == "flash_attention_2":
        # 2d mask is passed through the layers
        attention_mask = attention_mask if (attention_mask is not None and 0 in attention_mask) else None
    else:
        # 4d mask is passed through the layers
        attention_mask = _gaudi_prepare_4d_causal_attention_mask(
            attention_mask, (batch_size, seq_length), inputs_embeds, past_key_values_length
        )

    hidden_states = inputs_embeds

    # decoder layers
    all_hidden_states = () if output_hidden_states else None
    all_self_attns = () if output_attentions else None
    next_decoder_cache = None

    for decoder_layer in 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,
                attention_mask,
                position_ids,
                past_key_values,
                output_attentions,
                use_cache,
                cache_position,
                None,
            )
        else:
            layer_outputs = decoder_layer(
                hidden_states,
                attention_mask=attention_mask,
                position_ids=position_ids,
                past_key_value=past_key_values,
                output_attentions=output_attentions,
                use_cache=use_cache,
                cache_position=cache_position,
                token_idx=token_idx,
            )

        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],)

    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 use_legacy_cache else next_decoder_cache

    return BaseModelOutputWithPast(
        last_hidden_state=hidden_states,
        past_key_values=next_cache,
        hidden_states=all_hidden_states,
        attentions=all_self_attns,
    )


class GaudiStableLmForCausalLM(StableLmForCausalLM):
    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,
        cache_position: Optional[torch.LongTensor] = None,
        logits_to_keep: Union[int, torch.Tensor] = 0,
        token_idx: Optional[torch.Tensor] = None,
        **kwargs,
    ) -> CausalLMOutputWithPast:
        """
        Inherits from StableLmForCausalLM: https://github.com/huggingface/transformers/blob/v4.38.2/src/transformers/models/stablelm/modeling_stablelm.py
        The only differences are:
        - add new args token_idx
        """
        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
        )

        outputs: BaseModelOutputWithPast = 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,
            cache_position=cache_position,
            token_idx=token_idx,
        )

        hidden_states = outputs.last_hidden_state
        # No upscaling to float was ever done for StableLm
        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,
                vocab_size=self.config.vocab_size,
                **kwargs,
            )

        return CausalLMOutputWithPast(
            loss=loss,
            logits=logits,
            past_key_values=outputs.past_key_values,
            hidden_states=outputs.hidden_states,
            attentions=outputs.attentions,
        )

    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,
        **kwargs,
    ):
        """
        Inherits from StableLmForCausalLM: https://github.com/huggingface/transformers/blob/v4.38.2/src/transformers/models/stablelm/modeling_stablelm.py
        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
        """
        token_idx = kwargs.get("token_idx", None)
        if past_key_values is not None:
            if token_idx is None:
                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]
            else:
                idx = token_idx + kwargs.get("inputs_embeds_offset", 0) - 1
                input_ids = torch.index_select(input_ids, 1, 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] :]
                # This `clone` call is needed to avoid recapturing cuda graphs with `torch.compile`'s  `mode="reduce-overhead`, as otherwise the input `position_ids` would have various stride during the decoding. Here, simply using `.contiguous()` is not sufficient as in the batch size = 1 case, `position_ids` is already contiguous but with varying stride which retriggers a capture.
                position_ids = position_ids.clone(memory_format=torch.contiguous_format)

        # 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.clone(memory_format=torch.contiguous_format)
            }  # `contiguous()` needed for compilation use cases

        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,
            }
        )
        return model_inputs