modeling_bloom.py

# coding=utf-8
###############################################################################
# Copyright (C) 2022-2023 Habana Labs, Ltd. an Intel Company
###############################################################################
# Changes:
# - added import os
# - overrided __package__ to allow loading from different dir
# - switched GELU impl to nn.GELU
# - added token_idx to prepare_inputs_for_generation, BloomModel, BloomBlock, BloomAttention
# - added alibi tensor code adopted from Megatron transformer.py
# - added 'update' function to update KV-cache when static shapes are enabled
# - updated cache conversion methods to use n_head from config
# - added option to preallocate and reuse kv-cache
# - transposed 'key' in kv-cache to use the same shape as 'value'
# - extracted parts of BloomAttention.forward to sub_forward to reduce number of intermediate tensors
#
# Copyright 2022 HuggingFace Inc. team and BigScience workshop.
#
# 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 BLOOM model."""

import os
import math
import warnings
from typing import Optional, Tuple, Union, Callable

__package__ = 'transformers.models.bloom'

import torch
import torch.utils.checkpoint
from torch import nn
from torch.nn import BCEWithLogitsLoss, CrossEntropyLoss, LayerNorm, MSELoss
from torch.nn import functional as F

from ...file_utils import add_code_sample_docstrings, add_start_docstrings, add_start_docstrings_to_model_forward
from ...modeling_outputs import (
    BaseModelOutputWithPastAndCrossAttentions,
    CausalLMOutputWithCrossAttentions,
    QuestionAnsweringModelOutput,
    SequenceClassifierOutputWithPast,
    TokenClassifierOutput,
)
from ...modeling_utils import PreTrainedModel
from ...utils import logging
from .configuration_bloom import BloomConfig


logger = logging.get_logger(__name__)

_CHECKPOINT_FOR_DOC = "bigscience/bloom-560m"
_CONFIG_FOR_DOC = "BloomConfig"

BLOOM_PRETRAINED_MODEL_ARCHIVE_LIST = [
    "bigscience/bigscience-small-testing",
    "bigscience/bloom-560m",
    "bigscience/bloom-1b1",
    "bigscience/bloom-1b7",
    "bigscience/bloom-3b",
    "bigscience/bloom-7b1",
    "bigscience/bloom",
]

def pre_all_reduce(layer, input):
    if layer.__class__ is nn.Linear:
        return input
    output = torch.matmul(input, layer.weight.transpose(-1, -2))
    return output

def all_reduce(layer, input):
    if layer.__class__ is nn.Linear:
        return input
    else:
        if layer.mp_group is not None:
            from deepspeed import comm as dist
            dist.all_reduce(input, group=layer.mp_group)
        return input

def post_all_reduce(layer, input):
    if layer.__class__ is nn.Linear:
         return layer(input)
    if layer.bias is not None:
        output = input + layer.bias
    return output

try:
    in_place_interleave_hpu = torch.ops.hpu.in_place_interleave_
except AttributeError:
    in_place_interleave_hpu = None

try:
    kv_reorder_hpu = torch.ops.hpu.kv_reorder_
except AttributeError:
    kv_reorder_hpu = None

def _make_causal_mask(
    input_ids_shape: torch.Size, device: torch.device, past_key_values_length: int
) -> torch.BoolTensor:
    """
    Make causal mask used for self-attention.
    """
    batch_size, target_length = input_ids_shape
    mask = torch.empty((target_length, target_length + past_key_values_length), dtype=torch.bool, device=device)
    # ONNX doesn't support `torch.Tensor.triu` properly, thus we use this workaround
    seq_ids = torch.arange(target_length, device=device)
    mask[:, past_key_values_length:] = seq_ids[:, None] < seq_ids[None, :]

    if past_key_values_length > 0:
        mask[:, :past_key_values_length] = False

    expanded_mask = mask[None, None, :, :].expand(batch_size, 1, target_length, target_length + past_key_values_length)
    return expanded_mask


def _expand_mask(mask: torch.Tensor, tgt_length: int) -> torch.BoolTensor:
    """
    Expands attention_mask from `[batch_size, src_length]` to `[batch_size, 1, tgt_length, src_length]`.
    """
    batch_size, src_length = mask.shape
    tgt_length = tgt_length if tgt_length is not None else src_length

    expanded_mask = ~(mask[:, None, None, :].to(torch.bool))
    return expanded_mask.expand(batch_size, 1, tgt_length, src_length)


def build_alibi_tensor(attention_mask: torch.Tensor, num_heads: int, dtype: torch.dtype) -> torch.Tensor:
    """
    Link to paper: https://arxiv.org/abs/2108.12409 Alibi tensor is not causal as the original paper mentions, it
    relies on a translation invariance of softmax for quick implementation: with l being a tensor, and a fixed value
    `softmax(l+a) = softmax(l)`. Based on
    https://github.com/ofirpress/attention_with_linear_biases/blob/a35aaca144e0eb6b789dfcb46784c4b8e31b7983/fairseq/models/transformer.py#L742
    TODO @thomasw21 this doesn't work as nicely due to the masking strategy, and so masking varies slightly.

    Args:
    Returns tensor shaped (batch_size * num_heads, 1, max_seq_len)
        attention_mask (`torch.Tensor`):
            Token-wise attention mask, this should be of shape (batch_size, max_seq_len).
        num_heads (`int`, *required*):
            number of heads
        dtype (`torch.dtype`, *optional*, default=`torch.bfloat16`):
            dtype of the output tensor
    """
    batch_size, seq_length = attention_mask.shape
    closest_power_of_2 = 2 ** math.floor(math.log2(num_heads))
    base = torch.tensor(
        2 ** (-(2 ** -(math.log2(closest_power_of_2) - 3))), device=attention_mask.device, dtype=torch.float32
    )
    powers = torch.arange(1, 1 + closest_power_of_2, device=attention_mask.device, dtype=torch.int32)
    slopes = torch.pow(base, powers)

    if closest_power_of_2 != num_heads:
        extra_base = torch.tensor(
            2 ** (-(2 ** -(math.log2(2 * closest_power_of_2) - 3))), device=attention_mask.device, dtype=torch.float32
        )
        num_remaining_heads = min(closest_power_of_2, num_heads - closest_power_of_2)
        extra_powers = torch.arange(1, 1 + 2 * num_remaining_heads, 2, device=attention_mask.device, dtype=torch.int32)
        slopes = torch.cat([slopes, torch.pow(extra_base, extra_powers)], dim=0)

    # Note: alibi will added to the attention bias that will be applied to the query, key product of attention
    # => therefore alibi will have to be of shape (batch_size, num_heads, query_length, key_length)
    # => here we set (batch_size=1, num_heads=num_heads, query_length=1, key_length=max_length)
    # => the query_length dimension will then be broadcasted correctly
    # This is more or less identical to T5's relative position bias:
    # https://github.com/huggingface/transformers/blob/f681437203baa7671de3174b0fa583c349d9d5e1/src/transformers/models/t5/modeling_t5.py#L527
    # code taken from Megatron transformer.py

    batch_size = attention_mask.size()[0]
    max_seq_len = attention_mask.size()[1]
    alibi = slopes.unsqueeze(1).unsqueeze(1) * torch.arange(max_seq_len, device=attention_mask.device).unsqueeze(0).unsqueeze(0).expand(
        num_heads, -1, -1)

    #Select the part of the tensor that corresponds to our tensor parallel index.
    tp_world_size = int(os.environ.get('WORLD_SIZE', 1))
    tp_index = int(os.environ.get('RANK', 0))
    alibi = alibi.reshape((tp_world_size, -1, *alibi.shape[1:]))[tp_index]

    alibi = alibi.repeat(batch_size, 1, 1)
    return alibi.to(dtype)


def dropout_add(x: torch.Tensor, residual: torch.Tensor, prob: float, training: bool) -> torch.Tensor:
    """
    Dropout add function

    Args:
        x (`torch.tensor`, *required*):
            input tensor
        residual (`torch.tensor`, *required*):
            esidual tensor
        prob (`float`, *required*):
            dropout probability
        training (`bool`, *required*):
            training mode
    """
    out = F.dropout(x, p=prob, training=training)
    out = residual + out
    return out


def bloom_gelu_forward(x: torch.Tensor) -> torch.Tensor:
    """
    Custom bias GELU function. Adapted from Megatron-DeepSpeed code. Here we use a simple implementation (inference) to
    make the model jitable.

    Args:
        x (`torch.tensor`, *required*):
            input hidden states
    """
    return x * 0.5 * (1.0 + torch.tanh(0.79788456 * x * (1 + 0.044715 * x * x)))


def bloom_gelu_back(g: torch.Tensor, x: torch.Tensor) -> torch.Tensor:
    """
    gradient of tanh approximation of gelu gradient of actual gelu is: 0.5 * (1. + torch.erf(x * 0.70710678)) +
    0.3989423 * x * torch.exp(-0.5 * x * x)

    Args:
        g (`torch.tensor`, *required*):
            gradient output tensor
        x (`torch.tensor`, *required*):
            input tensor
    """
    x = x[0]  # x is a tuple of 1 element, needs to unpack it first
    tanh_out = torch.tanh(0.79788456 * x * (1 + 0.044715 * x * x))
    # sqrt(2/pi) * 3 * 0.044715 -> 0.1070322243
    ff = 0.5 * x * ((1 - tanh_out * tanh_out) * (0.79788456 + 0.1070322243 * x * x)) + 0.5 * (1 + tanh_out)
    return ff * g


class GeLUFunction(torch.autograd.Function):
    @staticmethod
    def forward(ctx, input: torch.Tensor) -> torch.Tensor:
        ctx.save_for_backward(input)
        return bloom_gelu_forward(input)

    @staticmethod
    def backward(ctx, grad_output: torch.Tensor) -> torch.Tensor:
        input = ctx.saved_tensors
        tmp = bloom_gelu_back(grad_output, input)
        return tmp


class BloomGelu(nn.Module):
    """
    BloomBiasGelu wrapper function that make use of the simple function on inference mode to make the model
    torchscriptable and use the autograd function in training mode to get the accurate results of the gradients Partly
    copied from Megatron-DeepSpeed code and adapted for our needs

    See here why autograd functions are not torchscriptable: https://github.com/pytorch/pytorch/issues/22329
    """

    def __init__(self):
        super().__init__()

    def forward(self, x: torch.Tensor) -> torch.Tensor:
        if self.training:
            return GeLUFunction.apply(x)
        else:
            return bloom_gelu_forward(x)


class CacheReorder(object):
    def __init__(self):
        super().__init__()
        self.input_length = 0
        self.num_beams = 0
        self.device = torch.device('cpu')

    def prepare_for_new_input(self, input_length, num_beams, batch_size, device):
        self.num_beams = num_beams
        self.device = device
        if device.type == "hpu" and self.input_length != input_length:
            self.start = torch.full([batch_size], input_length, dtype=torch.int32, device=device)
            self.end = torch.full([batch_size], input_length, dtype=torch.int32, device=device)
            self.mul = torch.tensor([[64, 16, 4, 1]], dtype=torch.int32, device=device)
            self.input_length = input_length

    def prepare_next_beam_idx(self, beam_idx):
        if self.device.type != "hpu" or kv_reorder_hpu is None or self.num_beams != 4:
            return beam_idx
        else:
            self.end.add_(1)
            indices = beam_idx.view(-1, self.num_beams)
            indices = torch.sum(indices * self.mul, axis=-1).to(torch.uint8)
            return indices

    def reorder_beams_first_token(self, tensor, beam_idx):
        if self.device.type != "hpu" or in_place_interleave_hpu is None or self.num_beams != 4:
            updated = tensor.index_select(0, beam_idx)
            tensor.copy_(updated)
        else:
            in_place_interleave_hpu(tensor)

    def reorder_beams_next_token(self, tensor, beam_idx):
        if self.device.type != "hpu" or kv_reorder_hpu is None or self.num_beams != 4:
            updated = tensor.index_select(0, beam_idx)
            tensor.copy_(updated)
        else:
            kv_reorder_hpu(tensor, self.start, self.end, beam_idx)


class CacheUpdate(nn.Module):
    def __init__(self):
        super().__init__()

    def forward(self, prev, cur, dim, idx):
        orig_cur = cur
        cur = cur.to(dtype=prev.dtype)
        if prev.shape[0] != cur.shape[0]:
            assert prev.shape[0] % cur.shape[0] == 0, f'Cannot update kv-cache. BatchSize changed! {prev.shape[0]} vs {cur.shape[0]}'
            # Repeat to accomodate bs/beam changes
            cur = cur.repeat(prev.shape[0] // cur.shape[0], 1, 1)
        if prev.shape[1] != cur.shape[1] and cur.shape[1] != 1:
            # Pad to accomodate input bucketing
            padding_len = prev.shape[1] - cur.shape[1]
            cur = torch.nn.functional.pad(cur, (0, 0, 0, padding_len))
        if prev.shape == cur.shape:
            # Initialize
            prev.copy_(cur)
            return orig_cur
        if os.environ.get('SKIP_KV_CACHE_UPDATE', '0') != '0':
            # Skip update
            return prev
        assert cur.shape[1] == 1, f'Cannot update kv-cache. Unsupported shapes. prev:{prev.shape} cur:{cur.shape}'
        if idx is not None:
            prev.index_copy_(dim, idx - 1, cur)
            prev_cast = prev.to(orig_cur.dtype)
            return prev_cast
        else:
            return torch.cat((prev, cur), dim=dim)

class BloomAttention(nn.Module):
    def __init__(self, config: BloomConfig):
        super().__init__()

        self.pretraining_tp = config.pretraining_tp
        self.slow_but_exact = config.slow_but_exact

        self.hidden_size = config.hidden_size
        self.num_heads = config.n_head
        self.head_dim = self.hidden_size // self.num_heads
        self.split_size = self.hidden_size
        self.hidden_dropout = config.hidden_dropout
        self.past_key = None
        self.past_value = None
        self.kv_cache_fp8 = False
        self.v_update = CacheUpdate()
        self.k_update = CacheUpdate()
        self.cache_reorder = CacheReorder()

        if self.head_dim * self.num_heads != self.hidden_size:
            raise ValueError(
                f"`hidden_size` must be divisible by num_heads (got `hidden_size`: {self.hidden_size} and `num_heads`:"
                f" {self.num_heads})."
            )

        # Layer-wise attention scaling
        self.inv_norm_factor = 1.0 / math.sqrt(self.head_dim)
        self.beta = 1.0

        self.query_key_value = nn.Linear(self.hidden_size, 3 * self.hidden_size, bias=True)
        self.dense = nn.Linear(self.hidden_size, self.hidden_size)
        self.attention_dropout = nn.Dropout(config.attention_dropout)

    def _split_heads(self, fused_qkv: torch.Tensor) -> Tuple[torch.Tensor, torch.Tensor, torch.Tensor]:
        """
        Split the last dimension into (num_heads, head_dim) without making any copies, results share same memory
        storage as `fused_qkv`

        Args:
            fused_qkv (`torch.tensor`, *required*): [batch_size, seq_length, num_heads * 3 * head_dim]

        Returns:
            query: [batch_size, seq_length, num_heads, head_dim] key: [batch_size, seq_length, num_heads, head_dim]
            value: [batch_size, seq_length, num_heads, head_dim]
        """
        batch_size, seq_length, three_times_hidden_size = fused_qkv.shape
        fused_qkv = fused_qkv.view(batch_size, seq_length, self.num_heads, 3, self.head_dim)
        return fused_qkv[..., 0, :], fused_qkv[..., 1, :], fused_qkv[..., 2, :]

    def _merge_heads(self, x: torch.Tensor) -> torch.Tensor:
        """
        Merge heads together over the last dimenstion

        Args:
            x: (`torch.tensor`, *required*): [batch_size * num_heads, seq_length, head_dim]

        Returns:
            torch.tensor: [batch_size, seq_length, num_heads * head_dim]
        """
        # What we want to achieve is:
        # batch_size * num_heads, seq_length, head_dim -> batch_size, seq_length, num_heads * head_dim
        batch_size_and_num_heads, seq_length, _ = x.shape
        batch_size = batch_size_and_num_heads // self.num_heads

        # First view to decompose the batch size
        # batch_size * num_heads, seq_length, head_dim -> batch_size, num_heads, seq_length, head_dim
        x = x.view(batch_size, self.num_heads, seq_length, self.head_dim)

        # batch_size, num_heads, seq_length, head_dim -> batch_size, seq_length, num_heads, head_dim
        x = x.permute(0, 2, 1, 3)

        # batch_size, seq_length, num_heads, head_dim -> batch_size, seq_length, num_heads * head_dim
        return x.reshape(batch_size, seq_length, self.num_heads * self.head_dim)

    def allocate_kv_cache(self, batch_size, seq_len, kv_cache_fp8):
        key_shape = (batch_size * self.num_heads, seq_len, self.head_dim)
        value_shape = (batch_size * self.num_heads, seq_len, self.head_dim)
        if self.past_key is None or self.past_key.shape != key_shape:
            device = self.query_key_value.weight.device
            dtype = self.query_key_value.weight.dtype
            if kv_cache_fp8:
                self.kv_cache_fp8 = True
                dtype = torch.float8_e4m3fn
            self.past_key = torch.empty(key_shape, dtype=dtype, device=device)
            self.past_value = torch.empty(value_shape, dtype=dtype, device=device)

    def prepare_next_beam_idx(self, beam_idx: torch.LongTensor):
        return self.cache_reorder.prepare_next_beam_idx(beam_idx)

    def prepare_for_new_input(self, input_length: int, num_beams: int, batch_size: int, device: torch.device):
        self.cache_reorder.prepare_for_new_input(input_length, num_beams, batch_size, device)

    def reorder(self, tensor: torch.LongTensor, beam_idx: torch.LongTensor, reorder_fn : Callable):
        cache_4D = tensor.view(-1, self.num_heads, tensor.size(-2), tensor.size(-1))
        return reorder_fn(cache_4D, beam_idx)

    def reorder_kv_cache(self, beam_idx: torch.LongTensor, reorder_fn : Callable):
        if self.past_key is None:
            return (None, None)
        else:
            self.reorder(self.past_key, beam_idx, reorder_fn)
            self.reorder(self.past_value, beam_idx, reorder_fn)
            return (self.past_key.shape, self.past_value.shape)

    def reorder_kv_cache_first_token(self, beam_idx: torch.LongTensor):
        return self.reorder_kv_cache(beam_idx, self.cache_reorder.reorder_beams_first_token)

    def reorder_kv_cache_next_token(self, beam_idx: torch.LongTensor):
        return self.reorder_kv_cache(beam_idx, self.cache_reorder.reorder_beams_next_token)

    def pre_attn_forward(
        self,
        hidden_states: torch.Tensor,
        alibi: torch.Tensor,
        attention_mask: torch.Tensor,
        layer_past: Optional[Tuple[torch.Tensor, torch.Tensor]] = None,
        head_mask: Optional[torch.Tensor] = None,
        use_cache: bool = False,
        output_attentions: bool = False,
        token_idx: Optional[torch.Tensor] = None,
        reuse_cache: Optional[bool] = False
    ):
        fused_qkv = self.query_key_value(hidden_states)  # [batch_size, seq_length, 3 x hidden_size]

        # 3 x [batch_size, seq_length, num_heads, head_dim]
        (query_layer, key_layer, value_layer) = self._split_heads(fused_qkv)

        batch_size, q_length, _, _ = query_layer.shape
        query_layer = query_layer.transpose(1, 2).reshape(batch_size * self.num_heads, q_length, self.head_dim)
        key_layer = key_layer.transpose(1, 2).reshape(batch_size * self.num_heads, q_length, self.head_dim)
        value_layer = value_layer.transpose(1, 2).reshape(batch_size * self.num_heads, q_length, self.head_dim)

        if layer_past is not None or reuse_cache:
            if reuse_cache:
                past_key, past_value = self.past_key, self.past_value
            else:
                past_key, past_value = layer_past
            # concatenate along seq_length dimension:
            #  - key: [batch_size * self.num_heads, head_dim, kv_length]
            #  - value: [batch_size * self.num_heads, kv_length, head_dim]
            key_layer = self.k_update(past_key, key_layer, 1, token_idx)
            value_layer = self.v_update(past_value, value_layer, 1, token_idx)

        _, kv_length, _ = key_layer.shape

        if use_cache is True:
            if reuse_cache:
                present = (key_layer.shape, value_layer.shape)
            else:
                present = (key_layer, value_layer)
        else:
            present = None

        # [batch_size * num_heads, q_length, kv_length]
        # we use `torch.Tensor.baddbmm` instead of `torch.baddbmm` as the latter isn't supported by TorchScript v1.11
        matmul_result = alibi.baddbmm(
            batch1=query_layer,
            batch2=key_layer.transpose(1, 2),
            beta=self.beta,
            alpha=self.inv_norm_factor,
        )

        # change view to [batch_size, num_heads, q_length, kv_length]
        attention_scores = matmul_result.view(batch_size, self.num_heads, q_length, kv_length)

        # cast attention scores to fp32, compute scaled softmax and cast back to initial dtype - [batch_size, num_heads, q_length, kv_length]
        input_dtype = attention_scores.dtype
        # `float16` has a minimum value of -65504.0, whereas `bfloat16` and `float32` have a minimum value of `-3.4e+38`
        if input_dtype == torch.float16:
            attention_scores = attention_scores.to(torch.float)
        attn_weights = torch.masked_fill(attention_scores, attention_mask, torch.finfo(attention_scores.dtype).min)
        attention_probs = F.softmax(attn_weights, dim=-1, dtype=torch.float32).to(input_dtype)

        # [batch_size, num_heads, q_length, kv_length]
        attention_probs = self.attention_dropout(attention_probs)

        if head_mask is not None:
            attention_probs = attention_probs * head_mask

        # change view [batch_size x num_heads, q_length, kv_length]
        attention_probs_reshaped = attention_probs.view(batch_size * self.num_heads, q_length, kv_length)

        # matmul: [batch_size * num_heads, q_length, head_dim]
        context_layer = torch.bmm(attention_probs_reshaped, value_layer)

        # change view [batch_size, num_heads, q_length, head_dim]
        context_layer = self._merge_heads(context_layer)
        # hidden_states = context_layer
        hidden_states = pre_all_reduce(self.dense, context_layer)
        if output_attentions:
            return hidden_states, present, attention_probs
        return hidden_states, present, None
    
    def attention_all_reduce(self, hidden_states: torch.Tensor) -> torch.Tensor:
        all_reduce(self.dense, hidden_states)
    
    def post_attn_forward(self,
                          input: torch.Tensor,
                          residual: torch.Tensor):
        intermediate_output = post_all_reduce(self.dense, input)
        output_tensor = dropout_add(intermediate_output, residual, self.hidden_dropout, self.training)
        return output_tensor

    def forward(
        self,
        hidden_states: torch.Tensor,
        residual: torch.Tensor,
        alibi: torch.Tensor,
        attention_mask: torch.Tensor,
        layer_past: Optional[Tuple[torch.Tensor, torch.Tensor]] = None,
        head_mask: Optional[torch.Tensor] = None,
        use_cache: bool = False,
        output_attentions: bool = False,
        token_idx: Optional[torch.Tensor] = None,
        reuse_cache: Optional[bool] = False,
    ):
        context_layer, present, attention_probs = self.pre_attn_forward(hidden_states, alibi, attention_mask, layer_past, head_mask, use_cache, output_attentions, token_idx, reuse_cache)

        # aggregate results across tp ranks. See here: https://github.com/pytorch/pytorch/issues/76232
        if self.pretraining_tp > 1 and self.slow_but_exact:
            slices = self.hidden_size / self.pretraining_tp
            output_tensor = torch.zeros_like(context_layer)
            for i in range(self.pretraining_tp):
                output_tensor = output_tensor + F.linear(
                    context_layer[:, :, int(i * slices) : int((i + 1) * slices)],
                    self.dense.weight[:, int(i * slices) : int((i + 1) * slices)],
                )
        else:
            output_tensor = self.dense(context_layer)

        output_tensor = dropout_add(output_tensor, residual, self.hidden_dropout, self.training)

        outputs = (output_tensor, present)
        if output_attentions:
            outputs += (attention_probs,)

        return outputs


class BloomMLP(nn.Module):
    def __init__(self, config: BloomConfig):
        super().__init__()
        hidden_size = config.hidden_size

        self.pretraining_tp = config.pretraining_tp
        self.slow_but_exact = config.slow_but_exact
        self.dense_h_to_4h = nn.Linear(hidden_size, 4 * hidden_size)
        self.gelu_impl = nn.GELU(approximate='tanh')
        self.dense_4h_to_h = nn.Linear(4 * hidden_size, hidden_size)
        self.hidden_dropout = config.hidden_dropout

    def forward(self, hidden_states: torch.Tensor, residual: torch.Tensor) -> torch.Tensor:
        hidden_states = self.gelu_impl(self.dense_h_to_4h(hidden_states))

        if self.pretraining_tp > 1 and self.slow_but_exact:
            intermediate_output = torch.zeros_like(residual)
            slices = self.dense_4h_to_h.weight.shape[-1] / self.pretraining_tp
            for i in range(self.pretraining_tp):
                intermediate_output = intermediate_output + F.linear(
                    hidden_states[:, :, int(i * slices) : int((i + 1) * slices)],
                    self.dense_4h_to_h.weight[:, int(i * slices) : int((i + 1) * slices)],
                )
        else:
            intermediate_output = self.dense_4h_to_h(hidden_states)

        output = dropout_add(intermediate_output, residual, self.hidden_dropout, self.training)

        return output
    def pre_mlp_forward(self, hidden_states: torch.Tensor) -> torch.Tensor:
        hidden_states = self.gelu_impl(self.dense_h_to_4h(hidden_states))
        output = pre_all_reduce(self.dense_4h_to_h, hidden_states)
        return output

    def mlp_all_reduce(self, hidden_states: torch.Tensor) -> torch.Tensor:
        all_reduce(self.dense_4h_to_h, hidden_states)

    def post_mlp_forward(self, input: torch.Tensor, residual: torch.Tensor) -> torch.Tensor:
        intermediate_output = post_all_reduce(self.dense_4h_to_h, input)
        output = dropout_add(intermediate_output, residual, self.hidden_dropout, self.training)
        return output


class BloomBlock(nn.Module):
    def __init__(self, config: BloomConfig):
        super().__init__()
        hidden_size = config.hidden_size

        self.input_layernorm = LayerNorm(hidden_size, eps=config.layer_norm_epsilon)
        self.num_heads = config.n_head
        self.self_attention = BloomAttention(config)
        self.post_attention_layernorm = LayerNorm(hidden_size, eps=config.layer_norm_epsilon)

        self.mlp = BloomMLP(config)

        self.apply_residual_connection_post_layernorm = config.apply_residual_connection_post_layernorm
        self.hidden_dropout = config.hidden_dropout

    def allocate_kv_cache(self, batch_size, seq_len, kv_cache_fp8):
        self.self_attention.allocate_kv_cache(batch_size, seq_len, kv_cache_fp8)

    def prepare_for_new_input(self, input_length: int, num_beams: int, batch_size: int, device: torch.device):
        self.self_attention.prepare_for_new_input(input_length, num_beams, batch_size, device)

    def prepare_next_beam_idx(self, beam_idx: torch.LongTensor):
        return self.self_attention.prepare_next_beam_idx(beam_idx)

    def reorder_kv_cache_first_token(self, beam_idx: torch.LongTensor):
        return self.self_attention.reorder_kv_cache_first_token(beam_idx)

    def reorder_kv_cache_next_token(self, beam_idx: torch.LongTensor):
        return self.self_attention.reorder_kv_cache_next_token(beam_idx)

    def pre_attn(
        self,
        hidden_states: torch.Tensor,
        alibi: torch.Tensor,
        attention_mask: torch.Tensor,
        layer_past: Optional[Tuple[torch.Tensor, torch.Tensor]] = None,
        head_mask: Optional[torch.Tensor] = None,
        use_cache: bool = False,
        output_attentions: bool = False,
        token_idx: Optional[torch.Tensor] = None,
        reuse_cache: Optional[bool] = None,
    ):
        # hidden_states: [batch_size, seq_length, hidden_size]

        # Layer norm at the beginning of the transformer layer.
        layernorm_output = self.input_layernorm(hidden_states)

        # Layer norm post the self attention.
        if self.apply_residual_connection_post_layernorm:
            residual = layernorm_output
        else:
            residual = hidden_states
        hidden_states, present, attention_probs = self.self_attention.pre_attn_forward(layernorm_output, alibi, attention_mask, layer_past, head_mask, use_cache, output_attentions, token_idx, reuse_cache)
        return hidden_states, present, attention_probs, residual
    
    def post_attn_pre_mlp(self, input, residual):
        attention_output = self.self_attention.post_attn_forward(input, residual)

        layernorm_output = self.post_attention_layernorm(attention_output)
        # Get residual
        if self.apply_residual_connection_post_layernorm:
            residual = layernorm_output
        else:
            residual = attention_output
        output = self.mlp.pre_mlp_forward(layernorm_output)
        return output, residual
    
    def post_mlp(self, input, resiudal):
        output = self.mlp.post_mlp_forward(input, resiudal)

        return output

    def forward(
        self,
        hidden_states: torch.Tensor,
        alibi: torch.Tensor,
        attention_mask: torch.Tensor,
        layer_past: Optional[Tuple[torch.Tensor, torch.Tensor]] = None,
        head_mask: Optional[torch.Tensor] = None,
        use_cache: bool = False,
        output_attentions: bool = False,
        token_idx: Optional[torch.Tensor] = None,
        reuse_cache: Optional[bool] = None,
    ):
        output_pre_attn, present, attention_probs, resiudal_attn = self.pre_attn(hidden_states, alibi, attention_mask, layer_past, head_mask, use_cache, output_attentions, token_idx, reuse_cache)

        self.self_attention.attention_all_reduce(output_pre_attn)
        output_post_attn_pre_mlp, resiudal_mlp = self.post_attn_pre_mlp(output_pre_attn, resiudal_attn)
        self.mlp.mlp_all_reduce(output_post_attn_pre_mlp)
        output_mlp = self.post_mlp(output_post_attn_pre_mlp, resiudal_mlp)

        outputs = (present,)
        if output_attentions:
            outputs += (attention_probs,)
        if use_cache:
            outputs = (output_mlp,) + outputs
        else:
            outputs = (output_mlp,) + outputs[1:]
        return outputs


class BloomPreTrainedModel(PreTrainedModel):
    _keys_to_ignore_on_load_missing = [r"h.*.self_attention.scale_mask_softmax.causal_mask", r"lm_head.weight"]
    """
    An abstract class to handle weights initialization and a simple interface for downloading and loading pretrained
    models.
    """

    config_class = BloomConfig
    base_model_prefix = "transformer"
    supports_gradient_checkpointing = True
    _no_split_modules = ["BloomBlock"]

    def __init__(self, *inputs, **kwargs):
        super().__init__(*inputs, **kwargs)

    def _init_weights(self, module: nn.Module):
        """Initialize the weights."""
        if isinstance(module, nn.Linear):
            # Slightly different from the TF version which uses truncated_normal for initialization
            # cf https://github.com/pytorch/pytorch/pull/5617
            module.weight.data.normal_(mean=0.0, std=self.config.initializer_range)
            if module.bias is not None:
                module.bias.data.zero_()
        elif isinstance(module, nn.Embedding):
            module.weight.data.normal_(mean=0.0, std=self.config.initializer_range)
            if module.padding_idx is not None:
                module.weight.data[module.padding_idx].zero_()
        elif isinstance(module, LayerNorm):
            module.bias.data.zero_()
            module.weight.data.fill_(1.0)

    def _set_gradient_checkpointing(self, module: nn.Module, value: bool = False):
        if isinstance(module, BloomModel):
            module.gradient_checkpointing = value

    def _convert_to_standard_cache(
        self, past_key_value: Tuple[Tuple[torch.Tensor, torch.Tensor]]
    ) -> Tuple[Tuple[torch.Tensor, torch.Tensor]]:
        """
        Standardizes the format of the cache so as to match most implementations, i.e. to tuple(tuple([batch_size,
        num_heads, ...]))
        """
        batch_size_times_num_heads, seq_length, head_dim = past_key_value[0][0].shape
        tp_world_size = int(os.environ.get('WORLD_SIZE', 1))
        num_heads = self.config.n_head // tp_world_size
        batch_size = batch_size_times_num_heads // num_heads

        # key: [batch_size * num_heads, seq_length, head_dim] -> [batch_size, num_heads, seq_length, head_dim]
        # value: [batch_size * num_heads, seq_length, head_dim] -> [batch_size, num_heads, seq_length, head_dim]
        return tuple(
            (
                layer_past[0].view(batch_size, num_heads, seq_length, head_dim),
                layer_past[1].view(batch_size, num_heads, seq_length, head_dim),
            )
            for layer_past in past_key_value
        )

    def _convert_to_bloom_cache(
        self, past_key_value: Tuple[Tuple[torch.Tensor, torch.Tensor]]
    ) -> Tuple[Tuple[torch.Tensor, torch.Tensor]]:
        """
        Converts the cache to the format expected by Bloom, i.e. to tuple(tuple([batch_size * num_heads, ...]))
        """
        batch_size, num_heads, seq_length, head_dim = past_key_value[0][0].shape
        batch_size_times_num_heads = batch_size * num_heads
        # key:  [batch_size, num_heads, seq_length, head_dim] -> [batch_size * num_heads, seq_length, head_dim]
        # value: [batch_size, num_heads, seq_length, head_dim] -> [batch_size * num_heads, seq_length, head_dim]
        return tuple(
            (
                layer_past[0].view(batch_size_times_num_heads, seq_length, head_dim),
                layer_past[1].view(batch_size_times_num_heads, seq_length, head_dim),
            )
            for layer_past in past_key_value
        )


BLOOM_START_DOCSTRING = r"""

    This model inherits from [`PreTrainedModel`]. Check the superclass documentation for the generic methods the
    library implements for all its model (such as downloading or saving, resizing the input embeddings etc.)

    This model is also a PyTorch [torch.nn.Module](https://pytorch.org/docs/stable/nn.html#torch.nn.Module) subclass.
    Use it as a regular PyTorch Module and refer to the PyTorch documentation for all matter related to general usage
    and behavior.

    Parameters:
        config ([`BloomConfig`]): Model configuration class with all the parameters of the model.
            Initializing with a config file does not load the weights associated with the model, only the
            configuration. Check out the [`~PreTrainedModel.from_pretrained`] method to load the model weights.
"""

BLOOM_INPUTS_DOCSTRING = r"""
    Args:
        input_ids (`torch.LongTensor` of shape `(batch_size, input_ids_length)`):
            `input_ids_length` = `sequence_length` if `past_key_values` is `None` else `past_key_values[0][0].shape[2]`
            (`sequence_length` of input past key value states). Indices of input sequence tokens in the vocabulary.

            If `past_key_values` is used, only `input_ids` that do not have their past calculated should be passed as
            `input_ids`.

            Indices can be obtained using [`AutoTokenizer`]. See [`PreTrainedTokenizer.encode`] and
            [`PreTrainedTokenizer.__call__`] for details.

            [What are input IDs?](../glossary#input-ids)
        past_key_values (`Tuple[Tuple[torch.Tensor]]` of length `config.n_layers`):
            Contains precomputed hidden-states (key and values in the attention blocks) as computed by the model (see
            `past_key_values` output below). Can be used to speed up sequential decoding. The `input_ids` which have
            their past given to this model should not be passed as `input_ids` as they have already been computed.

            Each element of `past_key_values` is a tuple (past_key, past_value):
            - past_key: [batch_size * num_heads, head_dim, kv_length]
            - past_value: [batch_size * num_heads, kv_length, head_dim]
        attention_mask (`torch.FloatTensor` 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)
        head_mask (`torch.FloatTensor` of shape `(num_heads,)` or `(num_layers, num_heads)`, *optional*):
            Mask to nullify selected heads of the self-attention modules. Mask values selected in `[0, 1]`:

            - 1 indicates the head is **not masked**,
            - 0 indicates the head is **masked**.

        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.

            If `past_key_values` is used, optionally only the last `inputs_embeds` have to be input (see
            `past_key_values`).
        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 [`~file_utils.ModelOutput`] instead of a plain tuple.
"""


@add_start_docstrings(
    "The bare Bloom Model transformer outputting raw hidden-states without any specific head on top.",
    BLOOM_START_DOCSTRING,
)
class BloomModel(BloomPreTrainedModel):
    def __init__(self, config: BloomConfig):
        super().__init__(config)

        self.embed_dim = config.hidden_size
        self.num_heads = config.n_head

        # Embedding + LN Embedding
        self.word_embeddings = nn.Embedding(config.vocab_size, self.embed_dim)
        self.word_embeddings_layernorm = LayerNorm(self.embed_dim, eps=config.layer_norm_epsilon)

        # Transformer blocks
        self.h = nn.ModuleList([BloomBlock(config) for _ in range(config.num_hidden_layers)])

        # Final Layer Norm
        self.ln_f = LayerNorm(self.embed_dim, eps=config.layer_norm_epsilon)

        self.gradient_checkpointing = False

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

    def get_input_embeddings(self):
        return self.word_embeddings

    def _prepare_attn_mask(
        self, attention_mask: torch.Tensor, input_shape: Tuple[int, int], past_key_values_length: int
    ) -> torch.BoolTensor:
        # create causal mask
        # [batch_size, seq_length] -> [batch_size, 1, tgt_length, src_length]
        combined_attention_mask = None
        device = attention_mask.device
        _, src_length = input_shape

        if src_length > 1:
            combined_attention_mask = _make_causal_mask(
                input_shape, device=device, past_key_values_length=past_key_values_length
            )

        # [batch_size, seq_length] -> [batch_size, 1, tgt_length, src_length]
        expanded_attn_mask = _expand_mask(attention_mask, tgt_length=src_length)
        combined_attention_mask = (
            expanded_attn_mask if combined_attention_mask is None else expanded_attn_mask | combined_attention_mask
        )

        return combined_attention_mask

    def set_input_embeddings(self, new_embeddings: torch.Tensor):
        self.word_embeddings = new_embeddings

    def allocate_kv_cache(self, batch_size, max_seq_len, kv_cache_fp8):
        for layer in self.h:
            layer.allocate_kv_cache(batch_size, max_seq_len, kv_cache_fp8)

    def prepare_for_new_input(self, input_length: int, num_beams: int, batch_size: int, device: torch.device):
        for layer in self.h:
            layer.prepare_for_new_input(input_length, num_beams, batch_size, device)

    def reorder_kv_cache_first_token(self, beam_idx: torch.LongTensor):
        return tuple(layer.reorder_kv_cache_first_token(beam_idx) for layer in self.h)

    def reorder_kv_cache_next_token(self, beam_idx: torch.LongTensor):
        beam_idx = self.h[0].prepare_next_beam_idx(beam_idx) if self.h else None
        return tuple(layer.reorder_kv_cache_next_token(beam_idx) for layer in self.h)

    @add_start_docstrings_to_model_forward(BLOOM_INPUTS_DOCSTRING)
    @add_code_sample_docstrings(
        checkpoint=_CHECKPOINT_FOR_DOC,
        output_type=BaseModelOutputWithPastAndCrossAttentions,
        config_class=_CONFIG_FOR_DOC,
    )
    def forward(
        self,
        input_ids: Optional[torch.LongTensor] = None,
        past_key_values: Optional[Tuple[Tuple[torch.Tensor, torch.Tensor], ...]] = None,
        attention_mask: Optional[torch.Tensor] = None,
        head_mask: Optional[torch.LongTensor] = None,
        inputs_embeds: 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,
        token_idx: Optional[torch.Tensor] = None,
        reuse_cache: Optional[bool] = None,
        **deprecated_arguments,
    ) -> Union[Tuple[torch.Tensor, ...], BaseModelOutputWithPastAndCrossAttentions]:
        if deprecated_arguments.pop("position_ids", False) is not False:
            # `position_ids` could have been `torch.Tensor` or `None` so defaulting pop to `False` allows to detect if users were passing explicitly `None`
            warnings.warn(
                "`position_ids` have no functionality in BLOOM and will be removed in v5.0.0. You can safely ignore"
                " passing `position_ids`.",
                FutureWarning,
            )
        if len(deprecated_arguments) > 0:
            raise ValueError(f"Got unexpected arguments: {deprecated_arguments}")

        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
        return_dict = return_dict if return_dict is not None else self.config.use_return_dict

        if input_ids is not None and inputs_embeds is not None:
            raise ValueError("You cannot specify both input_ids and inputs_embeds at the same time")
        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 input_ids or inputs_embeds")

        if past_key_values is None:
            past_key_values = tuple([None] * len(self.h))

        # Prepare head mask if needed
        # 1.0 in head_mask indicate we keep the head
        # attention_probs has shape batch_size x num_heads x N x N
        # head_mask has shape n_layer x batch x num_heads x N x N
        head_mask = self.get_head_mask(head_mask, self.config.n_layer)

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

        hidden_states = self.word_embeddings_layernorm(inputs_embeds)

        presents = () if use_cache else None
        all_self_attentions = () if output_attentions else None
        all_hidden_states = () if output_hidden_states else None

        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

        # Compute alibi tensor: check build_alibi_tensor documentation
        seq_length_with_past = seq_length
        past_key_values_length = 0
        if past_key_values[0] is not None:
            if reuse_cache:
                past_key_values_length = past_key_values[0][0][2]
            else:
                past_key_values_length = past_key_values[0][0].shape[2]
            seq_length_with_past = seq_length_with_past + past_key_values_length
        if attention_mask is None:
            attention_mask = torch.ones((batch_size, seq_length_with_past), device=hidden_states.device)
        else:
            attention_mask = attention_mask.to(hidden_states.device)

        alibi = build_alibi_tensor(attention_mask, self.num_heads, dtype=hidden_states.dtype)

        causal_mask = self._prepare_attn_mask(
            attention_mask,
            input_shape=(batch_size, seq_length),
            past_key_values_length=past_key_values_length,
        )

        for i, (block, layer_past) in enumerate(zip(self.h, past_key_values)):
            if output_hidden_states:
                all_hidden_states = all_hidden_states + (hidden_states,)

            if self.gradient_checkpointing and self.training:

                def create_custom_forward(module):
                    def custom_forward(*inputs):
                        # None for past_key_value
                        return module(*inputs, use_cache=use_cache, output_attentions=output_attentions)

                    return custom_forward

                outputs = torch.utils.checkpoint.checkpoint(
                    create_custom_forward(block),
                    hidden_states,
                    alibi,
                    causal_mask,
                    layer_past,
                    head_mask[i],
                )
            else:
                outputs = block(
                    hidden_states,
                    layer_past=layer_past,
                    attention_mask=causal_mask,
                    head_mask=head_mask[i],
                    use_cache=use_cache,
                    output_attentions=output_attentions,
                    alibi=alibi,
                    token_idx=token_idx,
                    reuse_cache=reuse_cache,
                )

            hidden_states = outputs[0]
            if use_cache is True:
                presents = presents + (outputs[1],)

            if output_attentions:
                all_self_attentions = all_self_attentions + (outputs[2 if use_cache else 1],)

        # Add last hidden state
        hidden_states = self.ln_f(hidden_states)

        if output_hidden_states:
            all_hidden_states = all_hidden_states + (hidden_states,)

        if not return_dict:
            return tuple(v for v in [hidden_states, presents, all_hidden_states, all_self_attentions] if v is not None)

        return BaseModelOutputWithPastAndCrossAttentions(
            last_hidden_state=hidden_states,
            past_key_values=presents,
            hidden_states=all_hidden_states,
            attentions=all_self_attentions,
        )


@add_start_docstrings(
    """
    The Bloom Model transformer with a language modeling head on top (linear layer with weights tied to the input
    embeddings).
    """,
    BLOOM_START_DOCSTRING,
)
class BloomForCausalLM(BloomPreTrainedModel):
    _keys_to_ignore_on_load_missing = [r"h.*.self_attention.scale_mask_softmax.causal_mask", r"lm_head.weight"]

    def __init__(self, config: BloomConfig):
        super().__init__(config)
        self.transformer = BloomModel(config)
        self.lm_head = nn.Linear(config.hidden_size, config.vocab_size, bias=False)

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

    def get_output_embeddings(self):
        return self.lm_head

    def set_output_embeddings(self, new_embeddings: torch.Tensor):
        self.lm_head = new_embeddings

    def prepare_inputs_for_generation(
        self,
        input_ids: torch.LongTensor,
        past_key_values: Optional[torch.Tensor] = None,
        attention_mask: Optional[torch.Tensor] = None,
        inputs_embeds: Optional[torch.Tensor] = None,
        token_idx: Optional[torch.Tensor] = None,
        **kwargs,
    ) -> dict:
        # only last token for input_ids if past is not None
        if past_key_values:
            if token_idx is None:
                input_ids = input_ids[:, -1].unsqueeze(-1)
            else:
                input_ids = torch.index_select(input_ids, 1, token_idx - 1)

            # the cache may be in the stardard format (e.g. in contrastive search), convert to bloom's format if needed
            if past_key_values[0][0].shape[0] == input_ids.shape[0]:
                past_key_values = self._convert_to_bloom_cache(past_key_values)

        # 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}

        model_inputs.update(
            {
                "past_key_values": past_key_values,
                "use_cache": kwargs.get("use_cache"),
                "attention_mask": attention_mask,
                "token_idx": token_idx,
            }
        )
        return model_inputs

    def allocate_kv_cache(self, batch_size, max_seq_len, kv_cache_fp8):
        self.transformer.allocate_kv_cache(batch_size, max_seq_len, kv_cache_fp8)

    def prepare_for_new_input(self, input_length: int, num_beams: int, batch_size: int, device: torch.device):
        self.transformer.prepare_for_new_input(input_length, num_beams, batch_size, device)

    @add_start_docstrings_to_model_forward(BLOOM_INPUTS_DOCSTRING)
    @add_code_sample_docstrings(
        checkpoint=_CHECKPOINT_FOR_DOC,
        output_type=CausalLMOutputWithCrossAttentions,
        config_class=_CONFIG_FOR_DOC,
    )
    def forward(
        self,
        input_ids: Optional[torch.LongTensor] = None,
        past_key_values: Optional[Tuple[Tuple[torch.Tensor, torch.Tensor], ...]] = None,
        attention_mask: Optional[torch.Tensor] = None,
        head_mask: Optional[torch.Tensor] = None,
        inputs_embeds: Optional[torch.Tensor] = None,
        labels: Optional[torch.Tensor] = None,
        use_cache: Optional[bool] = None,
        output_attentions: Optional[bool] = None,
        output_hidden_states: Optional[bool] = None,
        return_dict: Optional[bool] = None,
        token_idx: Optional[torch.Tensor] = None,
        trim_logits: Optional[bool] = None,
        reuse_cache: Optional[bool] = None,
        **deprecated_arguments,
    ) -> Union[Tuple[torch.Tensor], CausalLMOutputWithCrossAttentions]:
        r"""
        labels (`torch.LongTensor` of shape `(batch_size, sequence_length)`, *optional*):
            Labels for language modeling. Note that the labels **are shifted** inside the model, i.e. you can set
            `labels = input_ids` Indices are selected in `[-100, 0, ..., config.vocab_size]` All labels set to `-100`
            are ignored (masked), the loss is only computed for labels in `[0, ..., config.vocab_size]`
        """
        if deprecated_arguments.pop("position_ids", False) is not False:
            # `position_ids` could have been `torch.Tensor` or `None` so defaulting pop to `False` allows to detect if users were passing explicitly `None`
            warnings.warn(
                "`position_ids` have no functionality in BLOOM and will be removed in v5.0.0. You can safely ignore"
                " passing `position_ids`.",
                FutureWarning,
            )
        if len(deprecated_arguments) > 0:
            raise ValueError(f"Got unexpected arguments: {deprecated_arguments}")

        return_dict = return_dict if return_dict is not None else self.config.use_return_dict

        transformer_outputs = self.transformer(
            input_ids,
            past_key_values=past_key_values,
            attention_mask=attention_mask,
            head_mask=head_mask,
            inputs_embeds=inputs_embeds,
            use_cache=use_cache,
            output_attentions=output_attentions,
            output_hidden_states=output_hidden_states,
            return_dict=return_dict,
            token_idx=token_idx,
            reuse_cache=reuse_cache,
        )
        hidden_states = transformer_outputs[0]
        _, seq_len, _ = hidden_states.shape

        if seq_len > 1 and trim_logits:
            if token_idx is not None:
                hidden_states = hidden_states.index_select(1, token_idx - 1)
            else:
                hidden_states = hidden_states[:, -1, :].unsqueeze(-2)

        lm_logits = self.lm_head(hidden_states)

        loss = None
        if labels is not None:
            # Shift so that tokens < n predict n
            shift_logits = lm_logits[..., :-1, :].contiguous()
            shift_labels = labels[..., 1:].contiguous()
            batch_size, seq_length, vocab_size = shift_logits.shape
            # Flatten the tokens
            loss_fct = CrossEntropyLoss()
            loss = loss_fct(
                shift_logits.view(batch_size * seq_length, vocab_size), shift_labels.view(batch_size * seq_length)
            )

        if not return_dict:
            output = (lm_logits,) + transformer_outputs[1:]
            return ((loss,) + output) if loss is not None else output

        return CausalLMOutputWithCrossAttentions(
            loss=loss,
            logits=lm_logits,
            past_key_values=transformer_outputs.past_key_values,
            hidden_states=transformer_outputs.hidden_states,
            attentions=transformer_outputs.attentions,
        )

    def reorder_kv_cache_first_token(self, beam_idx: torch.LongTensor):
        return self.transformer.reorder_kv_cache_first_token(beam_idx)

    def reorder_kv_cache_next_token(self, beam_idx: torch.LongTensor):
        return self.transformer.reorder_kv_cache_next_token(beam_idx)

    def _reorder_cache(
        self, 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`.
        """
        standardized_past = self._convert_to_standard_cache(past)

        # Get a copy of `beam_idx` on all the devices where we need those indices.
        device_to_beam_idx = {
            past_state.device: beam_idx.to(past_state.device) for layer_past in past for past_state in layer_past
        }
        reordered_past = tuple(
            (
                layer_past[0].index_select(0, device_to_beam_idx[layer_past[0].device]),
                layer_past[1].index_select(0, device_to_beam_idx[layer_past[0].device]),
            )
            for layer_past in standardized_past
        )
        return self._convert_to_bloom_cache(reordered_past)


@add_start_docstrings(
    """
    The Bloom Model transformer with a sequence classification head on top (linear layer).

    [`BloomForSequenceClassification`] uses the last token in order to do the classification, as other causal models
    (e.g. GPT-1) do.

    Since it does classification on the last token, it requires to know the position of the last token. If a
    `pad_token_id` is defined in the configuration, it finds the last token that is not a padding token in each row. If
    no `pad_token_id` is defined, it simply takes the last value in each row of the batch. Since it cannot guess the
    padding tokens when `inputs_embeds` are passed instead of `input_ids`, it does the same (take the last value in
    each row of the batch).
    """,
    BLOOM_START_DOCSTRING,
)
class BloomForSequenceClassification(BloomPreTrainedModel):
    _keys_to_ignore_on_load_missing = [r"h.*.self_attention.scale_mask_softmax.causal_mask", r"lm_head.weight"]

    def __init__(self, config: BloomConfig):
        super().__init__(config)
        self.num_labels = config.num_labels
        self.transformer = BloomModel(config)
        self.score = nn.Linear(config.hidden_size, config.num_labels, bias=False)

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

    @add_start_docstrings_to_model_forward(BLOOM_INPUTS_DOCSTRING)
    @add_code_sample_docstrings(
        checkpoint=_CHECKPOINT_FOR_DOC,
        output_type=SequenceClassifierOutputWithPast,
        config_class=_CONFIG_FOR_DOC,
    )
    def forward(
        self,
        input_ids: Optional[torch.LongTensor] = None,
        past_key_values: Optional[Tuple[Tuple[torch.Tensor, torch.Tensor], ...]] = None,
        attention_mask: Optional[torch.Tensor] = None,
        head_mask: Optional[torch.Tensor] = None,
        inputs_embeds: Optional[torch.Tensor] = None,
        labels: Optional[torch.Tensor] = None,
        use_cache: Optional[bool] = None,
        output_attentions: Optional[bool] = None,
        output_hidden_states: Optional[bool] = None,
        return_dict: Optional[bool] = None,
        **deprecated_arguments,
    ) -> Union[Tuple[torch.Tensor], SequenceClassifierOutputWithPast]:
        r"""
        labels (`torch.LongTensor` of shape `(batch_size,)`, *optional*):
            Labels for computing the sequence classification/regression loss. Indices should be in `[0, ...,
            config.num_labels - 1]`. If `config.num_labels == 1` a regression loss is computed (Mean-Square loss), If
            `config.num_labels > 1` a classification loss is computed (Cross-Entropy).
        """
        if deprecated_arguments.pop("position_ids", False) is not False:
            # `position_ids` could have been `torch.Tensor` or `None` so defaulting pop to `False` allows to detect if users were passing explicitly `None`
            warnings.warn(
                "`position_ids` have no functionality in BLOOM and will be removed in v5.0.0. You can safely ignore"
                " passing `position_ids`.",
                FutureWarning,
            )
        if len(deprecated_arguments) > 0:
            raise ValueError(f"Got unexpected arguments: {deprecated_arguments}")

        return_dict = return_dict if return_dict is not None else self.config.use_return_dict

        transformer_outputs = self.transformer(
            input_ids,
            past_key_values=past_key_values,
            attention_mask=attention_mask,
            head_mask=head_mask,
            inputs_embeds=inputs_embeds,
            use_cache=use_cache,
            output_attentions=output_attentions,
            output_hidden_states=output_hidden_states,
            return_dict=return_dict,
        )

        hidden_states = transformer_outputs[0]
        logits = self.score(hidden_states)

        if input_ids is not None:
            batch_size = input_ids.shape[0]
        else:
            batch_size = inputs_embeds.shape[0]

        if self.config.pad_token_id is None and batch_size != 1:
            raise ValueError("Cannot handle batch sizes > 1 if no padding token is defined.")
        if self.config.pad_token_id is None:
            sequence_lengths = -1
        else:
            if input_ids is not None:
                sequence_lengths = (torch.ne(input_ids, self.config.pad_token_id).sum(-1) - 1).to(logits.device)
            else:
                sequence_lengths = -1
                logger.warning(
                    f"{self.__class__.__name__} will not detect padding tokens in `inputs_embeds`. Results may be "
                    "unexpected if using padding tokens in conjunction with `inputs_embeds.`"
                )

        pooled_logits = logits[torch.arange(batch_size, device=logits.device), sequence_lengths]

        loss = None
        if labels is not None:
            if self.config.problem_type is None:
                if self.num_labels == 1:
                    self.config.problem_type = "regression"
                elif self.num_labels > 1 and (labels.dtype == torch.long or labels.dtype == torch.int):
                    self.config.problem_type = "single_label_classification"
                else:
                    self.config.problem_type = "multi_label_classification"

            if self.config.problem_type == "regression":
                loss_fct = MSELoss()
                if self.num_labels == 1:
                    loss = loss_fct(pooled_logits.squeeze(), labels.squeeze())
                else:
                    loss = loss_fct(pooled_logits, labels)
            elif self.config.problem_type == "single_label_classification":
                loss_fct = CrossEntropyLoss()
                loss = loss_fct(pooled_logits, labels)
            elif self.config.problem_type == "multi_label_classification":
                loss_fct = BCEWithLogitsLoss()
                loss = loss_fct(pooled_logits, labels)
        if not return_dict:
            output = (pooled_logits,) + transformer_outputs[1:]
            return ((loss,) + output) if loss is not None else output

        return SequenceClassifierOutputWithPast(
            loss=loss,
            logits=pooled_logits,
            past_key_values=transformer_outputs.past_key_values,
            hidden_states=transformer_outputs.hidden_states,
            attentions=transformer_outputs.attentions,
        )


@add_start_docstrings(
    """
    Bloom Model with a token classification head on top (a linear layer on top of the hidden-states output) e.g. for
    Named-Entity-Recognition (NER) tasks.
    """,
    BLOOM_START_DOCSTRING,
)
class BloomForTokenClassification(BloomPreTrainedModel):
    _keys_to_ignore_on_load_missing = [r"h.*.self_attention.scale_mask_softmax.causal_mask", r"lm_head.weight"]

    def __init__(self, config: BloomConfig):
        super().__init__(config)
        self.num_labels = config.num_labels

        self.transformer = BloomModel(config)
        if hasattr(config, "classifier_dropout") and config.classifier_dropout is not None:
            classifier_dropout = config.classifier_dropout
        elif hasattr(config, "hidden_dropout") and config.hidden_dropout is not None:
            classifier_dropout = config.hidden_dropout
        else:
            classifier_dropout = 0.1
        self.dropout = nn.Dropout(classifier_dropout)
        self.classifier = nn.Linear(config.hidden_size, config.num_labels)

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

    @add_start_docstrings_to_model_forward(BLOOM_INPUTS_DOCSTRING)
    @add_code_sample_docstrings(
        checkpoint=_CHECKPOINT_FOR_DOC,
        output_type=TokenClassifierOutput,
        config_class=_CONFIG_FOR_DOC,
    )
    def forward(
        self,
        input_ids: Optional[torch.LongTensor] = None,
        past_key_values: Optional[Tuple[Tuple[torch.Tensor, torch.Tensor], ...]] = None,
        attention_mask: Optional[torch.Tensor] = None,
        head_mask: Optional[torch.Tensor] = None,
        inputs_embeds: Optional[torch.Tensor] = None,
        labels: Optional[torch.Tensor] = None,
        use_cache: Optional[bool] = None,
        output_attentions: Optional[bool] = None,
        output_hidden_states: Optional[bool] = None,
        return_dict: Optional[bool] = None,
        **deprecated_arguments,
    ) -> Union[Tuple[torch.Tensor], TokenClassifierOutput]:
        r"""
        labels (`torch.LongTensor` of shape `(batch_size,)`, *optional*):
            Labels for computing the sequence classification/regression loss. Indices should be in `[0, ...,
            config.num_labels - 1]`. If `config.num_labels == 1` a regression loss is computed (Mean-Square loss), If
            `config.num_labels > 1` a classification loss is computed (Cross-Entropy).
        """
        if deprecated_arguments.pop("position_ids", False) is not False:
            # `position_ids` could have been `torch.Tensor` or `None` so defaulting pop to `False` allows to detect if users were passing explicitly `None`
            warnings.warn(
                "`position_ids` have no functionality in BLOOM and will be removed in v5.0.0. You can safely ignore"
                " passing `position_ids`.",
                FutureWarning,
            )
        if len(deprecated_arguments) > 0:
            raise ValueError(f"Got unexpected arguments: {deprecated_arguments}")

        return_dict = return_dict if return_dict is not None else self.config.use_return_dict

        transformer_outputs = self.transformer(
            input_ids,
            past_key_values=past_key_values,
            attention_mask=attention_mask,
            head_mask=head_mask,
            inputs_embeds=inputs_embeds,
            use_cache=use_cache,
            output_attentions=output_attentions,
            output_hidden_states=output_hidden_states,
            return_dict=return_dict,
        )

        hidden_states = transformer_outputs[0]
        hidden_states = self.dropout(hidden_states)
        logits = self.classifier(hidden_states)

        loss = None
        if labels is not None:
            batch_size, seq_length = labels.shape
            loss_fct = CrossEntropyLoss()
            loss = loss_fct(
                logits.view(batch_size * seq_length, self.num_labels), labels.view(batch_size * seq_length)
            )

        if not return_dict:
            output = (logits,) + transformer_outputs[2:]
            return ((loss,) + output) if loss is not None else output

        return TokenClassifierOutput(
            loss=loss,
            logits=logits,
            hidden_states=transformer_outputs.hidden_states,
            attentions=transformer_outputs.attentions,
        )


@add_start_docstrings(
    """
    The BLOOM Model transformer with a span classification head on top for extractive question-answering tasks like
    SQuAD (a linear layers on top of the hidden-states output to compute `span start logits` and `span end logits`).
    """,
    BLOOM_START_DOCSTRING,
)
class BloomForQuestionAnswering(BloomPreTrainedModel):
    _keys_to_ignore_on_load_missing = [r"h.*.self_attention.scale_mask_softmax.causal_mask", r"lm_head.weight"]

    def __init__(self, config):
        super().__init__(config)
        self.transformer = BloomModel(config)
        self.qa_outputs = nn.Linear(config.hidden_size, 2)

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

    @add_start_docstrings_to_model_forward(BLOOM_INPUTS_DOCSTRING.format("batch_size, sequence_length"))
    def forward(
        self,
        input_ids: Optional[torch.LongTensor] = None,
        attention_mask: Optional[torch.FloatTensor] = None,
        position_ids: Optional[torch.LongTensor] = None,
        head_mask: Optional[torch.FloatTensor] = None,
        inputs_embeds: Optional[torch.FloatTensor] = None,
        start_positions: Optional[torch.LongTensor] = None,
        end_positions: Optional[torch.LongTensor] = None,
        output_attentions: Optional[bool] = None,
        output_hidden_states: Optional[bool] = None,
        return_dict: Optional[bool] = None,
    ) -> Union[Tuple, QuestionAnsweringModelOutput]:
        r"""
        start_positions (`torch.LongTensor` of shape `(batch_size,)`, *optional*):
            Labels for position (index) of the start of the labelled span for computing the token classification loss.
            Positions are clamped to the length of the sequence (`sequence_length`). Position outside of the sequence
            are not taken into account for computing the loss.
        end_positions (`torch.LongTensor` of shape `(batch_size,)`, *optional*):
            Labels for position (index) of the end of the labelled span for computing the token classification loss.
            Positions are clamped to the length of the sequence (`sequence_length`). Position outside of the sequence
            are not taken into account for computing the loss.
        """
        return_dict = return_dict if return_dict is not None else self.config.use_return_dict

        outputs = self.transformer(
            input_ids,
            attention_mask=attention_mask,
            position_ids=position_ids,
            head_mask=head_mask,
            inputs_embeds=inputs_embeds,
            output_attentions=output_attentions,
            output_hidden_states=output_hidden_states,
            return_dict=return_dict,
        )

        sequence_output = outputs[0]

        logits = self.qa_outputs(sequence_output)
        start_logits, end_logits = logits.split(1, dim=-1)
        start_logits = start_logits.squeeze(-1).contiguous()
        end_logits = end_logits.squeeze(-1).contiguous()

        total_loss = None
        if start_positions is not None and end_positions is not None:
            # If we are on multi-GPU, split add a dimension
            if len(start_positions.size()) > 1:
                start_positions = start_positions.squeeze(-1)
            if len(end_positions.size()) > 1:
                end_positions = end_positions.squeeze(-1)
            # sometimes the start/end positions are outside our model inputs, we ignore these terms
            ignored_index = start_logits.size(1)
            start_positions = start_positions.clamp(0, ignored_index)
            end_positions = end_positions.clamp(0, ignored_index)

            loss_fct = CrossEntropyLoss(ignore_index=ignored_index)
            start_loss = loss_fct(start_logits, start_positions)
            end_loss = loss_fct(end_logits, end_positions)
            total_loss = (start_loss + end_loss) / 2

        if not return_dict:
            output = (start_logits, end_logits) + outputs[2:]
            return ((total_loss,) + output) if total_loss is not None else output

        return QuestionAnsweringModelOutput(
            loss=total_loss,
            start_logits=start_logits,
            end_logits=end_logits,
            hidden_states=outputs.hidden_states,
            attentions=outputs.attentions,
        )