from typing import Callable, List, Optional, Tuple, Union import random import torch import torch.nn as nn from transformers.models.llama import LlamaForCausalLM, LlamaConfig from transformers.utils import logging from transformers.modeling_utils import ALL_ATTENTION_FUNCTIONS logger = logging.get_logger(__name__) class SharperLlamaForCausalLM(LlamaForCausalLM): def __init__(self, config: LlamaConfig): super().__init__(config) logger.info(f"\n* SLlama: {self.model}") for layer_i, layer in enumerate(self.model.layers): hidden_dim = config.hidden_size n_heads = config.num_attention_heads depth = layer_i # or pass 0 layer.self_attn = SharperLlamaAttention( config, layer_idx=layer_i ) class SharperLlamaAttention(nn.Module): """Multi-headed attention from 'Attention Is All You Need' paper""" def __init__(self, config: LlamaConfig, layer_idx: int): super().__init__() self.config = config self.layer_idx = layer_idx self.head_dim = getattr(config, "head_dim", config.hidden_size // config.num_attention_heads) self.num_key_value_groups = config.num_attention_heads // config.num_key_value_heads self.scaling = self.head_dim**-0.5 self.attention_dropout = config.attention_dropout self.is_causal = True self.q_proj = nn.Linear( config.hidden_size, config.num_attention_heads * self.head_dim, bias=config.attention_bias ) self.k_proj = nn.Linear( config.hidden_size, config.num_key_value_heads * self.head_dim, bias=config.attention_bias ) self.v_proj = nn.Linear( config.hidden_size, config.num_key_value_heads * self.head_dim, bias=config.attention_bias ) self.o_proj = nn.Linear( config.num_attention_heads * self.head_dim, config.hidden_size, bias=config.attention_bias ) if layer_idx < 2: self.temperature = nn.Parameter(torch.ones(config.num_attention_heads, 2048) / config.temperature_div) # num_heads, seq_len self.temperature_value = 1.0 / config.temperature_div print(f"[params] Layer {layer_idx} temperature: {self.temperature_value}") else: self.register_buffer("temperature", torch.ones(config.num_attention_heads, 2048)) self.temperature_value = 1.0 print(f"[buffer] Layer {layer_idx} temperature: {self.temperature.shape}") self.q_projs = None self.k_projs = None self.v_projs = None def forward( self, hidden_states: torch.Tensor, position_embeddings: Tuple[torch.Tensor, torch.Tensor], attention_mask: Optional[torch.Tensor], past_key_value = None, cache_position: Optional[torch.LongTensor] = None, **kwargs, ) -> Tuple[torch.Tensor, Optional[torch.Tensor], Optional[Tuple[torch.Tensor]]]: input_shape = hidden_states.shape[:-1] hidden_shape = (*input_shape, -1, self.head_dim) seq_len = input_shape[-1] temperature = self.temperature[:, :seq_len].unsqueeze(0).unsqueeze(-1) # (1, num_heads, seq_len, 1) query_states = self.q_proj(hidden_states).view(hidden_shape).transpose(1, 2) # (batch, num_heads, seq_len, head_dim) key_states = self.k_proj(hidden_states).view(hidden_shape).transpose(1, 2) value_states = self.v_proj(hidden_states).view(hidden_shape).transpose(1, 2) cos, sin = position_embeddings query_states, key_states = apply_rotary_pos_emb(query_states, key_states, cos, sin) if past_key_value is not None: # sin and cos are specific to RoPE models; cache_position needed for the static cache cache_kwargs = {"sin": sin, "cos": cos, "cache_position": cache_position} key_states, value_states = past_key_value.update(key_states, value_states, self.layer_idx, cache_kwargs) attention_interface: Callable = eager_attention_forward if self.config._attn_implementation != "eager": if self.config._attn_implementation == "sdpa" and kwargs.get("output_attentions", False): logger.warning_once( "`torch.nn.functional.scaled_dot_product_attention` does not support `output_attentions=True`. Falling back to " 'eager attention. This warning can be removed using the argument `attn_implementation="eager"` when loading the model.' ) else: attention_interface = ALL_ATTENTION_FUNCTIONS[self.config._attn_implementation] if random.random() < 0.8: query_states = query_states / (temperature + 1e-6) # pre-multiply temperature self.temperature_value = temperature.mean().item() self.q_projs = self.q_proj.weight.data.chunk(self.config.num_attention_heads, dim=0) self.q_projs = [torch.norm(q, p=2).cpu().item() for q in self.q_projs] self.k_projs = self.k_proj.weight.data.chunk(self.config.num_key_value_heads, dim=0) self.k_projs = [torch.norm(k, p=2).cpu().item() for k in self.k_projs] self.v_projs = self.v_proj.weight.data.chunk(self.config.num_key_value_heads, dim=0) self.v_projs = [torch.norm(v, p=2).cpu().item() for v in self.v_projs] attn_output, attn_weights = attention_interface( self, query_states, key_states, value_states, attention_mask, dropout=0.0 if not self.training else self.attention_dropout, scaling=self.scaling, **kwargs, ) attn_output = attn_output.reshape(*input_shape, -1).contiguous() attn_output = self.o_proj(attn_output) return attn_output, attn_weights def repeat_kv(hidden_states: torch.Tensor, n_rep: int) -> torch.Tensor: """ This is the equivalent of torch.repeat_interleave(x, dim=1, repeats=n_rep). The hidden states go from (batch, num_key_value_heads, seqlen, head_dim) to (batch, num_attention_heads, seqlen, head_dim) """ batch, num_key_value_heads, slen, head_dim = hidden_states.shape if n_rep == 1: return hidden_states hidden_states = hidden_states[:, :, None, :, :].expand(batch, num_key_value_heads, n_rep, slen, head_dim) return hidden_states.reshape(batch, num_key_value_heads * n_rep, slen, head_dim) def eager_attention_forward( module: nn.Module, query: torch.Tensor, key: torch.Tensor, value: torch.Tensor, attention_mask: Optional[torch.Tensor], scaling: float, dropout: float = 0.0, **kwargs, ): key_states = repeat_kv(key, module.num_key_value_groups) value_states = repeat_kv(value, module.num_key_value_groups) temperature = kwargs.get("temperature", 1.0) attn_weights = torch.matmul(query, key_states.transpose(2, 3)) * scaling / temperature if attention_mask is not None: causal_mask = attention_mask[:, :, :, : key_states.shape[-2]] attn_weights = attn_weights + causal_mask attn_weights = nn.functional.softmax(attn_weights, dim=-1, dtype=torch.float32).to(query.dtype) attn_weights = nn.functional.dropout(attn_weights, p=dropout, training=module.training) attn_output = torch.matmul(attn_weights, value_states) attn_output = attn_output.transpose(1, 2).contiguous() return attn_output, attn_weights class LlamaRMSNorm(nn.Module): def __init__(self, hidden_size, eps=1e-6): """ LlamaRMSNorm is equivalent to T5LayerNorm """ super().__init__() self.weight = nn.Parameter(torch.ones(hidden_size)) self.variance_epsilon = eps def forward(self, hidden_states): input_dtype = hidden_states.dtype hidden_states = hidden_states.to(torch.float32) variance = hidden_states.pow(2).mean(-1, keepdim=True) hidden_states = hidden_states * torch.rsqrt(variance + self.variance_epsilon) return self.weight * hidden_states.to(input_dtype) def extra_repr(self): return f"{tuple(self.weight.shape)}, eps={self.variance_epsilon}" class LlamaRotaryEmbedding(nn.Module): def __init__(self, config: LlamaConfig, device=None): super().__init__() # BC: "rope_type" was originally "type" if hasattr(config, "rope_scaling") and config.rope_scaling is not None: self.rope_type = config.rope_scaling.get("rope_type", config.rope_scaling.get("type")) else: self.rope_type = "default" self.max_seq_len_cached = config.max_position_embeddings self.original_max_seq_len = config.max_position_embeddings self.config = config self.rope_init_fn = ROPE_INIT_FUNCTIONS[self.rope_type] inv_freq, self.attention_scaling = self.rope_init_fn(self.config, device) self.register_buffer("inv_freq", inv_freq, persistent=False) self.original_inv_freq = self.inv_freq def _dynamic_frequency_update(self, position_ids, device): """ dynamic RoPE layers should recompute `inv_freq` in the following situations: 1 - growing beyond the cached sequence length (allow scaling) 2 - the current sequence length is in the original scale (avoid losing precision with small sequences) """ seq_len = torch.max(position_ids) + 1 if seq_len > self.max_seq_len_cached: # growth inv_freq, self.attention_scaling = self.rope_init_fn(self.config, device, seq_len=seq_len) self.register_buffer("inv_freq", inv_freq, persistent=False) # TODO joao: may break with compilation self.max_seq_len_cached = seq_len if seq_len < self.original_max_seq_len and self.max_seq_len_cached > self.original_max_seq_len: # reset # This .to() is needed if the model has been moved to a device after being initialized (because # the buffer is automatically moved, but not the original copy) self.original_inv_freq = self.original_inv_freq.to(device) self.register_buffer("inv_freq", self.original_inv_freq, persistent=False) self.max_seq_len_cached = self.original_max_seq_len @torch.no_grad() def forward(self, x, position_ids): if "dynamic" in self.rope_type: self._dynamic_frequency_update(position_ids, device=x.device) # Core RoPE block inv_freq_expanded = self.inv_freq[None, :, None].float().expand(position_ids.shape[0], -1, 1) position_ids_expanded = position_ids[:, None, :].float() # Force float32 (see https://github.com/huggingface/transformers/pull/29285) device_type = x.device.type device_type = device_type if isinstance(device_type, str) and device_type != "mps" else "cpu" with torch.autocast(device_type=device_type, enabled=False): freqs = (inv_freq_expanded.float() @ position_ids_expanded.float()).transpose(1, 2) emb = torch.cat((freqs, freqs), dim=-1) cos = emb.cos() sin = emb.sin() # Advanced RoPE types (e.g. yarn) apply a post-processing scaling factor, equivalent to scaling attention cos = cos * self.attention_scaling sin = sin * self.attention_scaling return cos.to(dtype=x.dtype), sin.to(dtype=x.dtype) def rotate_half(x): """Rotates half the hidden dims of the input.""" x1 = x[..., : x.shape[-1] // 2] x2 = x[..., x.shape[-1] // 2 :] return torch.cat((-x2, x1), dim=-1) def apply_rotary_pos_emb(q, k, cos, sin, position_ids=None, unsqueeze_dim=1): """Applies Rotary Position Embedding to the query and key tensors. Args: q (`torch.Tensor`): The query tensor. k (`torch.Tensor`): The key tensor. cos (`torch.Tensor`): The cosine part of the rotary embedding. sin (`torch.Tensor`): The sine part of the rotary embedding. position_ids (`torch.Tensor`, *optional*): Deprecated and unused. unsqueeze_dim (`int`, *optional*, defaults to 1): The 'unsqueeze_dim' argument specifies the dimension along which to unsqueeze cos[position_ids] and sin[position_ids] so that they can be properly broadcasted to the dimensions of q and k. For example, note that cos[position_ids] and sin[position_ids] have the shape [batch_size, seq_len, head_dim]. Then, if q and k have the shape [batch_size, heads, seq_len, head_dim], then setting unsqueeze_dim=1 makes cos[position_ids] and sin[position_ids] broadcastable to the shapes of q and k. Similarly, if q and k have the shape [batch_size, seq_len, heads, head_dim], then set unsqueeze_dim=2. Returns: `tuple(torch.Tensor)` comprising of the query and key tensors rotated using the Rotary Position Embedding. """ cos = cos.unsqueeze(unsqueeze_dim) sin = sin.unsqueeze(unsqueeze_dim) q_embed = (q * cos) + (rotate_half(q) * sin) k_embed = (k * cos) + (rotate_half(k) * sin) return q_embed, k_embed class LlamaMLP(nn.Module): def __init__(self, config): super().__init__() self.config = config self.hidden_size = config.hidden_size self.intermediate_size = config.intermediate_size self.gate_proj = nn.Linear(self.hidden_size, self.intermediate_size, bias=config.mlp_bias) self.up_proj = nn.Linear(self.hidden_size, self.intermediate_size, bias=config.mlp_bias) self.down_proj = nn.Linear(self.intermediate_size, self.hidden_size, bias=config.mlp_bias) self.act_fn = ACT2FN[config.hidden_act] def forward(self, x): down_proj = self.down_proj(self.act_fn(self.gate_proj(x)) * self.up_proj(x)) return down_proj