optimum/exporters/openvino/model_patcher.py [1862:1895]:
- - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - -
        cos = cos[position_ids].unsqueeze(1)  # [bs, 1, seq_len, dim]
        sin = sin[position_ids].unsqueeze(1)  # [bs, 1, seq_len, dim]
        q_embed = (q * cos) + (rotate_half(q) * sin)
        k_embed = (k * cos) + (rotate_half(k) * sin)
        return q_embed, k_embed

    if output_attentions:
        return self._orig_forward(
            hidden_states, attention_mask, position_ids, past_key_value, output_attentions, use_cache
        )
    bsz, q_len, _ = hidden_states.size()

    query_states = self.q_proj(hidden_states).view(bsz, q_len, self.num_heads, self.head_dim).transpose(1, 2)
    key_states = self.k_proj(hidden_states).view(bsz, q_len, self.num_heads, self.head_dim).transpose(1, 2)
    value_states = self.v_proj(hidden_states).view(bsz, q_len, self.num_heads, self.head_dim).transpose(1, 2)

    kv_seq_len = key_states.shape[-2]
    if past_key_value is not None:
        kv_seq_len += past_key_value[0].shape[-2]
    cos, sin = self.rotary_emb(value_states, seq_len=kv_seq_len)
    query_states, key_states = apply_rotary_pos_emb(query_states, key_states, cos, sin, position_ids)
    # [bsz, nh, t, hd]

    if past_key_value is not None:
        # reuse k, v, self_attention
        key_states = torch.cat([past_key_value[0], key_states], dim=2)
        value_states = torch.cat([past_key_value[1], value_states], dim=2)

    past_key_value = (key_states, value_states) if use_cache else None

    attn_output = torch.nn.functional.scaled_dot_product_attention(
        query_states, key_states, value_states, attention_mask, scale=(1 / math.sqrt(self.head_dim))
    )
    attn_weights = None
- - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - -



optimum/exporters/openvino/model_patcher.py [1921:1953]:
- - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - -
        cos = cos[position_ids].unsqueeze(1)
        sin = sin[position_ids].unsqueeze(1)
        q_embed = (q * cos) + (rotate_half(q) * sin)
        k_embed = (k * cos) + (rotate_half(k) * sin)
        return q_embed, k_embed

    if output_attentions:
        return self._orig_forward(
            hidden_states, attention_mask, position_ids, past_key_value, output_attentions, use_cache
        )

    bsz, q_len, _ = hidden_states.size()
    query_states = self.q_proj(hidden_states).view(bsz, q_len, self.num_heads, self.head_dim).transpose(1, 2)
    key_states = self.k_proj(hidden_states).view(bsz, q_len, self.num_heads, self.head_dim).transpose(1, 2)
    value_states = self.v_proj(hidden_states).view(bsz, q_len, self.num_heads, self.head_dim).transpose(1, 2)
    kv_seq_len = key_states.shape[-2]
    if past_key_value is not None:
        kv_seq_len += past_key_value[0].shape[-2]

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

    if past_key_value is not None:
        # reuse k, v, self_attention
        key_states = torch.cat([past_key_value[0], key_states], dim=2)
        value_states = torch.cat([past_key_value[1], value_states], dim=2)

    past_key_value = (key_states, value_states) if use_cache else None

    attn_output = torch.nn.functional.scaled_dot_product_attention(
        query_states, key_states, value_states, attention_mask, scale=(1 / math.sqrt(self.head_dim))
    )
    attn_weights = None
- - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - -