Kimi K2

def apply_rotary_pos_emb(q, k, cos, sin, position_ids, 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`):
            The position indices of the tokens corresponding to the query and key tensors. For example, this can be
            used to pass offsetted position ids when working with a KV-cache.
        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[position_ids].unsqueeze(unsqueeze_dim)
    sin = sin[position_ids].unsqueeze(unsqueeze_dim)

    b, h, s, d = q.shape
    q = q.view(b, h, s, d // 2, 2).transpose(4, 3).reshape(b, h, s, d)

    b, h, s, d = k.shape
    k = k.view(b, h, s, d // 2, 2).transpose(4, 3).reshape(b, h, s, d)

    q_embed = (q * cos) + (rotate_half(q) * sin)
    k_embed = (k * cos) + (rotate_half(k) * sin)
    return q_embed, k_embed

def yarn_find_correction_dim(
    num_rotations, dim, base=10000, max_position_embeddings=2048
):
    return (dim * math.log(max_position_embeddings / (num_rotations * 2 * math.pi))) / (
        2 * math.log(base)
    )


# Find dim range bounds based on rotations
def yarn_find_correction_range(
    low_rot, high_rot, dim, base=10000, max_position_embeddings=2048
):
    low = math.floor(
        yarn_find_correction_dim(low_rot, dim, base, max_position_embeddings)
    )
    high = math.ceil(
        yarn_find_correction_dim(high_rot, dim, base, max_position_embeddings)
    )
    return max(low, 0), min(high, dim - 1)  # Clamp values just in case


def yarn_get_mscale(scale=1, mscale=1):
    if scale <= 1:
        return 1.0
    return 0.1 * mscale * math.log(scale) + 1.0


def yarn_linear_ramp_mask(min, max, dim):
    if min == max:
        max += 0.001  # Prevent singularity

    linear_func = (torch.arange(dim, dtype=torch.float32) - min) / (max - min)
    ramp_func = torch.clamp(linear_func, 0, 1)
    return ramp_func

… lines 260–284 omitted …

    def _set_cos_sin_cache(self, seq_len, device, dtype):
        self.max_seq_len_cached = seq_len
        dim = self.dim

        freq_extra = 1.0 / (
            self.base
            ** (torch.arange(0, dim, 2, dtype=torch.float32, device=device) / dim)
        )
        freq_inter = 1.0 / (
            self.scaling_factor
            * self.base
            ** (torch.arange(0, dim, 2, dtype=torch.float32, device=device) / dim)
        )

        low, high = yarn_find_correction_range(
            self.beta_fast,
            self.beta_slow,
            dim,
            self.base,
            self.original_max_position_embeddings,
        )
        inv_freq_mask = 1.0 - yarn_linear_ramp_mask(low, high, dim // 2).to(
            device=device, dtype=torch.float32
        )
        inv_freq = freq_inter * (1 - inv_freq_mask) + freq_extra * inv_freq_mask
        self.register_buffer("inv_freq", inv_freq, persistent=False)

        t = torch.arange(seq_len, device=device, dtype=torch.float32)

        freqs = torch.outer(t, inv_freq)

        _mscale = float(
            yarn_get_mscale(self.scaling_factor, self.mscale)
            / yarn_get_mscale(self.scaling_factor, self.mscale_all_dim)
        )

        emb = torch.cat((freqs, freqs), dim=-1)
        self.register_buffer(
            "cos_cached", (emb.cos() * _mscale).to(dtype), persistent=False
        )
        self.register_buffer(
            "sin_cached", (emb.sin() * _mscale).to(dtype), persistent=False
        )

        cos, sin = self.rotary_emb(value_states, seq_len=kv_seq_len)

        q_pe, k_pe = apply_rotary_pos_emb(q_pe, k_pe, cos, sin, position_ids)

        query_states = k_pe.new_empty(bsz, self.num_heads, q_len, self.q_head_dim)
        query_states[:, :, :, : self.qk_nope_head_dim] = q_nope
        query_states[:, :, :, self.qk_nope_head_dim :] = q_pe

        key_states = k_pe.new_empty(bsz, self.num_heads, q_len, self.q_head_dim)
        key_states[:, :, :, : self.qk_nope_head_dim] = k_nope
        key_states[:, :, :, self.qk_nope_head_dim :] = k_pe
        if past_key_value is not None:
            cache_kwargs = {"sin": sin, "cos": cos}  # Specific to RoPE models
            key_states, value_states = past_key_value.update(
                key_states, value_states, self.layer_idx, cache_kwargs
            )

        attn_weights = (
            torch.matmul(query_states, key_states.transpose(2, 3)) * self.softmax_scale
        )

        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()}"
            )
        assert attention_mask is not None
        if attention_mask is not None:
            if attention_mask.size() != (bsz, 1, q_len, kv_seq_len):
                raise ValueError(
                    f"Attention mask should be of size {(bsz, 1, q_len, kv_seq_len)}, but is {attention_mask.size()}"
                )
            attn_weights = attn_weights + attention_mask

        # upcast attention to fp32
        attn_weights = nn.functional.softmax(
            attn_weights, dim=-1, dtype=torch.float32
        ).to(query_states.dtype)
        attn_weights = nn.functional.dropout(
            attn_weights, p=self.attention_dropout, training=self.training
        )
        attn_output = torch.matmul(attn_weights, value_states)