DeepSeek-V3.2-Exp · Mix and project

61 transformer blocks › 58 MoE blocks › Sparse latent attention › Sparse attention

Mix and project

learnedx · 589–607

Mix each head’s values by its weights, join the 128 heads and project back to 7,168 features.

O_i=\tilde O_i\,W_{UV,i}

\Delta_{\mathrm{attn}}=oW_{O}

All 4 steps

\tilde O_i=A_i\,\hat c^{KV}

O_i=\tilde O_i\,W_{UV,i}

o=\operatorname{reshape}(O)

\Delta_{\mathrm{attn}}=oW_{O}

in: $A$ [B batch, 128 heads, 1 query, 1001 keys] $\hat c^{KV}$ [B batch, 1001 cached positions, 512 kv latent]
out: $\Delta_{\mathrm{attn}}$ [B batch, 1 position, 7168 d]
parameters: 117,440,512 · × 58 MoE blocks = 6,811,549,696

Parameters

117,440,512 parameters per instance, across its learned components

Exact per-instance accounting
Component	Calculation	Parameters
Output projection	$16{,}384 \times 7{,}168$	117,440,512
Per-instance total		117,440,512

Across 58 independent MoE blocks: $117{,}440{,}512 \times 58 = 6{,}811{,}549{,}696$ parameters.

Derived from the pinned reference configuration and inference code. Checkpoint tensors have not been independently counted; FP8 block scales and the multi-token-prediction module are not counted.

Explanation

The output projection is the only step that mixes heads.

Codemodel.py 590–608

model.py 590–606

        else:                   # MQA decode
            if self.dequant_wkv_b is None and self.wkv_b.scale is not None:
                self.dequant_wkv_b = weight_dequant(self.wkv_b.weight, self.wkv_b.scale)
            wkv_b = self.wkv_b.weight if self.dequant_wkv_b is None else self.dequant_wkv_b
            wkv_b = wkv_b.view(self.n_local_heads, -1, self.kv_lora_rank)
            q_nope = torch.einsum("bshd,hdc->bshc", q_nope, wkv_b[:, :self.qk_nope_head_dim])
            scores = (torch.einsum("bshc,btc->bsht", q_nope, self.kv_cache[:bsz, :end_pos]) +
                      torch.einsum("bshr,btr->bsht", q_pe, self.pe_cache[:bsz, :end_pos])) * self.softmax_scale

            # indexer
            topk_indices = self.indexer(x, qr, start_pos, freqs_cis, mask)
            index_mask = torch.full((bsz, 1, end_pos), float("-inf"), device=x.device).scatter_(-1, topk_indices, 0)
            scores += index_mask.unsqueeze(2)

            scores = scores.softmax(dim=-1)
            x = torch.einsum("bsht,btc->bshc", scores, self.kv_cache[:bsz, :end_pos])
            x = torch.einsum("bshc,hdc->bshd", x, wkv_b[:, -self.v_head_dim:])

model.py 607–608

        x = self.wo(x.flatten(2))
        return x

Sources2

Direct evidence

When decoding one token, the implementation absorbs the key–value up-projection into queries and outputs and attends over the cached latent directly.

mla-decode · deepseek-ai/DeepSeek-V3.2-Exp / inference/model.py
Pinned revision 87e509a2e5a1 ↗ · lines 590–606

Read code excerpt

        else:                   # MQA decode
            if self.dequant_wkv_b is None and self.wkv_b.scale is not None:
                self.dequant_wkv_b = weight_dequant(self.wkv_b.weight, self.wkv_b.scale)
            wkv_b = self.wkv_b.weight if self.dequant_wkv_b is None else self.dequant_wkv_b
            wkv_b = wkv_b.view(self.n_local_heads, -1, self.kv_lora_rank)
            q_nope = torch.einsum("bshd,hdc->bshc", q_nope, wkv_b[:, :self.qk_nope_head_dim])
            scores = (torch.einsum("bshc,btc->bsht", q_nope, self.kv_cache[:bsz, :end_pos]) +
                      torch.einsum("bshr,btr->bsht", q_pe, self.pe_cache[:bsz, :end_pos])) * self.softmax_scale

            # indexer
            topk_indices = self.indexer(x, qr, start_pos, freqs_cis, mask)
            index_mask = torch.full((bsz, 1, end_pos), float("-inf"), device=x.device).scatter_(-1, topk_indices, 0)
            scores += index_mask.unsqueeze(2)

            scores = scores.softmax(dim=-1)
            x = torch.einsum("bsht,btc->bshc", scores, self.kv_cache[:bsz, :end_pos])
            x = torch.einsum("bshc,hdc->bshd", x, wkv_b[:, -self.v_head_dim:])

Direct evidence

Head outputs are flattened and projected back to the model width.

mla-output · deepseek-ai/DeepSeek-V3.2-Exp / inference/model.py
Pinned revision 87e509a2e5a1 ↗ · lines 607–608

Read code excerpt

        x = self.wo(x.flatten(2))
        return x