DeepSeek-V3.2-Exp · RoPE coefficients

RoPE coefficients

tensorfreqs_cis

Rotation coefficients for the current sequence, supplied to every block.

shape: $R$ [T positions, 32 pairs]

Sources

Contextual evidence from Generate RoPE coefficients

Rotary coefficients are precomputed once for the maximum sequence length, with YaRN frequency interpolation beyond the original 4,096 positions.

rope · deepseek-ai/DeepSeek-V3.2-Exp / inference/model.py
Pinned revision 87e509a2e5a1 ↗ · lines 334–402

Read code excerpt

    dim = args.qk_rope_head_dim
    seqlen = args.max_seq_len
    beta_fast = args.beta_fast
    beta_slow = args.beta_slow
    base = args.rope_theta
    factor = args.rope_factor

… lines 340–392 omitted …

    freqs = 1.0 / (base ** (torch.arange(0, dim, 2, dtype=torch.float32) / dim))
    if seqlen > args.original_seq_len:
        low, high = find_correction_range(beta_fast, beta_slow, dim, base, args.original_seq_len)
        smooth = 1 - linear_ramp_factor(low, high, dim // 2)
        freqs = freqs / factor * (1 - smooth) + freqs * smooth

    t = torch.arange(seqlen)
    freqs = torch.outer(t, freqs)
    freqs_cis = torch.polar(torch.ones_like(freqs), freqs)
    return freqs_cis

Contextual evidence from Generate RoPE coefficients

Reference defaults the configuration does not override: maximum sequence length, YaRN RoPE settings and the softmax scale correction.

model-args · deepseek-ai/DeepSeek-V3.2-Exp / inference/model.py
Pinned revision 87e509a2e5a1 ↗ · lines 56–86

Read code excerpt

    max_seq_len: int = 4096 * 4

… lines 57–80 omitted …

    original_seq_len: int = 4096
    rope_theta: float = 10000.0
    rope_factor: float = 40
    beta_fast: int = 32
    beta_slow: int = 1
    mscale: float = 1.

Contextual evidence from Transformer blocks

The model embeds tokens, applies all blocks with a causal mask, normalizes and projects to vocabulary logits with an untied head. The reference generator projects only the last position; training projects every position.

transformer · deepseek-ai/DeepSeek-V3.2-Exp / inference/model.py
Pinned revision 87e509a2e5a1 ↗ · lines 880–908

Read code excerpt

        self.embed = ParallelEmbedding(args.vocab_size, args.dim)
        self.layers = torch.nn.ModuleList()
        for layer_id in range(args.n_layers):
            self.layers.append(Block(layer_id, args))
        self.norm = RMSNorm(args.dim)
        # lm_head in the checkpoint is stored in bf16, while the parameter here is stored in fp32 for easier computation of logits later.
        self.head = ColumnParallelLinear(args.dim, args.vocab_size, dtype=torch.float32)
        self.register_buffer("freqs_cis", precompute_freqs_cis(args), persistent=False)

… lines 888–900 omitted …

        seqlen = tokens.size(1)
        freqs_cis = self.freqs_cis[start_pos:start_pos+seqlen]
        mask = torch.full((seqlen, seqlen), float("-inf"), device=tokens.device).triu_(1) if seqlen > 1 else None
        h, residual = self.embed(tokens), None
        for layer in self.layers:
            h, residual = layer(h, residual, start_pos, freqs_cis, mask)
        h, _ = self.norm(h, residual)
        logits = self.head(h[:, -1].float())

Contextual evidence from Transformer blocks

A block applies attention-norm, MLA, ffn-norm and the FFN; the first n_dense_layers blocks use the dense MLP, the rest MoE. Residual additions happen inside the next normalization.

block · deepseek-ai/DeepSeek-V3.2-Exp / inference/model.py
Pinned revision 87e509a2e5a1 ↗ · lines 826–851

Read code excerpt

        self.attn = MLA(args)
        self.ffn = MLP(args.dim, args.inter_dim) if layer_id < args.n_dense_layers else MoE(args)
        self.attn_norm = RMSNorm(args.dim)
        self.ffn_norm = RMSNorm(args.dim)

… lines 830–843 omitted …

        if residual is None:
            x, residual = self.attn_norm(x), x
        else:
            x, residual = self.attn_norm(x, residual)
        x = self.attn(x, start_pos, freqs_cis, mask)
        x, residual = self.ffn_norm(x, residual)
        x = self.ffn(x)
        return x, residual