rohitg00/ai-engineering-from-scratch · error · ValueError

hidden must divide heads

Error message

hidden must divide heads

What it means

Error "hidden must divide heads" thrown in rohitg00/ai-engineering-from-scratch.

Source

Thrown at phases/19-capstone-projects/40-dpo-from-scratch/code/main.py:65

        ids = [self.INST_ID]
        ids.extend(prompt.encode("utf-8", errors="ignore"))
        ids.append(self.RESP_ID)
        return ids

    def encode_completion(self, completion: str) -> List[int]:
        return list(completion.encode("utf-8", errors="ignore"))


# ---------------------------------------------------------------------------
# TinyGPT
# ---------------------------------------------------------------------------


class CausalSelfAttention(nn.Module):
    def __init__(self, hidden: int, heads: int, max_len: int):
        super().__init__()
        if hidden % heads != 0:
            raise ValueError("hidden must divide heads")
        self.heads = heads
        self.head_dim = hidden // heads
        self.qkv = nn.Linear(hidden, hidden * 3, bias=False)
        self.out = nn.Linear(hidden, hidden, bias=False)
        mask = torch.tril(torch.ones(max_len, max_len, dtype=torch.bool))
        self.register_buffer("causal_mask", mask, persistent=False)

    def forward(self, x: torch.Tensor) -> torch.Tensor:
        B, T, D = x.shape
        qkv = self.qkv(x).view(B, T, 3, self.heads, self.head_dim).permute(2, 0, 3, 1, 4)
        q, k, v = qkv[0], qkv[1], qkv[2]
        att = (q @ k.transpose(-2, -1)) / math.sqrt(self.head_dim)
        causal = self.causal_mask[:T, :T].view(1, 1, T, T)
        att = att.masked_fill(~causal, float("-inf"))
        weights = F.softmax(att, dim=-1)
        ctx = (weights @ v).transpose(1, 2).contiguous().view(B, T, D)
        return self.out(ctx)

View on GitHub (pinned to 39ea8a1c6d)

When it happens

Trigger: Thrown at phases/19-capstone-projects/40-dpo-from-scratch/code/main.py:65 when the library encounters an invalid state.

Common situations: See trigger scenarios.

AI-assisted analysis of rohitg00/ai-engineering-from-scratch@39ea8a1c6d (2026-08-26). Data as JSON: /api/errors/ea167241e1e5bc98. Report an issue: GitHub.