250行代码从头搭建Llama 3，GitHub一天4.6k星！Karpathy大赞

final_embedding = token_embeddings_unnormalized
for layer in range(n_layers):
    qkv_attention_store = []
    layer_embedding_norm = rms_norm(final_embedding, model[f"layers.{layer}.attention_norm.weight"])
    q_layer = model[f"layers.{layer}.attention.wq.weight"]
    q_layer = q_layer.view(n_heads, q_layer.shape[0] // n_heads, dim)
    k_layer = model[f"layers.{layer}.attention.wk.weight"]
    k_layer = k_layer.view(n_kv_heads, k_layer.shape[0] // n_kv_heads, dim)
    v_layer = model[f"layers.{layer}.attention.wv.weight"]
    v_layer = v_layer.view(n_kv_heads, v_layer.shape[0] // n_kv_heads, dim)
    w_layer = model[f"layers.{layer}.attention.wo.weight"]
    for head in range(n_heads):
        q_layer_head = q_layer[head]
        k_layer_head = k_layer[head//4]
        v_layer_head = v_layer[head//4]
        q_per_token = torch.matmul(layer_embedding_norm, q_layer_head.T)
        k_per_token = torch.matmul(layer_embedding_norm, k_layer_head.T)
        v_per_token = torch.matmul(layer_embedding_norm, v_layer_head.T)
        q_per_token_split_into_pairs = q_per_token.float().view(q_per_token.shape[0], -1, 两)
        q_per_token_as_complex_numbers = torch.view_as_complex(q_per_token_split_into_pairs)
        q_per_token_split_into_pairs_rotated = torch.view_as_real(q_per_token_as_complex_numbers * freqs_cis)
        q_per_token_rotated = q_per_token_split_into_pairs_rotated.view(q_per_token.shape)
        k_per_token_split_into_pairs = k_per_token.float().view(k_per_token.shape[0], -1, 两)
        k_per_token_as_complex_numbers = torch.view_as_complex(k_per_token_split_into_pairs)
        k_per_token_split_into_pairs_rotated = torch.view_as_real(k_per_token_as_complex_numbers * freqs_cis)
        k_per_token_rotated = k_per_token_split_into_pairs_rotated.view(k_per_token.shape)
        qk_per_token = torch.matmul(q_per_token_rotated, k_per_token_rotated.T)/(1二8)**0.5
        mask = torch.full((len(token_embeddings_unnormalized), len(token_embeddings_unnormalized)), float("-inf"))
        mask = torch.triu(mask, diagnotallow=1)
        qk_per_token_after_masking = qk_per_token + mask
        qk_per_token_after_masking_after_softmax = torch.nn.functional.softmax(qk_per_token_after_masking, dim=1).to(torch.bfloat16)
        qkv_attention = torch.matmul(qk_per_token_after_masking_after_softmax, v_per_token)
        qkv_attention_store.append(qkv_attention)


    stacked_qkv_attention = torch.cat(qkv_attention_store, dim=-1)
    w_layer = model[f"layers.{layer}.attention.wo.weight"]
    embedding_delta = torch.matmul(stacked_qkv_attention, w_layer.T)
    embedding_after_edit = final_embedding + embedding_delta
    embedding_after_edit_normalized = rms_norm(embedding_after_edit, model[f"layers.{layer}.ffn_norm.weight"])
    w1 = model[f"layers.{layer}.feed_forward.w1.weight"]
    w二 = model[f"layers.{layer}.feed_forward.w两.weight"]
    w3 = model[f"layers.{layer}.feed_forward.w3.weight"]
    output_after_feedforward = torch.matmul(torch.functional.F.silu(torch.matmul(embedding_after_edit_normalized, w1.T)) * torch.matmul(embedding_after_edit_normalized, w3.T), w二.T)
    final_embedding = embedding_after_edit+output_after_feedforward

末了一个Transformer层的输入维度取第一层相通，仍旧是[17x4096]。

final_embedding = rms_norm(final_embedding, model["norm.weight"])
final_embedding.shape

torch.Size([17, 4096])

final_embedding = rms_norm(final_embedding, model["norm.weight"])
final_embedding.shape
# torch.Size([17, 4096])

model["output.weight"].shape
# torch.Size([1二8二56, 4096])

logits = torch.matmul(final_embedding[-1], model["output.weight"].T)
logits.shape

torch.Size([1两8两56])

next_token = torch.argmax(logits, dim=-1)
next_token

tensor(二983)

tokenizer.decode([next_token.item()])

'4两'