Implement a KV cache + incremental single-token decode loop for a small transformer

Medium

Attention Efficiency & Long Context2 / 5

Chapter 032 / 5

Attention Efficiency & Long Context

Implement a KV cache + incremental single-token decode loop for a small transformer

Medium

Target interface

class CachedAttention

Implement the function/class skeleton in the editor. Any correct approach is accepted.

Hints0 / 2

Reference solution

Your own code stays in the editor.

solution.pyPython

Local draft

import torch
import torch.nn as nn
import torch.nn.functional as F

class CachedAttention(nn.Module):

    def __init__(self, d_model, n_heads):
        raise NotImplementedError

    def forward(self, x, cache=None):
        raise NotImplementedError

@torch.no_grad()
def decode(model, embed, unembed, prompt_ids, n_new):
    raise NotImplementedError

⌘/Ctrl↵to submit

AI review

No review yet

Submit your solution for a verdict, a score out of 100 and concrete feedback. Any correct approach passes.

Chapter 032 / 5

Attention Efficiency & Long Context

Implement a KV cache + incremental single-token decode loop for a small transformer

Medium

Target interface

class CachedAttention

Implement the function/class skeleton in the editor. Any correct approach is accepted.

Hints0 / 2

Reference solution

Your own code stays in the editor.