INDEX

Explanations

phrases indicating requests or recommendations

oai_token-act-pair · gpt-4o-mini Triggered by @bot

New Auto-Interp

Top Features by Cosine Similarity

Comparing With LLAMA3.1-8B @ 25-llamascope-res-32k

Configuration

fnlp/Llama3_1-8B-Base-LXR-8x/Llama3_1-8B-Base-L25R-8x

Prompts (Dashboard)

24,576 prompts, 128 tokens each

Dataset (Dashboard)

cerebras/SlimPajama-627B

Features

32,768

Data Type

bfloat16

Hook Name

blocks.25.hook_resid_post

Hook Layer

Architecture

jumprelu

Context Size

1,024

Dataset

cerebras/SlimPajama-627B

Activation Function

relu

Embeds

IFrame

Link

Not in Any Lists

No Comments

Negative Logits

 ê²ĥìĿ´ëĭ¤

-0.20

thy

-0.20

 ÑĤÐµÐ±Ñı

-0.17

she

-0.17

 pudo

-0.17

Thy

-0.17

Ķëĭ¤

-0.16

ertia

-0.16

thy

-0.16

Ley

-0.16

POSITIVE LOGITS

 yourselves

0.36

 Ð¼Ð¾Ð¶ÐµÑĤÐµ

0.28

æĤ¨

0.28

ä½łä»¬

0.24

Ð¸ÑĤÐµ

0.23

prite

0.23

ite

0.22

ï¼ĮæĤ¨

0.22

ÑĥÐ¹ÑĤÐµ

0.21

iez

0.20

Activations Density 0.078%

phrases indicating requests or recommendations

No Comments

No Known Activations