INDEX

Explanations

special characters or symbols, primarily from various language scripts and emoji-like characters

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

New Auto-Interp

Top Features by Cosine Similarity

Comparing With LLAMA3.1-8B @ 23-llamascope-res-32k

Configuration

fnlp/Llama3_1-8B-Base-LXR-8x/Llama3_1-8B-Base-L23R-8x

Prompts (Dashboard)

24,576 prompts, 128 tokens each

Dataset (Dashboard)

cerebras/SlimPajama-627B

Features

32,768

Data Type

bfloat16

Hook Name

blocks.23.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

habi

-0.17

 cult

-0.17

Ä¢

-0.16

inite

-0.16

 Levin

-0.15

 Cres

-0.15

akk

-0.14

loor

-0.14

ÅĻev

-0.14

obs

-0.14

POSITIVE LOGITS

Ã¼ÅŁ

0.17

activated

0.15

olley

0.15

 chÃŃ

0.14

Ð¶Ñĸ

0.14

Ã¨

0.14

 Ð¸ÑģÑħÐ¾Ð´

0.14

rÃ³

0.14

 mÃ³

0.14

zych

0.13

Activations Density 0.005%

special characters or symbols, primarily from various language scripts and emoji-like characters

No Comments

No Known Activations

special characters or symbols, primarily from various language scripts and emoji-like characters

No Comments

No Known Activations