INDEX

Explanations

mentions of categories within various contexts

oai_token-act-pair · gpt-4o-mini

Triggered by @bot

New Auto-Interp

Top Features by Cosine Similarity

Comparing With LLAMA3.1-8B @ 24-llamascope-res-32k

Configuration

fnlp/Llama3_1-8B-Base-LXR-8x/Llama3_1-8B-Base-L24R-8x

Prompts (Dashboard)

24,576 prompts, 128 tokens each

Dataset (Dashboard)

cerebras/SlimPajama-627B

Features

32,768

Data Type

bfloat16

Hook Name

blocks.24.hook_resid_post

Hook Layer

Architecture

jumprelu

Context Size

1,024

Dataset

cerebras/SlimPajama-627B

Activation Function

relu

Embeds

IFrame

Link

Negative Logits

ryo

-0.17

agers

-0.17

ery

-0.16

ÑģÑı

-0.16

ase

-0.16

pery

-0.15

vy

-0.15

enberg

-0.15

iversit

-0.15

ly

-0.15

POSITIVE LOGITS

Ã©gorie

0.20

åĪ¥

0.19

apult

0.18

etting

0.18

/class

0.18

/type

0.17

/sub

0.17

wide

0.17

alog

0.17

 Ø¨ÙĨØ¯ÛĮ

0.16

Activations Density 0.018%

mentions of categories within various contexts

No Known Activations