INDEX

Explanations

phrases containing the word "to" indicating purpose or direction

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

New Auto-Interp

Top Features by Cosine Similarity

Comparing With LLAMA3.1-8B @ 7-llamascope-res-32k

Configuration

fnlp/Llama3_1-8B-Base-LXR-8x/Llama3_1-8B-Base-L7R-8x

Prompts (Dashboard)

24,576 prompts, 128 tokens each

Dataset (Dashboard)

cerebras/SlimPajama-627B

Features

32,768

Data Type

bfloat16

Hook Name

blocks.7.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

adows

-0.16

-0.15

jt

-0.15

Bil

-0.15

mvc

-0.15

edom

-0.14

-sided

-0.14

Ð¾Ð±ÑĢÐ°Ð·

-0.14

xing

-0.14

utos

-0.14

POSITIVE LOGITS

 Î´ÎµÎ´Î¿

0.16

nel

0.15

inel

0.14

Porno

0.14

eken

0.14

ç²¾åĵģ

0.14

illis

0.14

VENTORY

0.14

ÙĦØ³ÙĦ

0.14

allen

0.14

Activations Density 0.034%

phrases containing the word "to" indicating purpose or direction

No Comments

No Known Activations