INDEX

Explanations

phrases indicating personal beliefs or criticisms

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

New Auto-Interp

Top Features by Cosine Similarity

Comparing With LLAMA3.1-8B @ 9-llamascope-res-32k

Configuration

fnlp/Llama3_1-8B-Base-LXR-8x/Llama3_1-8B-Base-L9R-8x

Prompts (Dashboard)

24,576 prompts, 128 tokens each

Dataset (Dashboard)

cerebras/SlimPajama-627B

Features

32,768

Data Type

bfloat16

Hook Name

blocks.9.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

riere

-0.15

RIES

-0.15

-0.14

 sink

-0.14

andler

-0.14

vg

-0.14

ries

-0.14

á»Ŀ

-0.14

mar

-0.14

iteur

-0.14

POSITIVE LOGITS

aha

0.16

éĳ

0.15

amik

0.15

rÃ¼n

0.14

 Mods

0.14

 rÃ³wn

0.14

tridge

0.14

uner

0.14

Anc

0.14

urv

0.14

Activations Density 0.011%

phrases indicating personal beliefs or criticisms

No Comments

No Known Activations