INDEX

Explanations

references to the concept of deception or lies

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

oki

-0.17

stown

-0.15

olian

-0.15

oded

-0.15

idan

-0.14

OME

-0.14

odable

-0.14

iei

-0.14

zyst

-0.14

ãĥĲãĥ¼

-0.14

POSITIVE LOGITS

uten

0.26

utenant

0.23

berman

0.23

chten

0.20

ê´Ģ

0.19

urance

0.16

eg

0.16

gth

0.16

apis

0.15

istrovstvÃŃ

0.15

Activations Density 0.015%

references to the concept of deception or lies

No Comments

No Known Activations