INDEX

Explanations

tokens or phrases associated with social or ethical discussions and specific content features

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

New Auto-Interp

Top Features by Cosine Similarity

Comparing With LLAMA3.1-8B @ 31-llamascope-res-32k

Configuration

fnlp/Llama3_1-8B-Base-LXR-8x/Llama3_1-8B-Base-L31R-8x

Prompts (Dashboard)

24,576 prompts, 128 tokens each

Dataset (Dashboard)

cerebras/SlimPajama-627B

Features

32,768

Data Type

bfloat16

Hook Name

blocks.31.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

inish

-0.15

tÄĽ

-0.13

icaret

-0.13

erce

-0.13

reon

-0.13

iginal

-0.13

igin

-0.13

 yeter

-0.13

okt

-0.13

OSC

-0.12

POSITIVE LOGITS

or

0.24

 possibly

0.20

 probably

0.19

our

0.18

 perhaps

0.18

probably

0.17

 Possibly

0.17

adÃŃ

0.17

æĪĸèĢħ

0.17

 Probably

0.17

Activations Density 0.023%

tokens or phrases associated with social or ethical discussions and specific content features

No Comments

No Known Activations