INDEX

Explanations

terms related to superiority and inferiority

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

New Auto-Interp

Top Features by Cosine Similarity

Comparing With LLAMA3.1-8B @ 21-llamascope-res-32k

Configuration

fnlp/Llama3_1-8B-Base-LXR-8x/Llama3_1-8B-Base-L21R-8x

Prompts (Dashboard)

24,576 prompts, 128 tokens each

Dataset (Dashboard)

cerebras/SlimPajama-627B

Features

32,768

Data Type

bfloat16

Hook Name

blocks.21.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

ned

-0.19

ych

-0.18

away

-0.17

ting

-0.17

nel

-0.17

tings

-0.17

erson

-0.17

ness

-0.17

Ð°ÑĢÐ°Ð¼

-0.17

yn

-0.16

POSITIVE LOGITS

iors

0.26

ity

0.26

ities

0.22

iets

0.21

haps

0.20

-quality

0.20

ior

0.18

 Ð²ÑģÐµÐ³Ð¾

0.18

-most

0.17

ITY

0.16

Activations Density 0.013%

terms related to superiority and inferiority

No Comments

No Known Activations