INDEX

Explanations

phrases and concepts relating to accountability and oversight

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

New Auto-Interp

Top Features by Cosine Similarity

Comparing With LLAMA3.1-8B @ 30-llamascope-res-32k

Configuration

fnlp/Llama3_1-8B-Base-LXR-8x/Llama3_1-8B-Base-L30R-8x

Prompts (Dashboard)

24,576 prompts, 128 tokens each

Dataset (Dashboard)

cerebras/SlimPajama-627B

Features

32,768

Data Type

bfloat16

Hook Name

blocks.30.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

already

-0.16

Ð¾Ð±Ð°

-0.16

 somewhat

-0.16

enga

-0.15

ãģłãģĳ

-0.15

ipel

-0.15

_already

-0.15

-0.14

iring

-0.14

elo

-0.14

POSITIVE LOGITS

 whatsoever

0.52

 WHATSOEVER

0.49

 whatever

0.36

 nÃło

0.36

nor

0.34

whatever

0.31

Whatever

0.31

 other

0.29

 anywhere

0.29

 Whatever

0.28

Activations Density 0.402%

phrases and concepts relating to accountability and oversight

No Comments

No Known Activations