INDEX

Explanations

references to the concept of "fool" or foolishness

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

New Auto-Interp

Top Features by Cosine Similarity

Comparing With LLAMA3.1-8B @ 24-llamascope-res-32k

Configuration

fnlp/Llama3_1-8B-Base-LXR-8x/Llama3_1-8B-Base-L24R-8x

Prompts (Dashboard)

24,576 prompts, 128 tokens each

Dataset (Dashboard)

cerebras/SlimPajama-627B

Features

32,768

Data Type

bfloat16

Hook Name

blocks.24.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

ello

-0.18

shire

-0.16

 Miracle

-0.16

etta

-0.16

zing

-0.15

vt

-0.15

çĽĸ

-0.15

oga

-0.14

wick

-0.14

SURE

-0.14

POSITIVE LOGITS

hard

0.29

proof

0.28

ishly

0.25

ery

0.22

-proof

0.20

osoph

0.19

oose

0.18

ardy

0.18

ERY

0.18

fo

0.17

Activations Density 0.012%

references to the concept of "fool" or foolishness

No Comments

No Known Activations