INDEX

Explanations

references to mitigation strategies or the acronym "MIT."

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

New Auto-Interp

Top Features by Cosine Similarity

Comparing With LLAMA3.1-8B @ 27-llamascope-res-32k

Configuration

fnlp/Llama3_1-8B-Base-LXR-8x/Llama3_1-8B-Base-L27R-8x

Prompts (Dashboard)

24,576 prompts, 128 tokens each

Dataset (Dashboard)

cerebras/SlimPajama-627B

Features

32,768

Data Type

bfloat16

Hook Name

blocks.27.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

yms

-0.16

ÐµÐ½Ð½Ñı

-0.16

hetto

-0.15

ãĥĩãĤ£ãĤ¢

-0.15

athan

-0.14

orian

-0.14

ä»°

-0.14

ities

-0.14

elier

-0.14

vrier

-0.14

POSITIVE LOGITS

ochond

0.31

igation

0.28

ubishi

0.23

igated

0.23

igate

0.22

ford

0.22

osis

0.20

IGATION

0.20

igator

0.18

otic

0.18

Activations Density 0.010%

references to mitigation strategies or the acronym "MIT."

No Comments

No Known Activations