INDEX

Explanations

mentions of "profile" and related terms indicating user identities or descriptions

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

reich

-0.15

rooms

-0.15

ivec

-0.15

oral

-0.15

ENO

-0.14

allele

-0.14

/tcp

-0.14

ivals

-0.14

etCode

-0.14

erm

-0.14

POSITIVE LOGITS

0.37

/profile

0.23

lla

0.20

tte

0.20

ty

0.19

matic

0.18

card

0.18

0.17

Pic

0.17

Pic

0.17

Activations Density 0.019%

mentions of "profile" and related terms indicating user identities or descriptions

No Comments

No Known Activations