INDEX

Explanations

mentions of products and films

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

New Auto-Interp

Top Features by Cosine Similarity

Comparing With LLAMA3.1-8B @ 12-llamascope-res-32k

Configuration

fnlp/Llama3_1-8B-Base-LXR-8x/Llama3_1-8B-Base-L12R-8x

Prompts (Dashboard)

24,576 prompts, 128 tokens each

Dataset (Dashboard)

cerebras/SlimPajama-627B

Features

32,768

Data Type

bfloat16

Hook Name

blocks.12.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

ills

-0.18

ollo

-0.17

holders

-0.15

UNG

-0.15

 real

-0.14

ãĥıãĤ¤

-0.14

iber

-0.14

ÑĸÐ»Ñĸ

-0.14

linger

-0.14

lo

-0.14

POSITIVE LOGITS

Ð¾Ð·Ð²

0.16

porno

0.14

DMI

0.14

mÄĽ

0.14

ctype

0.14

inery

0.13

DRV

0.13

 YÃ¶n

0.13

avras

0.13

Isl

0.13

Activations Density 0.011%

mentions of products and films

No Comments

No Known Activations