INDEX

Explanations

references to stereotypes and biases in various contexts

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

New Auto-Interp

Top Features by Cosine Similarity

Comparing With LLAMA3.1-8B @ 9-llamascope-res-32k

Configuration

fnlp/Llama3_1-8B-Base-LXR-8x/Llama3_1-8B-Base-L9R-8x

Prompts (Dashboard)

24,576 prompts, 128 tokens each

Dataset (Dashboard)

cerebras/SlimPajama-627B

Features

32,768

Data Type

bfloat16

Hook Name

blocks.9.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

uu

-0.15

ç§Ģ

-0.15

asive

-0.15

lier

-0.15

liers

-0.15

cl

-0.15

 Grove

-0.15

aned

-0.15

urf

-0.14

 Integrity

-0.14

POSITIVE LOGITS

apse

0.14

isini

0.14

Fay

0.14

éĺħ

0.14

ç®

0.13

 Caps

0.13

\Entities

0.13

 HOLDERS

0.13

==>

0.13

Ð¸Ð½

0.13

Activations Density 0.044%

references to stereotypes and biases in various contexts

No Comments

No Known Activations