INDEX

Explanations

references to specific organizations, groups, or entities, particularly related to media and politics

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

ainter

-0.18

stadt

-0.18

Ã¥

-0.16

ponge

-0.16

Ð¾Ð´

-0.16

aper

-0.15

ark

-0.15

ugg

-0.15

 padded

-0.15

omor

-0.15

POSITIVE LOGITS

egasus

0.19

ultimate

0.19

=P

0.19

ricia

0.19

bilt

0.17

RIORITY

0.17

axe

0.16

jabi

0.16

ifer

0.16

iatric

0.15

Activations Density 0.847%

references to specific organizations, groups, or entities, particularly related to media and politics

No Comments

No Known Activations