INDEX

Explanations

references to specific academic articles and their citation details

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

New Auto-Interp

Top Features by Cosine Similarity

Comparing With LLAMA3.1-8B @ 13-llamascope-res-32k

Configuration

fnlp/Llama3_1-8B-Base-LXR-8x/Llama3_1-8B-Base-L13R-8x

Prompts (Dashboard)

24,576 prompts, 128 tokens each

Dataset (Dashboard)

cerebras/SlimPajama-627B

Features

32,768

Data Type

bfloat16

Hook Name

blocks.13.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

ĵåĲį

-0.16

 Dodd

-0.15

ç·Ĵ

-0.15

tain

-0.14

lix

-0.14

Nov

-0.14

ause

-0.14

hawks

-0.13

icket

-0.13

crud

-0.13

POSITIVE LOGITS

AAC

0.15

iyim

0.15

insula

0.14

Ð°ÑĢÑĩ

0.14

Ø§Ø±Ø©

0.14

ieber

0.14

ÙģØ§ÙĤ

0.14

aan

0.14

Î¹Î´

0.13

ibo

0.13

Activations Density 0.016%

references to specific academic articles and their citation details

No Comments

No Known Activations