INDEX

Explanations

specific character sequences that could represent names or titles

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

-0.16

uong

-0.14

uur

-0.14

Cog

-0.14

agli

-0.14

enderit

-0.14

Ð·Ð°Ð±

-0.14

 entirety

-0.13

umi

-0.13

utenberg

-0.13

POSITIVE LOGITS

baugh

0.16

itz

0.16

ky

0.15

orth

0.15

iman

0.14

acre

0.14

by

0.14

åĬ¿

0.14

 Weather

0.14

Ð¼Ð¸Ð½Ð¸ÑģÑĤÑĢÐ°

0.14

Activations Density 0.240%

specific character sequences that could represent names or titles

No Comments

No Known Activations