INDEX

Explanations

parentheses and numerical representations, possibly indicating data or thematic categorization

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

New Auto-Interp

Top Features by Cosine Similarity

Comparing With LLAMA3.1-8B @ 4-llamascope-res-32k

Configuration

fnlp/Llama3_1-8B-Base-LXR-8x/Llama3_1-8B-Base-L4R-8x

Prompts (Dashboard)

24,576 prompts, 128 tokens each

Dataset (Dashboard)

cerebras/SlimPajama-627B

Features

32,768

Data Type

bfloat16

Hook Name

blocks.4.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

enge

-0.17

eed

-0.17

eam

-0.15

ÎºÎ¿Ïį

-0.14

ModelProperty

-0.14

caffold

-0.14

à¸²à¸£à¸°

-0.14

ç²

-0.14

earing

-0.14

KO

-0.14

POSITIVE LOGITS

agma

0.16

URA

0.16

ijo

0.15

olla

0.15

ETF

0.14

ÑħÐ¸

0.14

_MAIL

0.13

omo

0.13

shall

0.13

Activations Density 0.083%

parentheses and numerical representations, possibly indicating data or thematic categorization

No Comments

No Known Activations