INDEX

Explanations

dates and months mentioned in the text

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

New Auto-Interp

Top Features by Cosine Similarity

Comparing With LLAMA3.1-8B @ 11-llamascope-res-32k

Configuration

fnlp/Llama3_1-8B-Base-LXR-8x/Llama3_1-8B-Base-L11R-8x

Prompts (Dashboard)

24,576 prompts, 128 tokens each

Dataset (Dashboard)

cerebras/SlimPajama-627B

Features

32,768

Data Type

bfloat16

Hook Name

blocks.11.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

ond

-0.16

udo

-0.16

sure

-0.15

-Pro

-0.14

ÑĸÐ»ÑĮ

-0.14

arius

-0.14

chan

-0.14

PRO

-0.14

 darling

-0.14

ÄŁinden

-0.13

POSITIVE LOGITS

ainter

0.14

pent

0.14

ibe

0.14

ijo

0.14

inton

0.13

eph

0.13

igin

0.13

ango

0.13

ÑĢÐ°Ð´Ð¸

0.13

é§

0.13

Activations Density 0.017%

dates and months mentioned in the text

No Comments

No Known Activations