INDEX

Explanations

references to research articles, including their sources and metadata

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

ael

-0.15

à¸±à¸ĩà¸ģ

-0.15

á»ī

-0.14

endl

-0.14

Ä©

-0.14

 defaultMessage

-0.14

é«ĺåº¦

-0.14

andalone

-0.13

ende

-0.13

Ele

-0.13

POSITIVE LOGITS

itch

0.16

aken

0.16

éĢĶ

0.15

 Ð±Ð°Ð³Ð°ÑĤ

0.15

kea

0.15

ITCH

0.15

ake

0.15

quin

0.14

akin

0.14

cken

0.14

Activations Density 0.167%

references to research articles, including their sources and metadata

No Comments

No Known Activations