INDEX

Explanations

references to bibliographic citations or related terms

oai_token-act-pair · gpt-4o-mini

Triggered by @bot

New Auto-Interp

Top Features by Cosine Similarity

Comparing With LLAMA3.1-8B @ 15-llamascope-res-32k

Configuration

fnlp/Llama3_1-8B-Base-LXR-8x/Llama3_1-8B-Base-L15R-8x

Prompts (Dashboard)

24,576 prompts, 128 tokens each

Dataset (Dashboard)

cerebras/SlimPajama-627B

Features

32,768

Data Type

bfloat16

Hook Name

blocks.15.hook_resid_post

Hook Layer

Architecture

jumprelu

Context Size

1,024

Dataset

cerebras/SlimPajama-627B

Activation Function

relu

Embeds

IFrame

Link

Negative Logits

ellar

-0.17

iert

-0.16

lew

-0.15

rie

-0.15

cliffe

-0.14

rames

-0.14

ehler

-0.14

lied

-0.13

edImage

-0.13

lers

-0.13

POSITIVE LOGITS

iminal

0.15

à¸Ńà¸§

0.14

tement

0.14

 lean

0.14

 guard

0.14

.shiro

0.14

aille

0.14

anning

0.14

atches

0.14

sterol

0.14

Activations Density 0.068%

references to bibliographic citations or related terms

No Known Activations