INDEX

Explanations

conversational phrases and references to personal experiences or preferences

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

res

-0.18

iber

-0.16

avo

-0.15

 Buch

-0.15

 Hatch

-0.15

 decom

-0.15

rese

-0.15

íķ©

-0.14

CNC

-0.14

Gro

-0.14

POSITIVE LOGITS

à¸¥à¸ĩ

0.16

'&&

0.15

oldur

0.15

EMY

0.15

Ð±Ð°Ñģ

0.15

TRL

0.15

 defaultProps

0.14

ogui

0.14

á»ģn

0.14

 Replies

0.14

Activations Density 0.077%

conversational phrases and references to personal experiences or preferences

No Comments

No Known Activations