INDEX

Explanations

references to spiritual or religious concepts

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

New Auto-Interp

Top Features by Cosine Similarity

Comparing With LLAMA3.1-8B @ 26-llamascope-res-32k

Configuration

fnlp/Llama3_1-8B-Base-LXR-8x/Llama3_1-8B-Base-L26R-8x

Prompts (Dashboard)

24,576 prompts, 128 tokens each

Dataset (Dashboard)

cerebras/SlimPajama-627B

Features

32,768

Data Type

bfloat16

Hook Name

blocks.26.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

_CY

-0.16

ovny

-0.15

foy

-0.14

iasi

-0.14

Ã¥n

-0.14

á»Ļ

-0.14

 Sant

-0.14

mÃ¡n

-0.14

oksen

-0.14

Eck

-0.14

POSITIVE LOGITS

 sting

0.16

Unchecked

0.15

 Parallel

0.14

aru

0.14

Parallel

0.14

 Jeho

0.14

 Paul

0.14

tn

0.14

pro

0.13

rieg

0.13

Activations Density 0.052%

references to spiritual or religious concepts

No Comments

No Known Activations