INDEX

Explanations

references to brands and registered trademarks

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

New Auto-Interp

Top Features by Cosine Similarity

Comparing With LLAMA3.1-8B @ 13-llamascope-res-32k

Configuration

fnlp/Llama3_1-8B-Base-LXR-8x/Llama3_1-8B-Base-L13R-8x

Prompts (Dashboard)

24,576 prompts, 128 tokens each

Dataset (Dashboard)

cerebras/SlimPajama-627B

Features

32,768

Data Type

bfloat16

Hook Name

blocks.13.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

ob

-0.16

ag

-0.15

ãĥ¡ãĥ©

-0.15

bars

-0.14

agrant

-0.14

ÑĢÐ¾ÑģÑĤÐ¾

-0.14

roat

-0.14

åĻ

-0.14

 hinges

-0.14

hend

-0.14

POSITIVE LOGITS

ahr

0.16

GetMethod

0.15

eteria

0.15

@ResponseBody

0.14

__;↵

0.14

ekli

0.14

Ø§Ø¡

0.14

Ø§Ø²Ùĩ

0.14

aira

0.13

lectron

0.13

Activations Density 0.012%

references to brands and registered trademarks

No Comments

No Known Activations