INDEX

Explanations

numerical data, specifically related to addresses and measurements

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

New Auto-Interp

Top Features by Cosine Similarity

Comparing With LLAMA3.1-8B @ 21-llamascope-res-32k

Configuration

fnlp/Llama3_1-8B-Base-LXR-8x/Llama3_1-8B-Base-L21R-8x

Prompts (Dashboard)

24,576 prompts, 128 tokens each

Dataset (Dashboard)

cerebras/SlimPajama-627B

Features

32,768

Data Type

bfloat16

Hook Name

blocks.21.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

lane

-0.18

angelo

-0.17

iard

-0.16

-0.15

erne

-0.15

æ´²

-0.15

-league

-0.15

unning

-0.14

Www

-0.14

ade

-0.14

POSITIVE LOGITS

ingly

0.17

ongs

0.16

reater

0.16

oog

0.16

rones

0.16

bots

0.15

warts

0.15

Äįi

0.15

iswa

0.15

arem

0.15

Activations Density 0.093%

numerical data, specifically related to addresses and measurements

No Comments

No Known Activations