INDEX

Explanations

words and phrases associated with descriptions of processes and qualities

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

New Auto-Interp

Top Features by Cosine Similarity

Comparing With LLAMA3.1-8B @ 18-llamascope-res-32k

Configuration

fnlp/Llama3_1-8B-Base-LXR-8x/Llama3_1-8B-Base-L18R-8x

Prompts (Dashboard)

24,576 prompts, 128 tokens each

Dataset (Dashboard)

cerebras/SlimPajama-627B

Features

32,768

Data Type

bfloat16

Hook Name

blocks.18.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

imi

-0.15

unan

-0.15

bul

-0.15

zdy

-0.14

ëŀĢ

-0.14

illa

-0.14

esses

-0.14

æ£ļ

-0.14

inka

-0.13

RAL

-0.13

POSITIVE LOGITS

ãģ§ãģ®

0.18

 whence

0.17

 oÃ¹

0.16

APS

0.15

 waar

0.15

 manner

0.15

 where

0.15

pps

0.14

luv

0.14

Activations Density 0.194%

words and phrases associated with descriptions of processes and qualities

No Comments

No Known Activations