INDEX

Explanations

phrases indicating tasks, laws, and comparisons involving people's roles, performance, and accomplishments

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

New Auto-Interp

Top Features by Cosine Similarity

Comparing With LLAMA3.1-8B @ 16-llamascope-res-32k

Configuration

fnlp/Llama3_1-8B-Base-LXR-8x/Llama3_1-8B-Base-L16R-8x

Prompts (Dashboard)

24,576 prompts, 128 tokens each

Dataset (Dashboard)

cerebras/SlimPajama-627B

Features

32,768

Data Type

bfloat16

Hook Name

blocks.16.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

 Forbes

-0.17

 somehow

-0.16

ICODE

-0.15

vil

-0.15

bourne

-0.15

å¢

-0.14

 often

-0.14

ario

-0.14

 Cran

-0.14

vely

-0.14

POSITIVE LOGITS

 nÃło

0.23

 except

0.21

except

0.20

 ÐºÑĢÐ¾Ð¼Ðµ

0.18

 ever

0.18

 Except

0.18

 EVER

0.15

thood

0.15

Except

0.15

EDIUM

0.15

Activations Density 0.131%

phrases indicating tasks, laws, and comparisons involving people's roles, performance, and accomplishments

No Comments

No Known Activations