INDEX

Explanations

responses to yes/no questions or decisions

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

New Auto-Interp

Top Features by Cosine Similarity

Comparing With LLAMA3.1-8B @ 29-llamascope-res-32k

Configuration

fnlp/Llama3_1-8B-Base-LXR-8x/Llama3_1-8B-Base-L29R-8x

Prompts (Dashboard)

24,576 prompts, 128 tokens each

Dataset (Dashboard)

cerebras/SlimPajama-627B

Features

32,768

Data Type

bfloat16

Hook Name

blocks.29.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

ë¦¬ì§Ģ

-0.16

ieder

-0.15

.Apis

-0.15

gradation

-0.14

 Ä°b

-0.14

ÙĪÙĦÙĬ

-0.14

Î¬

-0.14

AtPath

-0.14

andom

-0.14

 defaultManager

-0.14

POSITIVE LOGITS

 Hell

0.27

NO

0.23

 hell

0.23

yes

0.23

res

0.22

 negative

0.21

 Maybe

0.20

Hell

0.20

yes

0.20

Yes

0.20

Activations Density 0.073%

responses to yes/no questions or decisions

No Comments

No Known Activations