INDEX

Explanations

references to community and inclusivity

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

emand

-0.18

isas

-0.16

eyh

-0.15

 ÄĳÃ²i

-0.15

 demanding

-0.15

abor

-0.15

aar

-0.14

quential

-0.14

cant

-0.14

oz

-0.14

POSITIVE LOGITS

 encouraged

0.42

 invited

0.37

 welcome

0.36

enc

0.32

ENC

0.32

Enc

0.31

welcome

0.30

 strongly

0.29

 encourage

0.28

 encour

0.28

Activations Density 0.072%

references to community and inclusivity

No Comments

No Known Activations