INDEX

Explanations

references to participation and engagement in programs or activities

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

New Auto-Interp

Top Features by Cosine Similarity

Comparing With LLAMA3.1-8B @ 25-llamascope-res-32k

Configuration

fnlp/Llama3_1-8B-Base-LXR-8x/Llama3_1-8B-Base-L25R-8x

Prompts (Dashboard)

24,576 prompts, 128 tokens each

Dataset (Dashboard)

cerebras/SlimPajama-627B

Features

32,768

Data Type

bfloat16

Hook Name

blocks.25.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

itou

-0.16

á»©

-0.15

ç¶ļ

-0.15

cho

-0.14

izik

-0.14

ÅĪ

-0.14

Ñĳ

-0.14

ÐµÐºÐ°ÑĢ

-0.14

anson

-0.13

udden

-0.13

POSITIVE LOGITS

 walk

0.54

 leave

0.47

 walked

0.47

 walks

0.46

 leaving

0.45

 walking

0.45

 Walk

0.44

walk

0.44

 Leave

0.41

Walk

0.41

Activations Density 0.168%

references to participation and engagement in programs or activities

No Comments

No Known Activations