INDEX

Explanations

references to meetings and collaborative training 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

okens

-0.15

entic

-0.15

ubyte

-0.14

 tempor

-0.14

iei

-0.14

icit

-0.14

æ¨Ĥ

-0.14

 Diploma

-0.14

 sÃ¶

-0.13

lico

-0.13

POSITIVE LOGITS

 meeting

0.59

 meetings

0.54

meeting

0.47

 Meeting

0.47

Meeting

0.41

meet

0.40

 Meetings

0.39

 meet

0.32

Meet

0.32

mee

0.32

Activations Density 0.199%

references to meetings and collaborative training activities

No Comments

No Known Activations