INDEX

Explanations

concepts and discussions related to questions, beliefs, and issues for further exploration and analysis

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

New Auto-Interp

Top Features by Cosine Similarity

Comparing With LLAMA3.1-8B @ 21-llamascope-res-32k

Configuration

fnlp/Llama3_1-8B-Base-LXR-8x/Llama3_1-8B-Base-L21R-8x

Prompts (Dashboard)

24,576 prompts, 128 tokens each

Dataset (Dashboard)

cerebras/SlimPajama-627B

Features

32,768

Data Type

bfloat16

Hook Name

blocks.21.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

composed

-0.17

theless

-0.15

akin

-0.15

è¿Ļæł·çļĦ

-0.15

eyond

-0.14

terra

-0.14

/exec

-0.14

ÑĢÐ¾Ð²Ð¸Ñĩ

-0.14

linger

-0.13

such

-0.13

POSITIVE LOGITS

ä¹ĭä¸Ģ

0.20

/question

0.16

oid

0.15

NING

0.14

/framework

0.14

/questions

0.14

iner

0.14

Ð¾Ð¼ÐµÐ½

0.14

0.13

anos

0.13

Activations Density 0.208%

concepts and discussions related to questions, beliefs, and issues for further exploration and analysis

No Comments

No Known Activations