INDEX

Explanations

references to bibliographic entries or citations in texts

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

ç½®

-0.17

://%

-0.15

ationToken

-0.15

onis

-0.15

ģ

-0.15

ToOne

-0.15

allen

-0.14

çģ

-0.14

irable

-0.14

ÏĦÎ¿Î¹

-0.14

POSITIVE LOGITS

oteca

0.30

oth

0.29

ography

0.27

lio

0.25

ographies

0.25

ographic

0.24

ographical

0.24

ografia

0.23

otec

0.21

ograph

0.21

Activations Density 0.004%

references to bibliographic entries or citations in texts

No Comments

No Known Activations