INDEX

Explanations

references to social interactions and events involving people

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

ENUM

-0.15

-0.14

onaut

-0.13

orrh

-0.13

 foresee

-0.13

adlo

-0.13

iento

-0.13

Fly

-0.12

ãģ£ãģ±

-0.12

åĵŃ

-0.12

POSITIVE LOGITS

 introduced

0.34

 introdu

0.34

 introduce

0.33

 introduction

0.31

 introducing

0.29

 Introduced

0.29

 introduces

0.28

 approach

0.26

 greeting

0.25

 shake

0.24

Activations Density 0.425%

references to social interactions and events involving people

No Comments

No Known Activations