INDEX

Explanations

phrases indicating coordination or connection between elements

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

yre

-0.13

ë·

-0.12

UDA

-0.12

 indem

-0.12

 Ø¯Ø§Ø®

-0.12

retty

-0.12

ulfill

-0.12

utton

-0.12

ieurs

-0.12

addtogroup

-0.12

POSITIVE LOGITS

how

0.60

 what

0.57

why

0.46

what

0.46

how

0.44

 where

0.41

 whom

0.38

who

0.37

How

0.35

 What

0.34

Activations Density 0.120%

phrases indicating coordination or connection between elements

No Comments

No Known Activations