INDEX

Explanations

instances of the word "simple" and related terms

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

New Auto-Interp

Top Features by Cosine Similarity

Comparing With LLAMA3.1-8B @ 16-llamascope-res-32k

Configuration

fnlp/Llama3_1-8B-Base-LXR-8x/Llama3_1-8B-Base-L16R-8x

Prompts (Dashboard)

24,576 prompts, 128 tokens each

Dataset (Dashboard)

cerebras/SlimPajama-627B

Features

32,768

Data Type

bfloat16

Hook Name

blocks.16.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

_sink

-0.15

ngr

-0.14

Ð³ÐµÐ½

-0.14

esimal

-0.14

lers

-0.14

ultimate

-0.14

zel

-0.14

è¾°

-0.13

lio

-0.13

ilm

-0.13

POSITIVE LOGITS

tons

0.34

ton

0.33

/simple

0.31

xes

0.26

-minded

0.24

/plain

0.23

TON

0.23

st

0.21

 minded

0.20

âĢĮØªØ±ÛĮÙĨ

0.20

Activations Density 0.039%

instances of the word "simple" and related terms

No Comments

No Known Activations