INDEX

Explanations

expressions indicating absence or loss

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

çĬ¬

-0.15

à¸§à¸£

-0.15

Č↵

-0.14

Ø§Ø³Ø§Ø³

-0.14

Ð¸Ð»ÑĮ

-0.14

antro

-0.14

lag

-0.13

@{

-0.13

xies

-0.13

azzo

-0.13

POSITIVE LOGITS

 khá»ıi

0.19

adar

0.16

veau

0.15

ukt

0.15

edom

0.15

inate

0.15

 trace

0.15

 forever

0.14

gone

0.14

Activations Density 0.054%

expressions indicating absence or loss

No Comments

No Known Activations