INDEX

Explanations

mentions of "courtesy" or related acknowledgments in the text

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

New Auto-Interp

Top Features by Cosine Similarity

Comparing With LLAMA3.1-8B @ 20-llamascope-res-32k

Configuration

fnlp/Llama3_1-8B-Base-LXR-8x/Llama3_1-8B-Base-L20R-8x

Prompts (Dashboard)

24,576 prompts, 128 tokens each

Dataset (Dashboard)

cerebras/SlimPajama-627B

Features

32,768

Data Type

bfloat16

Hook Name

blocks.20.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

hek

-0.17

pg

-0.17

ito

-0.17

stad

-0.17

ps

-0.16

ras

-0.15

quel

-0.15

wers

-0.14

lein

-0.14

athon

-0.14

POSITIVE LOGITS

ably

0.17

of

0.17

ately

0.16

ously

0.16

tesy

0.15

Ã±as

0.15

SHIP

0.14

inely

0.14

sez

0.14

tainment

0.14

Activations Density 0.005%

mentions of "courtesy" or related acknowledgments in the text

No Comments

No Known Activations