INDEX

Explanations

phrases indicating agreement or disagreement in conversation

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

New Auto-Interp

Top Features by Cosine Similarity

Comparing With LLAMA3.1-8B @ 21-llamascope-res-32k

Configuration

fnlp/Llama3_1-8B-Base-LXR-8x/Llama3_1-8B-Base-L21R-8x

Prompts (Dashboard)

24,576 prompts, 128 tokens each

Dataset (Dashboard)

cerebras/SlimPajama-627B

Features

32,768

Data Type

bfloat16

Hook Name

blocks.21.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

ÐµÐ²Ð¸Ñĩ

-0.18

alian

-0.15

avy

-0.15

à¸²à¸¢à¸Ļ

-0.15

ancode

-0.14

roz

-0.14

nemonic

-0.14

instein

-0.14

emark

-0.14

.called

-0.14

POSITIVE LOGITS

yes

0.41

YES

0.41

Yes

0.40

yes

0.37

YES

0.36

Yes

0.34

NO

0.32

_yes

0.31

"Yes

0.30

“Yes

0.29

Activations Density 0.073%

phrases indicating agreement or disagreement in conversation

No Comments

No Known Activations