INDEX

Explanations

names, particularly variations of the name "Keri" or similar

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

irty

-0.16

rette

-0.16

-archive

-0.15

é±

-0.15

poke

-0.15

ickt

-0.15

æĿŁ

-0.15

kili

-0.15

Difference

-0.15

ableView

-0.15

POSITIVE LOGITS

Ã¡v

0.17

oton

0.17

shire

0.17

/REC

0.17

heimer

0.16

ÑıÐ±

0.15

oman

0.14

ohan

0.14

íķĳ

0.14

esti

0.14

Activations Density 0.046%

names, particularly variations of the name "Keri" or similar

No Comments

No Known Activations