INDEX

Explanations

details related to publication information and book bibliographies

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

Lay

-0.16

–

-0.15

 retr

-0.15

uem

-0.15

zing

-0.14

WindowTitle

-0.14

Ara

-0.14

ittle

-0.14

 refer

-0.14

fan

-0.14

POSITIVE LOGITS

á»£

0.15

 nackte

0.15

fav

0.15

Ä±s

0.14

owitz

0.14

ogrÃ¡f

0.14

æį·

0.14

istrovstvÃŃ

0.14

æı

0.14

imuth

0.14

Activations Density 0.089%

details related to publication information and book bibliographies

No Comments

No Known Activations