INDEX

Explanations

critical terms and headings related to structured documents or reports

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

New Auto-Interp

Top Features by Cosine Similarity

Comparing With LLAMA3.1-8B @ 9-llamascope-res-32k

Configuration

fnlp/Llama3_1-8B-Base-LXR-8x/Llama3_1-8B-Base-L9R-8x

Prompts (Dashboard)

24,576 prompts, 128 tokens each

Dataset (Dashboard)

cerebras/SlimPajama-627B

Features

32,768

Data Type

bfloat16

Hook Name

blocks.9.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

gili

-0.16

esser

-0.16

/Q

-0.15

ãĤīãģı

-0.15

igion

-0.14

 tlak

-0.14

 Wanted

-0.14

Ð¼Ð¸

-0.14

OURNAL

-0.14

osal

-0.13

POSITIVE LOGITS

ilst

0.18

coc

0.17

ipc

0.15

uide

0.15

 bets

0.15

ucher

0.14

uche

0.14

æ¢¨

0.14

hausen

0.14

 Rico

0.14

Activations Density 0.018%

critical terms and headings related to structured documents or reports

No Comments

No Known Activations