INDEX

Explanations

titles and publication details

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

odore

-0.25

ris

-0.16

eden

-0.16

istrovstvÃŃ

-0.16

edar

-0.15

atre

-0.15

ÐµÑģÑı

-0.14

ucks

-0.14

redd

-0.14

amt

-0.14

POSITIVE LOGITS

é¡µéĿ¢åŃĺæ¡£å¤ĩä»½

0.20

 latter

0.19

amp

0.19

eniable

0.18

Ð°Ð¶

0.18

itemap

0.16

enos

0.15

tiv

0.15

enu

0.15

0.14

Activations Density 0.270%

titles and publication details

No Comments

No Known Activations