INDEX

Explanations

mentions of "draft" and related terms indicating revisions or proposals

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

New Auto-Interp

Top Features by Cosine Similarity

Comparing With LLAMA3.1-8B @ 4-llamascope-res-32k

Configuration

fnlp/Llama3_1-8B-Base-LXR-8x/Llama3_1-8B-Base-L4R-8x

Prompts (Dashboard)

24,576 prompts, 128 tokens each

Dataset (Dashboard)

cerebras/SlimPajama-627B

Features

32,768

Data Type

bfloat16

Hook Name

blocks.4.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

à¤¤à¤®

-0.15

paragraph

-0.14

 trait

-0.14

zial

-0.14

cie

-0.14

à¤¤à¤°

-0.14

jm

-0.13

amel

-0.13

zia

-0.13

etter

-0.13

POSITIVE LOGITS

 WARRANT

0.17

voÅĻ

0.16

vg

0.16

 GOODMAN

0.15

aroo

0.15

inger

0.15

ixel

0.15

rophic

0.14

ãģªãģı

0.14

inge

0.14

Activations Density 0.007%

mentions of "draft" and related terms indicating revisions or proposals

No Comments

No Known Activations