INDEX

Explanations

references to actions and behaviors, especially in a competitive context

oai_token-act-pair · gpt-4o-mini

New Auto-Interp

Configuration

fnlp/Llama-Scope-R1-Distill/800M-Slimpajama-0-OpenR1-Math-220k/L15R

Prompts (Dashboard)

24,576 prompts, 128 tokens each

Dataset (Dashboard)

Hzfinfdu/SlimPajama-3B

Features

32,768

Data Type

float32

Hook Name

blocks.15.hook_resid_post

Hook Layer

Architecture

jumprelu

Context Size

1,024

Dataset

cerebras/SlimPajama-627B

Activation Function

relu

Embeds

PlotsExplanationShow Test FieldDefault Test Text

IFrame

Link

Not in Any Lists

No Comments

Negative Logits

 Mile

-0.15

 tang

-0.14

 mileage

-0.14

fq

-0.14

æµ·å¤ĸ

-0.13

 Martin

-0.13

 Major

-0.13

bound

-0.13

set

-0.13

POSITIVE LOGITS

Ã©ru

0.16

NSNotification

0.15

 phá»¥

0.14

ëĭ¤ìļ´

0.14

ÑĤÐ¾ÑĢ

0.14

oÄŁ

0.13

yat

0.13

upertino

0.13

íĻ©

0.13

urdy

0.13

Activations Density 0.095%

references to actions and behaviors, especially in a competitive context

No Comments

No Known Activations