INDEX

Explanations

references to vehicles and driving-related topics

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

oppers

-0.16

ramework

-0.15

elon

-0.14

rimp

-0.14

 aboard

-0.14

reuse

-0.14

taire

-0.13

 ÑģÐ¾Ð±Ð¾Ð¹

-0.13

ŀĭ

-0.13

lama

-0.13

POSITIVE LOGITS

 driving

0.16

car

0.16

ovan

0.15

car

0.15

é©¶

0.15

 emissions

0.15

-road

0.15

itten

0.14

 driven

0.14

urette

0.14

Activations Density 0.124%

references to vehicles and driving-related topics

No Comments

No Known Activations