INDEX

Explanations

terms related to understanding or articulating plans and intentions

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

bit

-0.15

åĬ¨çĶŁæĪĲ

-0.14

acco

-0.14

optera

-0.13

ie

-0.13

iki

-0.13

 itself

-0.13

tor

-0.13

earer

-0.13

orus

-0.13

POSITIVE LOGITS

 therefore

0.18

renom

0.14

)>>

0.13

upertino

0.13

crement

0.13

},{↵

0.13

 Ø¨Ø¬

0.13

å§ĭ

0.13

mant

0.13

ValueCollection

0.13

Activations Density 0.207%

terms related to understanding or articulating plans and intentions

No Comments

No Known Activations