INDEX

Explanations

references to geometric relationships or pairings

oai_token-act-pair · gpt-4o-mini

New Auto-Interp

Configuration

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

Prompts (Dashboard)

24,576 prompts, 128 tokens each

Dataset (Dashboard)

open-r1/OpenR1-Math-220k

Features

32,768

Data Type

float32

Hook Name

blocks.15.hook_resid_post

Architecture

jumprelu

Context Size

1,024

Dataset

cerebras/SlimPajama-627B

Embeds

PlotsExplanationShow Test FieldDefault Test Text

IFrame

Link

Not in Any Lists

No Comments

Negative Logits

illard

-0.07

china

-0.07

laden

-0.07

ÑĢÐ°ÑĤ

-0.07

ãĥ¬ãĥ³

-0.06

kili

-0.06

(æĹ¥

-0.06

/ubuntu

-0.06

 Phonetic

-0.06

aki

-0.06

POSITIVE LOGITS

 ones

0.12

 Ones

0.08

ones

0.07

(er

0.07

ONES

0.06

maal

0.06

etimes

0.06

liest

0.06

inox

0.06

 thereof

0.06

Activations Density 0.091%

references to geometric relationships or pairings

No Comments

No Known Activations