INDEX

Explanations

patterns in mathematical expressions and equations

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

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

 Sink

-0.06

cob

-0.06

usu

-0.06

lid

-0.06

íķŃ

-0.06

crement

-0.06

NST

-0.06

ONS

-0.06

_hz

-0.06

POSITIVE LOGITS

ongyang

0.07

 accordingly

0.07

ottle

0.07

LogLevel

0.06

ãĥ³ãĥĲ

0.06

raits

0.06

illance

0.06

deb

0.06

ActionCreators

0.06

Activations Density 0.066%

patterns in mathematical expressions and equations

No Comments

No Known Activations