INDEX

Explanations

punctuation marks, particularly apostrophes and colons, which may indicate quotation or dialogue

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

uft

-0.07

lug

-0.07

ue

-0.06

achat

-0.06

ardy

-0.06

agle

-0.06

pone

-0.06

eniable

-0.06

Ìĥ

-0.06

oola

-0.06

POSITIVE LOGITS

à¥Įà¤¦

0.07

_mime

0.06

ï¼«

0.06

**************

0.06

ï¸

0.06

ivan

0.06

è°±

0.06

 tarz

0.06

ï¼Īå¹³æĪĲ

0.06

ddit

0.06

Activations Density 0.162%

punctuation marks, particularly apostrophes and colons, which may indicate quotation or dialogue

No Comments

No Known Activations