INDEX

Explanations

conditional phrases that present alternative scenarios or exceptions

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

ãĥ³ãĥĩãĤ£

-0.07

zano

-0.06

 kinky

-0.06

Browse

-0.06

errupt

-0.06

äº¤éĢļ

-0.06

erved

-0.06

peror

-0.06

zel

-0.06

stva

-0.06

POSITIVE LOGITS

Ens

0.07

Ens

0.07

 worst

0.06

 minimal

0.06

 weakest

0.06

oes

0.06

Å¯st

0.06

min

0.06

.backup

0.06

hear

0.06

Activations Density 0.027%

conditional phrases that present alternative scenarios or exceptions

No Comments

No Known Activations