INDEX

Explanations

certain correctness indicators or confirmation phrases in calculations or assessments

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

cess

-0.06

ETA

-0.06

/net

-0.06

ÑĢÐ¾ÑģÑĤ

-0.06

ording

-0.06

ayacak

-0.06

 jedn

-0.06

ÑĢÐ°ÑģÑĤ

-0.06

king

-0.06

Ø§Øª

-0.06

POSITIVE LOGITS

 also

0.07

 Ridley

0.07

 again

0.07

lbrace

0.06

askan

0.06

erd

0.06

//{{

0.06

Activations Density 0.104%

certain correctness indicators or confirmation phrases in calculations or assessments

No Comments

No Known Activations