INDEX

Explanations

information related to awards and nominations in various categories

oai_token-act-pair · gpt-4o-mini

New Auto-Interp

Configuration

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

Prompts (Dashboard)

24,576 prompts, 128 tokens each

Dataset (Dashboard)

Hzfinfdu/SlimPajama-3B

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

Å¾ÃŃ

-0.16

ebin

-0.15

ZO

-0.15

ascar

-0.15

unami

-0.14

 Champ

-0.14

anzi

-0.14

 yÄ±

-0.14

ague

-0.14

Å¡ka

-0.14

POSITIVE LOGITS

 selected

0.22

 selection

0.21

 selections

0.20

 Selection

0.20

selected

0.19

 batch

0.17

-selection

0.17

åĢĻ

0.17

 chosen

0.16

-selected

0.16

Activations Density 0.085%

information related to awards and nominations in various categories

No Comments

No Known Activations