INDEX

Explanations

topics and concepts related to customization and configuration settings

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

èĤ²

-0.16

iddi

-0.16

elin

-0.15

-0.14

ientes

-0.14

elia

-0.14

qrt

-0.13

 hmot

-0.13

ayi

-0.13

UDIO

-0.13

POSITIVE LOGITS

 Barker

0.14

Ã¤ge

0.14

Ð°Ð³Ð°ÑĤÐ¾

0.13

 Caleb

0.13

«a

0.13

GRE

0.13

ichick

0.12

essen

0.12

bye

0.12

.DAL

0.12

Activations Density 0.159%

topics and concepts related to customization and configuration settings

No Comments

No Known Activations