INDEX

Explanations

references to alternative options or conditions

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

adora

-0.17

ãĥĨãĥ«

-0.15

Civ

-0.15

 Kons

-0.14

api

-0.14

.tolist

-0.13

¿ł

-0.13

 forb

-0.13

 rather

-0.13

POSITIVE LOGITS

 otherwise

0.24

otherwise

0.21

 Otherwise

0.20

Otherwise

0.19

 else

0.19

 sonst

0.18

 Ð¿ÑĢÐ¾Ñĩ

0.16

_else

0.16

 jinak

0.16

ELSE

0.15

Activations Density 0.075%

references to alternative options or conditions

No Comments

No Known Activations