INDEX

Explanations

references to people and pronouns related to interactions and relationships

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

 itself

-0.26

taire

-0.17

tti

-0.15

arah

-0.15

apl

-0.14

leans

-0.14

wy

-0.14

acht

-0.13

laps

-0.13

POSITIVE LOGITS

/us

0.35

/her

0.24

-mÃªme

0.20

atically

0.19

self

0.18

zelf

0.18

/we

0.17

SELF

0.16

chy

0.15

/th

0.14

Activations Density 0.230%

references to people and pronouns related to interactions and relationships

No Comments

No Known Activations