INDEX

Explanations

mentions of the word "Princess" at varying intensities, possibly related to different contexts or relationships

oai_token-act-pair · gpt-3.5-turbo

references to various princesses

oai_token-act-pair · gpt-4o-mini Triggered by @bot

New Auto-Interp

Top Features by Cosine Similarity

Comparing With GPT2-SMALL @ 7-res-jb

Configuration

jbloom/GPT2-Small-SAEs-Reformatted/blocks.7.hook_resid_pre

Prompts (Dashboard)

24,576 prompts, 128 tokens each

Dataset (Dashboard)

Skylion007/openwebtext

Features

24,576

Data Type

torch.float32

Hook Point

blocks.7.hook_resid_pre

Architecture

standard

Context Size

128

Dataset

Skylion007/openwebtext

Hook Point Layer

Activation Function

relu

Embeds

IFrame

Link

Not in Any Lists

No Comments

Negative Logits

 spaced

-0.69

sych

-0.68

ophon

-0.65

ulhu

-0.63

 neur

-0.62

ucl

-0.62

rils

-0.61

 funn

-0.61

appa

-0.61

oller

-0.60

POSITIVE LOGITS

 Leia

1.12

 Bride

1.10

 Celest

1.01

 Diana

0.97

anova

0.97

 Princess

0.95

 Peach

0.92

 princess

0.88

 Fiona

0.84

cess

0.83

Activations Density 0.029%

mentions of the word "Princess" at varying intensities, possibly related to different contexts or relationships

references to various princesses

No Comments

No Known Activations