INDEX

Explanations

mentions of the character "Doctor" in a sci-fi or medical context

oai_token-act-pair · gpt-3.5-turbo

the word "Doctor" and its various occurrences in context

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

New Auto-Interp

Top Features by Cosine Similarity

Comparing With GPT2-SMALL @ 1-res-jb

Configuration

jbloom/GPT2-Small-SAEs-Reformatted/blocks.1.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.1.hook_resid_pre

Architecture

standard

Context Size

128

Dataset

Skylion007/openwebtext

Hook Point Layer

Activation Function

relu

Embeds

PlotsExplanationShow Test FieldDefault Test Text

IFrame

Link

Not in Any Lists

No Comments

Negative Logits

 actionGroup

-0.94

 deleting

-0.65

 rout

-0.65

tag

-0.64

 dispersed

-0.64

 envelope

-0.63

 nerve

-0.62

HUD

-0.62

twitter

-0.61

EFF

-0.61

POSITIVE LOGITS

 Doctor

1.18

Doctor

0.94

sonian

0.93

iatus

0.91

abase

0.88

 Doctors

0.87

Who

0.85

iatric

0.84

ARDIS

0.83

doctor

0.80

Activations Density 0.006%

mentions of the character "Doctor" in a sci-fi or medical context

the word "Doctor" and its various occurrences in context

No Comments

No Known Activations