INDEX

Explanations

contractions with "doesn't" or "don't"

oai_token-act-pair · gpt-3.5-turbo

instances of the word "doesn't" or its variations

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

fox

-0.75

hung

-0.67

armed

-0.65

 Remastered

-0.65

dar

-0.63

phant

-0.62

ANG

-0.62

 Mant

-0.61

 Ability

-0.61

tone

-0.60

POSITIVE LOGITS

't

1.47

kie

0.87

ettings

0.82

berra

0.77

paces

0.75

olulu

0.73

terness

0.73

ajor

0.72

NOT

0.71

\'

0.70

Activations Density 0.026%

contractions with "doesn't" or "don't"

instances of the word "doesn't" or its variations

No Comments

No Known Activations