INDEX

Explanations

the contraction "didn't" followed by the letter "t"

oai_token-act-pair · gpt-3.5-turbo

instances of the word "didn't" in various forms

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

New Auto-Interp

Top Features by Cosine Similarity

Comparing With GPT2-SMALL @ 6-res-jb

Configuration

jbloom/GPT2-Small-SAEs-Reformatted/blocks.6.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.6.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

soType

-0.69

 Reduced

-0.68

pter

-0.67

 Continued

-0.66

 Dise

-0.65

 SetTextColor

-0.64

 Gleaming

-0.64

itiz

-0.64

lining

-0.62

Reviewer

-0.62

POSITIVE LOGITS

 bother

1.03

 necessarily

1.00

 hesitate

0.95

apest

0.80

olkien

0.80

aylor

0.79

weet

0.79

 exactly

0.78

ificate

0.78

 mention

0.77

Activations Density 0.077%

the contraction "didn't" followed by the letter "t"

instances of the word "didn't" in various forms

No Comments

No Known Activations