INDEX

Explanations

phrases related to continuity or appearing throughout

oai_token-act-pair · gpt-3.5-turbo

phrases and words conveying frequency or recurrence of events

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

New Auto-Interp

Top Features by Cosine Similarity

Comparing With GPT2-SMALL @ 3-res-jb

Configuration

jbloom/GPT2-Small-SAEs-Reformatted/blocks.3.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.3.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

ahime

-0.72

 proverb

-0.72

edia

-0.68

instein

-0.67

ello

-0.61

 Sergeant

-0.58

 Citation

-0.57

gart

-0.56

 Aging

-0.55

hower

-0.55

POSITIVE LOGITS

 except

1.16

except

0.97

 whatsoever

0.88

 alike

0.85

 besides

0.85

together

0.85

cape

0.79

 imaginable

0.76

 facets

0.74

 irrespective

0.70

Activations Density 0.603%

phrases related to continuity or appearing throughout

phrases and words conveying frequency or recurrence of events

No Comments

No Known Activations