INDEX

Explanations

adjectives indicating clear differences or separations

oai_token-act-pair · gpt-3.5-turbo

references to distinctiveness or uniqueness

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

New Auto-Interp

Top Features by Cosine Similarity

Comparing With GPT2-SMALL @ 0-res-jb

Configuration

jbloom/GPT2-Small-SAEs-Reformatted/blocks.0.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.0.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

bu

-0.74

hire

-0.67

USE

-0.64

 Aires

-0.63

ista

-0.63

buy

-0.62

aru

-0.62

asta

-0.61

 Weekend

-0.61

luck

-0.60

POSITIVE LOGITS

 distinct

3.43

 distinctive

2.03

 distinctly

1.83

 separate

1.64

 unique

1.59

 distinguish

1.51

 differentiated

1.46

 distinguished

1.43

 identifiable

1.42

 distinguishing

1.37

Activations Density 0.017%

adjectives indicating clear differences or separations

references to distinctiveness or uniqueness

No Comments

No Known Activations