INDEX

Explanations

proper nouns

oai_token-act-pair · gpt-3.5-turbo

proper nouns, particularly names of places, animals, or notable items

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

New Auto-Interp

Top Features by Cosine Similarity

Comparing With GPT2-SMALL @ 7-res-jb

Configuration

jbloom/GPT2-Small-SAEs-Reformatted/blocks.7.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.7.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

matically

-0.83

ischer

-0.78

 notor

-0.75

angible

-0.73

 referen

-0.73

adobe

-0.68

bably

-0.67

 proble

-0.66

 complimentary

-0.66

 confir

-0.66

POSITIVE LOGITS

 Productions

1.28

 Studios

1.20

 Brewing

1.13

 Syndrome

1.07

 Farms

1.07

 Ranch

1.06

 Brewery

1.02

 Records

0.99

 Mountains

0.99

 Cafe

0.98

Activations Density 0.270%

proper nouns

proper nouns, particularly names of places, animals, or notable items

No Comments

No Known Activations