INDEX

Explanations

proper nouns related to location, particularly those related to streets or avenues

oai_token-act-pair · gpt-3.5-turbo

mentions of specific streets or avenues

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

PlotsExplanationShow Test FieldDefault Test Text

IFrame

Link

Not in Any Lists

No Comments

Negative Logits

ERAL

-0.94

sonian

-0.93

=-=-=-=-=-=-=-=-

-0.84

estamp

-0.79

berman

-0.77

MER

-0.75

milo

-0.73

ebook

-0.70

advertisement

-0.67

wed

-0.66

POSITIVE LOGITS

 Avenue

1.09

Rd

0.89

 Junction

0.85

Ave

0.84

 Parkway

0.81

 Street

0.81

 Route

0.81

 streetcar

0.81

 Plaza

0.78

Street

0.78

Activations Density 0.018%

proper nouns related to location, particularly those related to streets or avenues

mentions of specific streets or avenues

No Comments

No Known Activations