INDEX

Explanations

proper nouns related to neighborhoods or locations

oai_token-act-pair · gpt-3.5-turbo

mentions of specific geographic locations

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

New Auto-Interp

Top Features by Cosine Similarity

Comparing With GPT2-SMALL @ 5-res-jb

Configuration

jbloom/GPT2-Small-SAEs-Reformatted/blocks.5.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.5.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

unal

-0.88

ificial

-0.87

onom

-0.84

isc

-0.83

uments

-0.82

ional

-0.80

ivity

-0.79

cies

-0.77

adic

-0.76

inent

-0.76

POSITIVE LOGITS

 Heights

0.94

 Borough

0.85

tones

0.80

 neighbourhood

0.76

creen

0.76

 Dwell

0.75

 Shopping

0.73

 Neighborhood

0.73

 Village

0.70

 neighbourhoods

0.69

Activations Density 0.029%

proper nouns related to neighborhoods or locations

mentions of specific geographic locations

No Comments

No Known Activations