INDEX

Explanations

references to specific businesses or locations

oai_token-act-pair · gpt-3.5-turbo

references to locations and establishments, particularly restaurants and bars

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

New Auto-Interp

Top Features by Cosine Similarity

Comparing With GPT2-SMALL @ 8-res-jb

Configuration

jbloom/GPT2-Small-SAEs-Reformatted/blocks.8.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.8.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

 embryos

-0.88

 declass

-0.85

 FISA

-0.82

 simulations

-0.81

 Supplementary

-0.81

 epigen

-0.81

 statistical

-0.80

pse

-0.79

 deposition

-0.78

 probes

-0.78

POSITIVE LOGITS

 Cafe

1.50

 CafÃ©

1.49

 restaurants

1.46

 Restaurant

1.45

 Restaur

1.39

 restaurant

1.35

 storefront

1.28

 Lounge

1.23

 Brewery

1.21

 restaur

1.17

Activations Density 0.964%

references to specific businesses or locations

references to locations and establishments, particularly restaurants and bars

No Comments

No Known Activations