INDEX

Explanations

introductions or mentions of introducing something

oai_token-act-pair · gpt-3.5-turbo

instances of the word "introducing" and its variations

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

New Auto-Interp

Top Features by Cosine Similarity

Comparing With GPT2-SMALL @ 4-res-jb

Configuration

jbloom/GPT2-Small-SAEs-Reformatted/blocks.4.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.4.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

 Grail

-0.71

 Immunity

-0.70

 firsthand

-0.67

 Bulgarian

-0.66

 Frankfurt

-0.63

 Organisation

-0.63

 Danish

-0.62

 bearer

-0.62

 Romanian

-0.62

raints

-0.62

POSITIVE LOGITS

ctory

1.32

ctions

1.31

cer

1.18

cing

1.17

ctive

1.10

ction

1.09

cÃ©

1.06

ce

1.06

cedes

1.05

plet

1.02

Activations Density 0.023%

introductions or mentions of introducing something

instances of the word "introducing" and its variations

No Comments

No Known Activations