INDEX

Explanations

questions at the end of a sentence

oai_token-act-pair · gpt-3.5-turbo

questions that propose hypothetical scenarios or prompt further inquiry

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

 inclusive

-0.71

apan

-0.69

 binge

-0.63

icro

-0.63

ulla

-0.62

aper

-0.61

 booked

-0.61

 fart

-0.60

enos

-0.60

 fabric

-0.60

POSITIVE LOGITS

 Nope

1.36

 Probably

1.22

 Certainly

1.21

 Possibly

1.14

 Absolutely

1.12

 Surely

1.09

 Perhaps

1.08

Yes

1.06

 Answer

1.06

 Maybe

1.05

Activations Density 0.089%

questions at the end of a sentence

questions that propose hypothetical scenarios or prompt further inquiry

No Comments

No Known Activations