INDEX

Explanations

mentions of actions related to granting permissions or approvals

oai_token-act-pair · gpt-3.5-turbo

New Auto-Interp

Top Features by Cosine Similarity

Comparing With GPT2-SMALL @ 9-res-jb

Configuration

jbloom/GPT2-Small-SAEs-Reformatted/blocks.9.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.9.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

 Maker

-0.71

 Cups

-0.71

 Citation

-0.70

ancial

-0.69

 virtue

-0.66

 Prosper

-0.61

FTWARE

-0.60

 Nadu

-0.60

 Solitaire

-0.60

 Beir

-0.59

POSITIVE LOGITS

raham

1.10

usable

1.09

rog

1.07

omination

1.07

rid

1.03

ject

1.02

stract

0.98

dullah

0.97

duction

0.95

bey

0.95

Activations Density 0.024%

mentions of actions related to granting permissions or approvals

No Comments

No Known Activations