INDEX

Explanations

proper nouns or names of entities such as individuals, companies, countries, and politicians

oai_token-act-pair · gpt-3.5-turbo

prominent figures or entities in various contexts

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

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

IFrame

Link

Not in Any Lists

No Comments

Negative Logits

 Reson

-0.69

Gir

-0.62

spr

-0.62

jac

-0.61

dx

-0.61

(?,

-0.61

gallery

-0.60

Roz

-0.60

BOOK

-0.59

 boils

-0.59

POSITIVE LOGITS

 traditionally

0.90

 longtime

0.88

 favored

0.86

 historically

0.83

 notoriously

0.82

 bedrock

0.81

 favoured

0.81

 bast

0.79

 precursor

0.79

 birthplace

0.78

Activations Density 0.282%

proper nouns or names of entities such as individuals, companies, countries, and politicians

prominent figures or entities in various contexts

No Comments

No Known Activations