INDEX

Explanations

names followed by the abbreviation 'ab'

oai_token-act-pair · gpt-3.5-turbo

occurrences of the prefix "ab."

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

New Auto-Interp

Top Features by Cosine Similarity

Comparing With GPT2-SMALL @ 3-res-jb

Configuration

jbloom/GPT2-Small-SAEs-Reformatted/blocks.3.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.3.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

enegger

-0.79

geist

-0.76

DragonMagazine

-0.74

zsche

-0.71

 Maker

-0.69

ESSION

-0.69

 Dodd

-0.68

ancial

-0.67

 Inquisitor

-0.66

 Eisen

-0.65

POSITIVE LOGITS

ibi

0.96

elled

0.94

raham

0.93

usable

0.89

ril

0.89

rog

0.88

duct

0.88

ject

0.87

rid

0.87

obo

0.86

Activations Density 0.009%

names followed by the abbreviation 'ab'

occurrences of the prefix "ab."

No Comments

No Known Activations