INDEX

Explanations

percentages mentioned in text

oai_token-act-pair · gpt-3.5-turbo

references to percentages and statistical data

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

New Auto-Interp

Top Features by Cosine Similarity

Comparing With GPT2-SMALL @ 5-res-jb

Configuration

jbloom/GPT2-Small-SAEs-Reformatted/blocks.5.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.5.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

Attributes

-0.72

ITED

-0.70

kefeller

-0.70

Reviewer

-0.69

GROUP

-0.68

anca

-0.66

DragonMagazine

-0.66

Merit

-0.66

æ³

-0.65

psons

-0.64

POSITIVE LOGITS

imet

1.28

imeter

1.08

oning

0.88

rate

0.83

urion

0.79

urance

0.78

+)

0.77

enary

0.76

rated

0.76

een

0.73

Activations Density 0.024%

percentages mentioned in text

references to percentages and statistical data

No Comments

No Known Activations