INDEX

Explanations

words related to actions of control or authority

oai_token-act-pair · gpt-3.5-turbo

instances of empty or placeholder text sections

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

PlotsExplanationShow Test FieldDefault Test Text

IFrame

Link

Not in Any Lists

No Comments

Negative Logits

 Vaugh

-0.72

esides

-0.60

 Moroc

-0.59

 predec

-0.57

dating

-0.56

luster

-0.56

recorded

-0.55

vertisement

-0.55

venth

-0.54

enty

-0.54

POSITIVE LOGITS

 oneself

0.62

 thee

0.60

 abortions

0.55

)?

0.55

 Xperia

0.53

him

0.52

jriwal

0.51

 Aadhaar

0.51

 yourselves

0.51

 Corbyn

0.51

Activations Density 0.721%

words related to actions of control or authority

instances of empty or placeholder text sections

No Comments

No Known Activations