INDEX

Explanations

versions or different editions of something

oai_token-act-pair · gpt-3.5-turbo

phrases that indicate different versions or adaptations of something

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

New Auto-Interp

Top Features by Cosine Similarity

Comparing With GPT2-SMALL @ 10-res-jb

Configuration

jbloom/GPT2-Small-SAEs-Reformatted/blocks.10.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.10.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

urers

-0.76

earchers

-0.74

hya

-0.70

 merits

-0.68

roma

-0.66

arers

-0.66

zinski

-0.66

cks

-0.65

erity

-0.65

rings

-0.65

POSITIVE LOGITS

 sorts

0.90

 Sleeping

0.75

 theirs

0.75

 himself

0.71

Hoo

0.71

 Yourself

0.70

 ourselves

0.70

 themselves

0.69

 Cinderella

0.69

 Solitaire

0.68

Activations Density 0.122%

versions or different editions of something

phrases that indicate different versions or adaptations of something

No Comments

No Known Activations