INDEX

Explanations

phrases indicating negation or potential misunderstanding

oai_token-act-pair · gpt-3.5-turbo

phrases that express caution or reassurance

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

New Auto-Interp

Top Features by Cosine Similarity

Comparing With GPT2-SMALL @ 6-res-jb

Configuration

jbloom/GPT2-Small-SAEs-Reformatted/blocks.6.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.6.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

ilogy

-0.76

anded

-0.68

alist

-0.66

ially

-0.64

figured

-0.64

azo

-0.62

ranch

-0.62

ettlement

-0.62

atar

-0.62

erial

-0.61

POSITIVE LOGITS

 yourself

1.05

 yourselves

1.02

 anymore

0.93

 Yourself

0.85

ANY

0.80

any

0.79

 your

0.75

 whining

0.70

 anything

0.70

 YOUR

0.68

Activations Density 0.123%

phrases indicating negation or potential misunderstanding

phrases that express caution or reassurance

No Comments

No Known Activations