INDEX

Explanations

instances where someone suggests or implies something

oai_token-act-pair · gpt-3.5-turbo

instances of the word "suggested" and its variants

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

gie

-0.82

tesy

-0.76

percent

-0.74

cedented

-0.73

pes

-0.73

respective

-0.72

brance

-0.70

haar

-0.69

onder

-0.67

FO

-0.66

POSITIVE LOGITS

 otherwise

0.94

 that

0.89

 alternatives

0.82

 abandoning

0.79

 lowering

0.79

 remedies

0.78

ively

0.77

 eliminating

0.76

 reconsider

0.75

 caution

0.75

Activations Density 0.082%

instances where someone suggests or implies something

instances of the word "suggested" and its variants

No Comments

No Known Activations