INDEX

Explanations

the word "yes" with varying degrees of emphasis

oai_token-act-pair · gpt-3.5-turbo

occurrences of the word "yes."

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

bage

-0.90

perial

-0.83

actionDate

-0.78

RAW

-0.75

lets

-0.73

leted

-0.72

uterte

-0.72

ufact

-0.70

stem

-0.70

recated

-0.69

POSITIVE LOGITS

terday

1.56

yes

0.84

sir

0.74

 matter

0.72

YES

0.72

YES

0.71

 bake

0.71

etheless

0.70

 indeed

0.67

yes

0.65

Activations Density 0.013%

the word "yes" with varying degrees of emphasis

occurrences of the word "yes."

No Comments

No Known Activations