INDEX

Explanations

questions or statements ending in a question mark

oai_token-act-pair · gpt-3.5-turbo

rhetorical questions and expressions of uncertainty

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

New Auto-Interp

Top Features by Cosine Similarity

Comparing With GPT2-SMALL @ 11-res-jb

Configuration

jbloom/GPT2-Small-SAEs-Reformatted/blocks.11.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.11.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

anwhile

-0.71

escription

-0.64

arde

-0.63

 laun

-0.63

İĭ

-0.62

luence

-0.61

ronics

-0.61

CRIP

-0.61

 Skydragon

-0.60

monton

-0.60

POSITIVE LOGITS

yes

1.98

yes

1.97

Yes

1.90

Yes

1.89

YES

1.88

YES

1.83

 Nope

1.75

Absolutely

1.73

 Absolutely

1.60

No

1.47

Activations Density 0.302%

questions or statements ending in a question mark

rhetorical questions and expressions of uncertainty

No Comments

No Known Activations