INDEX

Explanations

phrases indicating options, choices, or alternatives

oai_token-act-pair · gpt-3.5-turbo

conditional phrases related to options or alternatives

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

New Auto-Interp

Top Features by Cosine Similarity

Comparing With GPT2-SMALL @ 9-res-jb

Configuration

jbloom/GPT2-Small-SAEs-Reformatted/blocks.9.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.9.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

 understandably

-0.61

 then

-0.61

 also

-0.59

 foremost

-0.57

 Aden

-0.56

now

-0.56

 Abram

-0.56

 Amph

-0.54

 Ramos

-0.54

 doubtless

-0.54

POSITIVE LOGITS

abouts

1.20

Else

1.11

depending

1.02

chard

0.96

 combination

0.92

omever

0.92

nam

0.91

 thereof

0.89

 equivalent

0.89

 Marketable

0.84

Activations Density 0.296%

phrases indicating options, choices, or alternatives

conditional phrases related to options or alternatives

No Comments

No Known Activations