INDEX

Explanations

words or phrases surrounded by quotation marks (") where the quotation marks follow 'I'

oai_token-act-pair · gpt-3.5-turbo

occurrences of the word "I"

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

New Auto-Interp

Top Features by Cosine Similarity

Comparing With GPT2-SMALL @ 0-res-jb

Configuration

jbloom/GPT2-Small-SAEs-Reformatted/blocks.0.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.0.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

 grooming

-0.65

esville

-0.59

 aver

-0.56

 stocking

-0.56

 turf

-0.56

 finalists

-0.54

 sustain

-0.53

 significant

-0.53

bes

-0.53

sted

-0.53

POSITIVE LOGITS

2.93

My

1.70

IAS

1.44

Is

1.39

We

1.34

II

1.33

Oh

1.31

Honestly

1.31

Personally

1.30

Activations Density 0.081%

words or phrases surrounded by quotation marks (") where the quotation marks follow 'I'

occurrences of the word "I"

No Comments

No Known Activations