INDEX

Explanations

direct quotations

oai_token-act-pair · gpt-3.5-turbo

quotation marks or speech indicators in the text

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

 accomp

-0.85

 carrier

-0.84

 removable

-0.82

 cleanup

-0.81

 adjud

-0.78

 disproportionately

-0.77

 cubic

-0.76

 dominate

-0.75

 flared

-0.75

 replacement

-0.75

POSITIVE LOGITS

We

1.46

Absolutely

1.44

There

1.42

1.41

Whoever

1.40

It

1.38

If

1.36

Obviously

1.36

You

1.35

Personally

1.34

Activations Density 0.126%

direct quotations

quotation marks or speech indicators in the text

No Comments

No Known Activations