INDEX

Explanations

various content-related artifacts like line breaks, special characters, headings, and other structural elements within the text

oai_token-act-pair · gpt-3.5-turbo

expressions of quantity or measurements

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

New Auto-Interp

Top Features by Cosine Similarity

Comparing With GPT2-SMALL @ 7-res-jb

Configuration

jbloom/GPT2-Small-SAEs-Reformatted/blocks.7.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.7.hook_resid_pre

Architecture

standard

Context Size

128

Dataset

Skylion007/openwebtext

Hook Point Layer

Activation Function

relu

Embeds

IFrame

Link

Not in Any Lists

No Comments

Negative Logits

 bloc

-0.75

 coral

-0.74

 Aval

-0.72

 grapp

-0.72

 camp

-0.71

 symb

-0.71

 approved

-0.70

 tent

-0.69

 hatch

-0.69

 consum

-0.68

POSITIVE LOGITS

However

1.57

But

1.54

Therefore

1.47

Moreover

1.47

Thus

1.45

That

1.44

Furthermore

1.44

Likewise

1.43

Then

1.43

And

1.43

Activations Density 0.526%

various content-related artifacts like line breaks, special characters, headings, and other structural elements within the text

expressions of quantity or measurements

No Comments

No Known Activations