INDEX

Explanations

quotations within texts

oai_token-act-pair · gpt-3.5-turbo

dialogue or quotes 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 @ 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

IFrame

Link

Not in Any Lists

No Comments

Negative Logits

 describ

-0.80

 looph

-0.79

rul

-0.78

 favour

-0.78

 edged

-0.78

 moder

-0.77

 vegetarian

-0.76

 spill

-0.75

¥ŀ

-0.75

 discont

-0.74

POSITIVE LOGITS

Everybody

1.87

We

1.83

They

1.81

Honestly

1.81

It

1.78

Obviously

1.76

Absolutely

1.75

1.74

Yeah

1.74

Everything

1.72

Activations Density 0.153%

quotations within texts

dialogue or quotes in the text

No Comments

No Known Activations