INDEX

Explanations

instructions or steps in a task or process

oai_token-act-pair · gpt-3.5-turbo

language related to instructions or guidance for technical tasks

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

PlotsExplanationShow Test FieldDefault Test Text

IFrame

Link

Not in Any Lists

No Comments

Negative Logits

ufact

-0.59

atories

-0.59

 Kurds

-0.59

 Islamists

-0.59

 appell

-0.58

 Arabs

-0.58

 objected

-0.58

Iraq

-0.58

 Kazakh

-0.57

Said

-0.57

POSITIVE LOGITS

 your

0.87

 YOUR

0.86

your

0.84

 Your

0.80

Your

0.77

 personalized

0.75

 yourself

0.72

 beginner

0.72

 yours

0.72

 Tips

0.70

Activations Density 1.247%

instructions or steps in a task or process

language related to instructions or guidance for technical tasks

No Comments

No Known Activations