INDEX

Explanations

keywords related to notes, updates, and documentation

oai_token-act-pair · gpt-3.5-turbo

references to notes and documentation

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

New Auto-Interp

Top Features by Cosine Similarity

Comparing With GPT2-SMALL @ 6-res-jb

Configuration

jbloom/GPT2-Small-SAEs-Reformatted/blocks.6.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.6.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

Interstitial

-0.79

ILLE

-0.75

LOS

-0.75

ãĥĩ

-0.67

fare

-0.63

namese

-0.62

 indignation

-0.62

cester

-0.62

duct

-0.61

 renewal

-0.61

POSITIVE LOGITS

books

0.91

notes

0.91

 Notes

0.85

BOOK

0.85

ppelin

0.83

ense

0.81

 Gained

0.81

book

0.78

 notes

0.76

Pad

0.75

Activations Density 0.022%

keywords related to notes, updates, and documentation

references to notes and documentation

No Comments

No Known Activations