INDEX

Explanations

notes or comments at the end of text blocks

oai_token-act-pair · gpt-3.5-turbo

statements or annotations indicating important notes or comments

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

New Auto-Interp

Top Features by Cosine Similarity

Comparing With GPT2-SMALL @ 5-res-jb

Configuration

jbloom/GPT2-Small-SAEs-Reformatted/blocks.5.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.5.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

 verte

-0.74

 neighb

-0.71

 jaws

-0.70

atom

-0.70

ravel

-0.64

 carbohyd

-0.62

enium

-0.62

 whale

-0.61

esc

-0.61

raft

-0.61

POSITIVE LOGITS

books

1.23

BOOK

1.19

book

1.12

ably

0.84

Pad

0.84

andum

0.80

 Note

0.77

 Notes

0.76

 note

0.76

 Edited

0.73

Activations Density 0.023%

notes or comments at the end of text blocks

statements or annotations indicating important notes or comments

No Comments

No Known Activations