INDEX

Explanations

words related to collections or being part of a set

oai_token-act-pair · gpt-3.5-turbo

references to various types of collections

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

New Auto-Interp

Top Features by Cosine Similarity

Comparing With GPT2-SMALL @ 2-res-jb

Configuration

jbloom/GPT2-Small-SAEs-Reformatted/blocks.2.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.2.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

pher

-0.72

NAS

-0.62

yl

-0.61

twitch

-0.60

 unden

-0.59

than

-0.58

 childbirth

-0.58

mic

-0.58

sl

-0.56

forth

-0.56

POSITIVE LOGITS

 collection

1.00

collection

0.96

 Collection

0.92

 collections

0.91

 Collections

0.87

ĸļ

0.86

DragonMagazine

0.80

OGR

0.79

ieu

0.78

 collectors

0.76

Activations Density 0.015%

words related to collections or being part of a set

references to various types of collections

No Comments

No Known Activations