INDEX

Explanations

verbs that indicate a change or impact on someone or something

oai_token-act-pair · gpt-3.5-turbo

distinct structural markers or separators within text

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

 Auditor

-0.63

 Dome

-0.62

 haul

-0.60

 pursuit

-0.59

NH

-0.58

Naj

-0.57

Tah

-0.56

 timet

-0.55

 Sydney

-0.55

 pursuing

-0.55

POSITIVE LOGITS

 partName

0.85

ŃĶ

0.80

obyl

0.78

Ĥİ

0.78

Allows

0.76

hower

0.76

ould

0.73

amples

0.73

si

0.71

acters

0.69

Activations Density 0.547%

verbs that indicate a change or impact on someone or something

distinct structural markers or separators within text

No Comments

No Known Activations