INDEX

Explanations

mentions of additions or enhancements in different contexts

oai_token-act-pair · gpt-3.5-turbo

occurrences of the word "addition."

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

New Auto-Interp

Top Features by Cosine Similarity

Comparing With GPT2-SMALL @ 4-res-jb

Configuration

jbloom/GPT2-Small-SAEs-Reformatted/blocks.4.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.4.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

bis

-0.70

zees

-0.69

bane

-0.68

zh

-0.67

walking

-0.63

cdn

-0.63

zi

-0.62

raz

-0.62

uchs

-0.61

meta

-0.61

POSITIVE LOGITS

xual

0.88

endum

0.83

ition

0.81

itious

0.78

 thereto

0.75

 Flavoring

0.74

 insult

0.72

itive

0.70

olean

0.69

itives

0.68

Activations Density 0.013%

mentions of additions or enhancements in different contexts

occurrences of the word "addition."

No Comments

No Known Activations