INDEX

Explanations

references to individuals named Roy

oai_token-act-pair · gpt-3.5-turbo

references to individuals named Roy

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

TSA

-0.66

CBI

-0.65

 Disorder

-0.64

 oper

-0.61

 Indones

-0.60

 Articles

-0.60

 opacity

-0.60

 containment

-0.59

 grease

-0.59

ctica

-0.58

POSITIVE LOGITS

alties

1.47

ce

1.03

dy

1.02

Roy

0.90

bal

0.89

ston

0.88

 Hodg

0.87

der

0.84

 Moore

0.84

croft

0.83

Activations Density 0.012%

references to individuals named Roy

references to individuals named Roy

No Comments

No Known Activations