INDEX

Explanations

references to the Premier League, a popular football league in England

oai_token-act-pair · gpt-3.5-turbo

references to the Premier League

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

New Auto-Interp

Top Features by Cosine Similarity

Comparing With GPT2-SMALL @ 0-res-jb

Configuration

jbloom/GPT2-Small-SAEs-Reformatted/blocks.0.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.0.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

agh

-0.81

 hallucinations

-0.73

 thereof

-0.69

 Adin

-0.69

ieg

-0.68

Î³

-0.68

nik

-0.68

Cod

-0.67

azel

-0.65

eth

-0.65

POSITIVE LOGITS

 Premier

3.71

Prem

2.17

 premier

1.90

 Premiership

1.68

 Bundesliga

1.59

 Prime

1.41

 Serie

1.37

 Champions

1.35

 Provincial

1.25

 Liga

1.24

Activations Density 0.019%

references to the Premier League, a popular football league in England

references to the Premier League

No Comments

No Known Activations