INDEX

Explanations

the word "break" or related phrases

oai_token-act-pair · gpt-3.5-turbo

references to the word "break."

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

uni

-0.77

 Poster

-0.70

 Shap

-0.69

ongh

-0.69

anooga

-0.68

landish

-0.68

ushi

-0.67

judicial

-0.66

emed

-0.66

imir

-0.66

POSITIVE LOGITS

 break

3.67

 breaks

2.57

break

2.45

 Break

2.35

Break

2.12

 breaking

1.87

breaks

1.78

 broke

1.76

 breakdown

1.70

 breakup

1.60

Activations Density 0.011%

the word "break" or related phrases

references to the word "break."

No Comments

No Known Activations