INDEX

Explanations

references to Microsoft Office-related terms

oai_token-act-pair · gpt-3.5-turbo

mentions of Microsoft Office and its related products

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

New Auto-Interp

Top Features by Cosine Similarity

Comparing With GPT2-SMALL @ 3-res-jb

Configuration

jbloom/GPT2-Small-SAEs-Reformatted/blocks.3.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.3.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

ONSORED

-0.81

ablishment

-0.75

ĸļ

-0.74

SHIP

-0.71

éĹĺ

-0.69

shi

-0.69

hr

-0.67

aired

-0.67

ocrine

-0.66

Murray

-0.65

POSITIVE LOGITS

 Suite

0.86

 Linux

0.82

0.80

 Express

0.79

 codec

0.78

 Gmail

0.77

Windows

0.77

 apps

0.76

 Keyboard

0.76

 GNOME

0.75

Activations Density 0.122%

references to Microsoft Office-related terms

mentions of Microsoft Office and its related products

No Comments

No Known Activations