INDEX

Explanations

number-related terms and elements

oai_token-act-pair · gpt-3.5-turbo

references to collections or groups of items

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

New Auto-Interp

Top Features by Cosine Similarity

Comparing With GPT2-SMALL @ 10-res-jb

Configuration

jbloom/GPT2-Small-SAEs-Reformatted/blocks.10.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.10.hook_resid_pre

Architecture

standard

Context Size

128

Dataset

Skylion007/openwebtext

Hook Point Layer

Activation Function

relu

Embeds

IFrame

Link

Not in Any Lists

No Comments

Negative Logits

Nonetheless

-0.72

 thereof

-0.71

umably

-0.69

['

-0.68

 outwe

-0.68

 presumably

-0.67

None

-0.66

abal

-0.65

empt

-0.64

oubt

-0.64

POSITIVE LOGITS

 curated

1.05

 immersive

0.93

 FREE

0.89

 interactive

0.85

 customizable

0.80

 unforgettable

0.77

 âĢº

0.76

 Explore

0.76

 innovative

0.76

 personalized

0.74

Activations Density 0.786%

number-related terms and elements

references to collections or groups of items

No Comments

No Known Activations