INDEX

Explanations

descriptions of features and characteristics of various items or products

oai_token-act-pair · gpt-3.5-turbo

New Auto-Interp

Top Features by Cosine Similarity

Comparing With GPT2-SMALL @ 9-res-jb

Configuration

jbloom/GPT2-Small-SAEs-Reformatted/blocks.9.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.9.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

 Transcript

-0.80

 Facts

-0.71

 Written

-0.70

 documents

-0.67

 Names

-0.67

 Announce

-0.67

Aad

-0.65

 Newsp

-0.65

 Politics

-0.64

POSITIVE LOGITS

 manageable

0.89

 comfortably

0.88

 noticeably

0.88

 heavier

0.87

 finer

0.86

 overpowered

0.85

 smoother

0.85

 bottleneck

0.85

 overpower

0.83

 cramped

0.81

Activations Density 0.960%

descriptions of features and characteristics of various items or products

No Comments

No Known Activations