INDEX

Explanations

percentage values expressed in different formats

oai_token-act-pair · gpt-3.5-turbo

phrases related to statistical data and quantities

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

New Auto-Interp

Top Features by Cosine Similarity

Comparing With GPT2-SMALL @ 8-res-jb

Configuration

jbloom/GPT2-Small-SAEs-Reformatted/blocks.8.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.8.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

®

-0.70

Ü

-0.68

hovah

-0.67

\\\\\\\\

-0.64

 millenn

-0.62

ÃĥÃĤ

-0.60

rique

-0.59

PDATE

-0.58

SEE

-0.57

DEBUG

-0.56

POSITIVE LOGITS

 subtract

0.76

 oranges

0.76

 thickness

0.75

 ratio

0.74

 deductions

0.73

 cubes

0.73

 sizes

0.72

 ounces

0.72

 ratios

0.72

tein

0.71

Activations Density 1.222%

percentage values expressed in different formats

phrases related to statistical data and quantities

No Comments

No Known Activations