INDEX

Explanations

numerical values and measurements

oai_token-act-pair · gpt-3.5-turbo

numerical values and statistical data

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

 withd

-0.73

 proble

-0.68

 spokes

-0.66

 undermin

-0.63

 accordingly

-0.62

 foremost

-0.61

 obser

-0.61

.}

-0.60

 respons

-0.60

 conscientious

-0.60

POSITIVE LOGITS

 APPLIC

0.85

 âľĵ

0.78

 Snapdragon

0.77

0.76

 Flavoring

0.73

Avg

0.72

Avg

0.69

 Female

0.68

 TOTAL

0.66

Activations Density 0.217%

numerical values and measurements

numerical values and statistical data

No Comments

No Known Activations