INDEX

Explanations

words related to rules and regulations

oai_token-act-pair · gpt-3.5-turbo

terms related to governance and regulation

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

 Finder

-0.70

 Slay

-0.69

 Stanley

-0.66

 Depot

-0.66

Plot

-0.65

 proverb

-0.64

;;;;;;;;;;;;

-0.63

Champ

-0.62

 Seymour

-0.61

orge

-0.59

POSITIVE LOGITS

 regulating

0.94

 commerce

0.84

 dealing

0.83

regulated

0.80

 protecting

0.80

 accessing

0.79

 regulated

0.78

 enforcing

0.78

glers

0.77

erous

0.74

Activations Density 0.292%

words related to rules and regulations

terms related to governance and regulation

No Comments

No Known Activations