INDEX

Explanations

mentions of individuals or things at a high level in a hierarchy or organization

oai_token-act-pair · gpt-3.5-turbo

references to high-level positions or roles in organizations and government

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

New Auto-Interp

Top Features by Cosine Similarity

Comparing With GPT2-SMALL @ 4-res-jb

Configuration

jbloom/GPT2-Small-SAEs-Reformatted/blocks.4.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.4.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

 Lucia

-0.85

Stars

-0.73

 Predators

-0.72

 Giov

-0.71

 Stars

-0.71

 Norris

-0.69

 Andromeda

-0.69

OPLE

-0.68

 Kenobi

-0.67

Frames

-0.67

POSITIVE LOGITS

level

0.86

 level

0.84

 levels

0.74

 sophistication

0.74

 guidance

0.74

 representation

0.72

 delegation

0.72

abulary

0.70

 elevation

0.68

holder

0.67

Activations Density 0.019%

mentions of individuals or things at a high level in a hierarchy or organization

references to high-level positions or roles in organizations and government

No Comments

No Known Activations