INDEX

Explanations

words related to legitimacy

oai_token-act-pair · gpt-3.5-turbo

concepts related to legitimacy and legality

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

New Auto-Interp

Top Features by Cosine Similarity

Comparing With GPT2-SMALL @ 0-res-jb

Configuration

jbloom/GPT2-Small-SAEs-Reformatted/blocks.0.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.0.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

feld

-0.67

Visual

-0.65

TL

-0.63

Sym

-0.63

severe

-0.62

LIN

-0.62

 Tank

-0.61

 Annotations

-0.61

reflect

-0.60

 diving

-0.60

POSITIVE LOGITS

 legitimacy

3.32

 illegitimate

2.09

 legality

1.83

 validity

1.77

 legitim

1.71

 legitimate

1.44

 authenticity

1.34

 illeg

1.32

 legit

1.30

 unlaw

1.24

Activations Density 0.015%

words related to legitimacy

concepts related to legitimacy and legality

No Comments

No Known Activations