INDEX

Explanations

phrases related to discrediting or disproving information

oai_token-act-pair · gpt-3.5-turbo

terms related to the disproval or debunking of claims and theories

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

Trade

-0.73

Cho

-0.70

 Gifts

-0.67

 adventurous

-0.64

 Warehouse

-0.64

 Controls

-0.64

 Overse

-0.62

 Trading

-0.62

ole

-0.62

Control

-0.61

POSITIVE LOGITS

 debunked

2.83

 dispro

2.58

 refuted

2.46

 debunk

2.31

 refute

2.21

 discredited

2.16

 corrobor

1.89

 vind

1.81

 contradicted

1.74

 substant

1.74

Activations Density 0.034%

phrases related to discrediting or disproving information

terms related to the disproval or debunking of claims and theories

No Comments

No Known Activations