INDEX

Explanations

keywords related to reported or alleged events or actions

oai_token-act-pair · gpt-3.5-turbo

references to allegations or claims of wrongdoing

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

New Auto-Interp

Top Features by Cosine Similarity

Comparing With GPT2-SMALL @ 3-res-jb

Configuration

jbloom/GPT2-Small-SAEs-Reformatted/blocks.3.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.3.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

OVA

-0.75

lite

-0.74

 Clicker

-0.73

ajo

-0.73

bern

-0.72

eyes

-0.70

avascript

-0.69

ARCH

-0.68

sv

-0.68

xtap

-0.68

POSITIVE LOGITS

 allegations

0.90

 accusations

0.82

 accuser

0.79

 alleged

0.73

 disclosures

0.73

 alleges

0.71

ities

0.70

 misrepresent

0.70

 abuser

0.70

 abuses

0.70

Activations Density 0.016%

keywords related to reported or alleged events or actions

references to allegations or claims of wrongdoing

No Comments

No Known Activations