INDEX

Explanations

instances of verbal and physical harassment

oai_token-act-pair · gpt-3.5-turbo

occurrences and discussions surrounding harassment, particularly sexual harassment

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

New Auto-Interp

Top Features by Cosine Similarity

Comparing With GPT2-SMALL @ 10-res-jb

Configuration

jbloom/GPT2-Small-SAEs-Reformatted/blocks.10.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.10.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

éĹĺ

-0.96

ethe

-0.74

NAS

-0.72

ACTED

-0.72

export

-0.71

 Wonders

-0.70

zyme

-0.69

rient

-0.69

olesterol

-0.67

coat

-0.67

POSITIVE LOGITS

 harassment

1.06

 harassing

1.00

 accus

0.98

 allegations

0.97

 tactics

0.96

 accusations

0.93

 harass

0.92

 leveled

0.91

 complaints

0.90

 stalking

0.88

Activations Density 0.072%

instances of verbal and physical harassment

occurrences and discussions surrounding harassment, particularly sexual harassment

No Comments

No Known Activations