INDEX

Explanations

phrases related to forceful actions or interventions

oai_token-act-pair · gpt-3.5-turbo

references to coercive actions and violence

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

New Auto-Interp

Top Features by Cosine Similarity

Comparing With GPT2-SMALL @ 1-res-jb

Configuration

jbloom/GPT2-Small-SAEs-Reformatted/blocks.1.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.1.hook_resid_pre

Architecture

standard

Context Size

128

Dataset

Skylion007/openwebtext

Hook Point Layer

Activation Function

relu

Embeds

PlotsExplanationShow Test FieldDefault Test Text

IFrame

Link

Not in Any Lists

No Comments

Negative Logits

 Prediction

-0.71

paragraph

-0.70

ancial

-0.69

purpose

-0.69

daily

-0.68

ership

-0.68

orno

-0.67

sal

-0.67

 Purpose

-0.67

rug

-0.64

POSITIVE LOGITS

 forcefully

1.13

 forcibly

1.07

dru

0.83

 steril

0.82

 kissed

0.82

 shoved

0.82

 awoken

0.79

 overpowered

0.79

 violently

0.78

avage

0.77

Activations Density 0.011%

phrases related to forceful actions or interventions

references to coercive actions and violence

No Comments

No Known Activations