INDEX

Explanations

phrases related to accusations or denials of wrongdoing

oai_token-act-pair · gpt-3.5-turbo

references to violations or misconduct

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

New Auto-Interp

Top Features by Cosine Similarity

Comparing With GPT2-SMALL @ 6-res-jb

Configuration

jbloom/GPT2-Small-SAEs-Reformatted/blocks.6.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.6.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

etically

-0.84

NetMessage

-0.83

pop

-0.67

osphere

-0.66

etics

-0.64

etic

-0.64

iku

-0.62

population

-0.61

ofi

-0.61

 butterflies

-0.59

POSITIVE LOGITS

 wrongdoing

1.03

 misconduct

0.94

 allegations

0.85

 perpetrated

0.85

 unfocusedRange

0.79

 involving

0.78

 targeting

0.78

 disclosures

0.77

ionage

0.77

onduct

0.76

Activations Density 0.066%

phrases related to accusations or denials of wrongdoing

references to violations or misconduct

No Comments

No Known Activations