INDEX

Explanations

references to nurses or nursing-related terms

oai_token-act-pair · gpt-3.5-turbo

mentions of nurses in various contexts

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

PlotsExplanationShow Test FieldDefault Test Text

IFrame

Link

Not in Any Lists

No Comments

Negative Logits

eous

-0.86

Dest

-0.85

 Dest

-0.75

 Formation

-0.72

Vers

-0.71

 Compatibility

-0.71

classic

-0.70

 Armageddon

-0.67

 RELE

-0.66

Classic

-0.65

POSITIVE LOGITS

 nurse

3.67

 nurses

3.46

 Nurse

2.82

 Nurs

2.33

 nursing

2.25

 Nursing

2.07

 nurs

1.65

 paramed

1.63

Doctors

1.61

 paramedics

1.59

Activations Density 0.017%

references to nurses or nursing-related terms

mentions of nurses in various contexts

No Comments

No Known Activations