INDEX

Explanations

keywords related to medical conditions and treatments

oai_token-act-pair · gpt-3.5-turbo Triggered by @bot

references to diseases and health-related issues

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

New Auto-Interp

Top Features by Cosine Similarity

Comparing With GPT2-SMALL @ 12-res-jb

Configuration

jbloom/GPT2-Small-SAEs-Reformatted/blocks.11.hook_resid_post

Prompts (Dashboard)

24,576 prompts, 128 tokens each

Dataset (Dashboard)

Skylion007/openwebtext

Features

24,576

Data Type

torch.float32

Hook Point

blocks.11.hook_resid_post

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

displayText

-0.99

 Aerospace

-0.78

 GoPro

-0.76

soDeliveryDate

-0.76

Catalog

-0.76

 Garage

-0.74

 Intercept

-0.73

VO

-0.73

DF

-0.72

 Electronic

-0.72

POSITIVE LOGITS

 plague

1.89

 tuberculosis

1.72

 disease

1.71

 diseases

1.69

 malaria

1.66

pox

1.65

 Plague

1.58

 typh

1.58

 sickness

1.56

 parasites

1.52

Activations Density 0.610%

keywords related to medical conditions and treatments

references to diseases and health-related issues

No Comments

No Known Activations