INDEX

Explanations

medical conditions related to inflammation

oai_token-act-pair · gpt-3.5-turbo

words related to health conditions and injuries

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

 conclusion

-0.72

 intervals

-0.70

 review

-0.70

 pole

-0.70

 poles

-0.70

 initiation

-0.65

Fin

-0.64

 Quit

-0.63

 buckle

-0.63

 premises

-0.62

POSITIVE LOGITS

amed

4.52

aming

2.62

ames

2.09

ame

1.80

ammed

1.64

amer

1.58

ams

1.53

AMES

1.51

AME

1.49

ammad

1.46

Activations Density 0.005%

medical conditions related to inflammation

words related to health conditions and injuries

No Comments

No Known Activations