INDEX

Explanations

dates or specific numbers

oai_token-act-pair · gpt-3.5-turbo

New Auto-Interp

Top Features by Cosine Similarity

Comparing With GPT2-SMALL @ 9-res-jb

Configuration

jbloom/GPT2-Small-SAEs-Reformatted/blocks.9.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.9.hook_resid_pre

Architecture

standard

Context Size

128

Dataset

Skylion007/openwebtext

Hook Point Layer

Activation Function

relu

Embeds

IFrame

Link

Not in Any Lists

No Comments

Negative Logits

 serpent

-0.80

dem

-0.79

 phased

-0.78

 mach

-0.75

 predic

-0.75

 brid

-0.74

 estranged

-0.74

 glac

-0.74

 antit

-0.74

 Rebell

-0.73

POSITIVE LOGITS

Conclusion

1.39

Secondly

1.37

Again

1.33

Advertisement

1.32

VII

1.32

Bonus

1.32

Others

1.30

Comment

1.29

Also

1.28

Loading

1.27

Activations Density 0.197%

dates or specific numbers

No Comments

No Known Activations