INDEX

Explanations

sentences or phrases followed by a specific symbol or character sequence

oai_token-act-pair · gpt-3.5-turbo

punctuation and sentences that imply completion or conclusion

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

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

 presumably

-0.93

 grop

-0.92

 hypot

-0.85

pse

-0.83

 upset

-0.82

 speculated

-0.82

 sucker

-0.81

 unexplained

-0.81

 censored

-0.81

 unidentified

-0.80

POSITIVE LOGITS

Features

1.61

Learn

1.61

Our

1.51

Join

1.50

Contact

1.48

Whether

1.43

Discover

1.43

Through

1.42

Together

1.39

Visit

1.39

Activations Density 0.374%

sentences or phrases followed by a specific symbol or character sequence

punctuation and sentences that imply completion or conclusion

No Comments

No Known Activations