INDEX

Explanations

phrases indicating the completion or finalization of an investigation or judgment

oai_token-act-pair · gpt-3.5-turbo

instances of the word "concluded" 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 @ 1-res-jb

Configuration

jbloom/GPT2-Small-SAEs-Reformatted/blocks.1.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.1.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

enta

-0.81

cler

-0.77

activated

-0.73

arta

-0.72

vag

-0.71

ede

-0.69

heter

-0.67

GB

-0.66

houses

-0.66

otin

-0.66

POSITIVE LOGITS

 concludes

1.04

 concluded

1.00

 concluding

0.98

 conclude

0.95

 conclusion

0.79

reement

0.78

lehem

0.75

ãĤ¤ãĥĪ

0.75

 Conclusion

0.73

olicy

0.72

Activations Density 0.006%

phrases indicating the completion or finalization of an investigation or judgment

instances of the word "concluded" in various contexts

No Comments

No Known Activations