INDEX

Explanations

numerical quantities or comparatives followed by a verb

oai_token-act-pair · gpt-3.5-turbo

instances of the word "more" indicating significant quantities or increases

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

New Auto-Interp

Top Features by Cosine Similarity

Comparing With GPT2-SMALL @ 6-res-jb

Configuration

jbloom/GPT2-Small-SAEs-Reformatted/blocks.6.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.6.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

ĺħ

-0.83

xtap

-0.82

 Conclusion

-0.75

imaru

-0.73

Fram

-0.72

ÃŁ

-0.72

ishops

-0.70

Quote

-0.69

Guard

-0.68

 Practices

-0.67

POSITIVE LOGITS

 than

1.28

 stringent

0.94

 importantly

0.85

 sophisticated

0.84

than

0.82

 Than

0.81

 detailed

0.80

 expensive

0.79

 comprehensive

0.78

ado

0.78

Activations Density 0.139%

numerical quantities or comparatives followed by a verb

instances of the word "more" indicating significant quantities or increases

No Comments

No Known Activations