INDEX

Explanations

phrases indicating suggestions or implications

oai_token-act-pair · gpt-3.5-turbo

suggestive phrases that indicate reasoning or conclusion

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

New Auto-Interp

Top Features by Cosine Similarity

Comparing With GPT2-SMALL @ 5-res-jb

Configuration

jbloom/GPT2-Small-SAEs-Reformatted/blocks.5.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.5.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

fare

-0.75

uss

-0.69

EEE

-0.68

 presided

-0.67

ppa

-0.63

skill

-0.61

gie

-0.60

 fought

-0.60

FO

-0.60

imb

-0.59

POSITIVE LOGITS

ively

0.78

 indications

0.76

 indicators

0.71

evidence

0.70

uggest

0.70

unct

0.70

kinson

0.69

ered

0.68

ausible

0.68

ifiable

0.67

Activations Density 0.068%

phrases indicating suggestions or implications

suggestive phrases that indicate reasoning or conclusion

No Comments

No Known Activations