INDEX

Explanations

conclusion-related terms and phrases

oai_token-act-pair · gpt-3.5-turbo

references to conclusions or results drawn from analysis or discussion

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

ophon

-0.70

bid

-0.68

nas

-0.68

amen

-0.68

maid

-0.66

inventory

-0.65

cler

-0.64

unte

-0.63

 talents

-0.63

tu

-0.62

POSITIVE LOGITS

 conclusion

0.96

 conclusions

0.94

0.81

éĸ

0.79

 Conclusion

0.73

 regarding

0.68

 concludes

0.67

 drawn

0.67

 paralysis

0.66

 concluding

0.66

Activations Density 0.014%

conclusion-related terms and phrases

references to conclusions or results drawn from analysis or discussion

No Comments

No Known Activations