INDEX

Explanations

phrases related to holding law enforcement accountable for their actions

oai_token-act-pair · gpt-3.5-turbo

numerical references or counts in the text

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

PlotsExplanationShow Test FieldDefault Test Text

IFrame

Link

Not in Any Lists

No Comments

Negative Logits

 peripher

-0.88

 values

-0.83

 temperament

-0.81

 permitted

-0.81

 intent

-0.81

 observers

-0.80

 writ

-0.80

 excluded

-0.79

 ruled

-0.77

 means

-0.76

POSITIVE LOGITS

[+

1.08

 Ibid

1.01

 Fargo

0.88

 Beware

0.88

 Dresden

0.87

 Conclusion

0.87

 Alto

0.87

 Kinnikuman

0.84

 Glac

0.83

 Connie

0.80

Activations Density 0.072%

phrases related to holding law enforcement accountable for their actions

numerical references or counts in the text

No Comments

No Known Activations