INDEX

Explanations

terms related to conclusions or verdicts in legal contexts

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

New Auto-Interp

Top Features by Cosine Similarity

Comparing With GEMMA-2-2B @ 1-gemmascope-mlp-16k

Configuration

google/gemma-scope-2b-pt-mlp/layer_1/width_16k/average_l0_105

Prompts (Dashboard)

24,576 prompts, 128 tokens each

Dataset (Dashboard)

monology/pile-uncopyrighted

Features

16,384

Data Type

float32

Hook Name

blocks.1.hook_mlp_out

Hook Layer

Architecture

jumprelu

Context Size

1,024

Dataset

monology/pile-uncopyrighted

Activation Function

relu

Embeds

IFrame

Link

Not in Any Lists

No Comments

Negative Logits

 conclusion

-2.66

 conclude

-2.45

conclusion

-2.33

 concluded

-2.25

 Conclusion

-2.14

 concludes

-2.14

 CONCLUSION

-1.99

Conclusion

-1.99

 conclusions

-1.97

 concluding

-1.91

POSITIVE LOGITS

utsches

0.56

MLLoader

0.54

 scala

0.51

uminense

0.51

채

0.50

 Penh

0.49

mixin

0.49

 solide

0.48

rairie

0.47

 referenties

0.47

Activations Density 0.196%

terms related to conclusions or verdicts in legal contexts

No Comments

No Known Activations