INDEX

Explanations

terms related to medical conditions and their associations with risk and treatment

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

Follows "is," "are," or "has"

np_acts-logits-general · gemini-2.0-flash

is, are, has

np_acts-logits-general · gemini-2.5-flash-lite

New Auto-Interp

Top Features by Cosine Similarity

Comparing With GEMMA-2-9B-IT @ 20-gemmascope-res-16k

Configuration

google/gemma-scope-9b-it-res/layer_20/width_16k/average_l0_91

Prompts (Dashboard)

24,576 prompts, 128 tokens each

Dataset (Dashboard)

monology/pile-uncopyrighted

Features

16,384

Data Type

float32

Hook Name

blocks.20.hook_resid_post

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

NameInMap

-0.69

:✨

-0.57

AndEndTag

-0.45

VolleyError

-0.44

Blas

-0.43

 floppy

-0.43

buff

-0.42

存于互联网档案馆

-0.42

ograms

-0.42

letter

-0.42

POSITIVE LOGITS

 frecuentemente

0.44

 príncipe

0.41

 frequently

0.41

}))

0.39

 commonly

0.39

 lentejuelas

0.39

}}_

0.38

 princesas

0.38

PYX

0.38

 Nachweis

0.37

Activations Density 2.234%

terms related to medical conditions and their associations with risk and treatment

Follows "is," "are," or "has"

is, are, has

No Comments

No Known Activations

terms related to medical conditions and their associations with risk and treatment

Follows "is," "are," or "has"

is, are, has

No Comments

No Known Activations