INDEX

Explanations

terms related to medical risks and physical conditions

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

time, level, damage, populations, amplitude

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-131k

Configuration

google/gemma-scope-9b-it-res/layer_20/width_131k/average_l0_81

Prompts (Dashboard)

24,576 prompts, 128 tokens each

Dataset (Dashboard)

monology/pile-uncopyrighted

Features

131,072

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

pos

-0.32

 factors

-0.31

tehen

-0.30

 urbanas

-0.30

niaus

-0.30

 loadData

-0.29

 zand

-0.28

 Simple

-0.28

 planas

-0.28

 otomatig

-0.28

POSITIVE LOGITS



0.70

awtextra

0.57

tvguidetime

0.57

الحياه

0.57

 propOrder

0.54

RenderAtEndOf

0.54

存于互联网档案馆

0.52

complexType

0.52

şört

0.50

новниш

0.50

Activations Density 0.113%

terms related to medical risks and physical conditions

time, level, damage, populations, amplitude

No Comments

No Known Activations