INDEX

Explanations

terms related to specific types of medical treatments and conditions

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

solid lines

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

UnusedPrivate

-0.77

 tartalomajánló

-0.75

aarrggbb

-0.68

richTextPanel

-0.68

PreferredItem

-0.66

matchCondition

-0.63

AutoScaleMode

-0.61

writeField

-0.61

HtmlAttribute

-0.60

 للمعارف

-0.59

POSITIVE LOGITS

 turística

0.34

र्शन

0.31

čenje

0.30

 hosped

0.30

ütfen

0.30

but

0.29

Flammability

0.29

 clientèle

0.29

yet

0.28

(«

0.28

Activations Density 0.099%

terms related to specific types of medical treatments and conditions

solid lines

No Comments

No Known Activations