INDEX

Explanations

technical terms and concepts related to medical or biological topics

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

material structure components

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

the neuron detects technical, scientific, or domain-specific terminology (specialized nouns and descriptors found in medical, biological, engineering, and patent-like texts).

oai_token-act-pair · gpt-5-mini Triggered by @vetterc0

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

تقاوى

-0.66

ніципалі

-0.59

+#+

-0.59

 betweenstory

-0.56



-0.55

省市镇

-0.54

 виправивши

-0.54

 незавершена

-0.50

 tartalomajánló

-0.49

 Alquiler

-0.48

POSITIVE LOGITS

########.

0.45

 resourceCulture

0.45

 AssemblyCulture

0.43

 مرئيه

0.40

GHIJKLM

0.37

styleType

0.36

зулта

0.35

Personendaten

0.35

parsedMessage

0.35

TextWatcher

0.34

Activations Density 39.194%

technical terms and concepts related to medical or biological topics

material structure components

the neuron detects technical, scientific, or domain-specific terminology (specialized nouns and descriptors found in medical, biological, engineering, and patent-like texts).

No Comments

No Known Activations