INDEX

Explanations

descriptive terms related to body size and shape

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

obesity and weight

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

pulseira

-0.45

 miniaturka

-0.44

 buurt

-0.42

 verhalen

-0.42

Abitanti

-0.42

 penutup

-0.42

 Verhält

-0.41

seamnă

-0.41

Tarih

-0.41

 bluzka

-0.41

POSITIVE LOGITS

fat

0.84

 overweight

0.73

 obese

0.72

 obesity

0.67

Fat

0.66

Fat

0.65

FAT

0.62

 gordo

0.61

 obes

0.61

 Obesity

0.59

Activations Density 0.018%

descriptive terms related to body size and shape

obesity and weight

No Comments

No Known Activations