INDEX

Explanations

terms related to scale and scaling in various contexts

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

Texts containing the word "scale"

np_acts-logits-general · gemini-2.0-flash

scale, scalable, and scope

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



-0.50

Pax

-0.44

 Mite

-0.44

 Fort

-0.44

 Position

-0.43

argout

-0.43

Fir

-0.42

 Thing

-0.42

ite

-0.41

iron

-0.41

POSITIVE LOGITS

SequentialGroup

0.79

scale

0.72

 pleaſure

0.72

 jScrollPane

0.71

Scale

0.70

scope

0.68

SBATCH

0.67

 SCALE

0.66

 scale

0.65

 escala

0.65

Activations Density 0.290%

terms related to scale and scaling in various contexts

Texts containing the word "scale"

scale, scalable, and scope

No Comments

No Known Activations