INDEX

Explanations

technical terms and metrics related to algorithms and statistical methods

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

Machine learning classifiers and training

np_acts-logits-general · gemini-2.0-flash

classifier classification and training

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

New Auto-Interp

Top Features by Cosine Similarity

Comparing With GEMMA-2-9B @ 30-gemmascope-res-16k

Configuration

google/gemma-scope-9b-pt-res/layer_30/width_16k/average_l0_120

Prompts (Dashboard)

24,576 prompts, 128 tokens each

Dataset (Dashboard)

monology/pile-uncopyrighted

Features

16,384

Data Type

float32

Hook Name

blocks.30.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

RefNanny

-0.47

plotly

-0.38

IGraph

-0.37

 Rhyth

-0.36

 Mino

-0.35

 JSONException

-0.35

teardown

-0.35

rix

-0.34

 JSTOR

-0.34

gae

-0.34

POSITIVE LOGITS

 classifiers

0.72

 classifier

0.69

 training

0.69

 discriminant

0.67

 TRAINING

0.65

training

0.63

Training

0.63

 kasarigan

0.62

classifiers

0.61

 classification

0.60

Activations Density 0.868%

technical terms and metrics related to algorithms and statistical methods

Machine learning classifiers and training

classifier classification and training

No Comments

No Known Activations