INDEX

Explanations

references to the term "automatic."

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

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

PlotsExplanationShow Test FieldDefault Test Text

IFrame

Link

Not in Any Lists

No Comments

Negative Logits

 Gegend

-0.51

 kajian

-0.49

 ámbitos

-0.46

urilor

-0.46

 Inscrivez

-0.45

 Encuentra

-0.44

 Söz

-0.44

 muestras

-0.44

Retrouvez

-0.42

Partager

-0.42

POSITIVE LOGITS

Automatic

1.07

automatic

1.02

 Automatic

1.02

 automatic

0.92

 Automat

0.89

 AUTOMATIC

0.88

 Autom

0.88

 automat

0.87

Automat

0.87

Autowired

0.79

Activations Density 0.005%

references to the term "automatic."

No Comments

No Known Activations