INDEX

Explanations

references to the software package manager "pip."

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

pip words and phrases

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

 autorytatywna

-0.61

 Center

-0.55

hasMany

-0.48

Center

-0.48

Authorize

-0.47

 Werth

-0.47

 amaç

-0.47

NSDate

-0.46

 World

-0.46

 longitude

-0.46

POSITIVE LOGITS

Pip

0.98

pip

0.93

Pip

0.91

PIP

0.87

PIP

0.87

pip

0.87

 pipelines

0.77

 pipette

0.68

 pipeline

0.68

 piped

0.66

Activations Density 0.002%

references to the software package manager "pip."

pip words and phrases

No Comments

No Known Activations