INDEX

Explanations

references to the use of the init method and related file attributes in Python code

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

`init` and `all`

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

 Puls

-0.52

iecie

-0.51

 tans

-0.49

 Saks

-0.47

 puls

-0.47

 drain

-0.47

瘩

-0.47

 haft

-0.46

})^{

-0.46

POSITIVE LOGITS

 feroit

0.73

tvguidetime

0.68

 maș

0.67

 avoient

0.66

MemoryWarning

0.65

 ainfi

0.65

 étoient

0.65

 esimer

0.63

zzleHttp

0.63

 săptăm

0.63

Activations Density 0.012%

references to the use of the __init__ method and related file attributes in Python code

`__init__` and `__all__`

No Comments

No Known Activations

references to the use of the init method and related file attributes in Python code

`init` and `all`