INDEX

Explanations

references to traditional practices and concepts across various contexts

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

preceding "traditional" or "conventional"

np_acts-logits-general · gemini-2.0-flash

traditional versus conventional methods

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

New Auto-Interp

Top Features by Cosine Similarity

Comparing With GEMMA-2-2B @ 20-gemmascope-res-16k

Configuration

google/gemma-scope-2b-pt-res/layer_20/width_16k/average_l0_71

Prompts (Dashboard)

36,864 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

PlotsExplanationShow Test FieldDefault Test Text

IFrame

Link

Not in Any Lists

No Comments

Negative Logits

xase

-0.59

ValueStyle

-0.59

AddTagHelper

-0.54

InstanceState

-0.54

 AssemblyTitle

-0.53

 officielles

-0.53

ніципалі

-0.53

 generally

-0.53

 virš

-0.52

 keber

-0.51

POSITIVE LOGITS

ists

0.92

 wisdom

0.81

izing

0.77

적으로

0.76

적인

0.73

ising

0.70

izes

0.70

ism

0.68

 methods

0.68

weise

0.67

Activations Density 0.158%

references to traditional practices and concepts across various contexts

preceding "traditional" or "conventional"

traditional versus conventional methods

No Comments

No Known Activations