INDEX

Explanations

patterns related to autogenerated code and comments

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

Mentions auto-generated code or methods

np_acts-logits-general · gemini-2.0-flash

auto-generated code and comments

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

New Auto-Interp

Top Features by Cosine Similarity

Comparing With GEMMA-2-2B @ 22-gemmascope-res-16k

Configuration

google/gemma-scope-2b-pt-res/layer_22/width_16k/average_l0_72

Prompts (Dashboard)

36,864 prompts, 128 tokens each

Dataset (Dashboard)

monology/pile-uncopyrighted

Features

16,384

Data Type

float32

Hook Name

blocks.22.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

ArgumentParser

-0.68

AddHtmlAttribute

-0.68

ErrorListener

-0.68

ItemBackground

-0.66

nocache

-0.63

 SqlDbType

-0.62

pulumi

-0.61

 Monfieur

-0.61

bitField

-0.61

parsedMessage

-0.60

POSITIVE LOGITS

GEBURTSDATUM

0.66

 code

0.66



0.59

Boiler

0.54

 délib

0.54

.');

0.51

 generated

0.48

первых

0.48

 boiler

0.48

skeleton

0.48

Activations Density 0.134%

patterns related to autogenerated code and comments

Mentions auto-generated code or methods

auto-generated code and comments

No Comments

No Known Activations