INDEX

Explanations

code-related keywords and constructs in programming languages

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

C++ or Elixir namespace and module declarations

np_acts-logits-general · gemini-2.0-flash

programming keywords and statements

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-16k

Configuration

google/gemma-scope-9b-it-res/layer_20/width_16k/average_l0_91

Prompts (Dashboard)

24,576 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

IFrame

Link

Not in Any Lists

No Comments

Negative Logits

volezza

-0.48

 للمعارف

-0.45

fhir

-0.44

BeginContext

-0.42

itinéraire

-0.41

 lenker

-0.41

ainville

-0.40

InputBorder

-0.40

 ModelRenderer

-0.40

mdl

-0.39

POSITIVE LOGITS

帖最后由

0.64

 static

0.54

static

0.53

 dynamique

0.52

usercontent

0.51

staticmethod

0.49

optionalTypeArgs

0.49

CppMethod

0.49

ंदीखरीदारी

0.48

 Wikiseite

0.47

Activations Density 0.011%

code-related keywords and constructs in programming languages

C++ or Elixir namespace and module declarations

programming keywords and statements

No Comments

No Known Activations