INDEX

Explanations

code-related structures or keywords in programming context

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

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

forChild

-0.51

staticmethod

-0.50

Parcelize

-0.49

 especialistas

-0.46

AddField

-0.45

quettes

-0.43

 Polsek

-0.43

Inputs

-0.42

smarty

-0.41

 specialist

-0.41

POSITIVE LOGITS

contentLoaded

0.89

########.

0.82

 MaterialApp

0.81

GraphicsUnit

0.78

 Roskov

0.77

 kasarigan

0.77

 ReactDOM

0.73

forRoot

0.73

 betweenstory

0.72

 snippetHide

0.71

Activations Density 0.120%

code-related structures or keywords in programming context

No Comments

No Known Activations