INDEX

Explanations

structural elements and identifiers within code documentation

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

user input follows `<start_of_turn>`

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

apimachinery

-0.43

="@+

-0.41

Chham

-0.40

WriteBarrier

-0.39

 uLocal

-0.39

XmlIgnore

-0.38

 AssemblyCulture

-0.36

AndEndTag

-0.36

 Vikipedi

-0.36

tagext

-0.36

POSITIVE LOGITS

integrating

0.50

hyrchwyd

0.47

mapStateToProps

0.45

 Integrating

0.45

mbggenerated

0.44

WriteLiteral

0.43

wakilan

0.42

joueurs

0.42

Integrating

0.42

 Alva

0.41

Activations Density 0.190%

structural elements and identifiers within code documentation

user input follows `<start_of_turn>`

No Comments

No Known Activations