INDEX

Explanations

code structure elements, specifically focusing on class and method identifiers

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

code declaration

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

ryd

-0.45

adesivo

-0.43

 SafeMath

-0.42

 advised

-0.41

geist

-0.40

⟡

-0.40

selaer

-0.40

nă

-0.40

 Qaraldi

-0.40

reathe

-0.39

POSITIVE LOGITS

 propOrder

0.69

Personendaten

0.68

 initComponents

0.60

setVerticalGroup

0.60

WithIOException

0.54

 <<<<<<<<<<<<<<

0.53

 الحره

0.49

aarrggbb

0.46

IntoConstraints

0.46

fillType

0.45

Activations Density 0.001%

code structure elements, specifically focusing on class and method identifiers

code declaration

No Comments

No Known Activations