INDEX

Explanations

elements of code and syntax used in programming or data structures

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

sides, temporary, SC

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

InitVars

-0.57

 zwiſchen

-0.56

EReference

-0.53

onViewCreated

-0.53

 anastom

-0.52

 papilla

-0.52

 itemName

-0.52

 setName

-0.51

 Negotiating

-0.51

 Dapper

-0.51

POSITIVE LOGITS

:✨

0.63

 respectively

0.49

 respectivamente

0.43

 poř

0.37

respectively

0.37

 Egipto

0.36

package

0.35

 rumahnya

0.35

Personensuche

0.35

 bandeira

0.34

Activations Density 0.008%

elements of code and syntax used in programming or data structures

sides, temporary, SC

No Comments

No Known Activations