INDEX

Explanations

identifiers or references that appear to be part of a programming or software context

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

hexadecimal identifiers

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

RenderAtEndOf

-0.76

########.

-0.69

 gynhyrchwyd

-0.69

ReusableCell

-0.68

хьтан

-0.65

Personendaten

-0.65

 المعيارى

-0.64

="@+

-0.63

 MainAxisSize

-0.63

TokenNameLBRACE

-0.63

POSITIVE LOGITS

 recomendado

0.41

 seguido

0.41

 transparencia

0.39

0.38

 calidad

0.35

 diversidad

0.34

 participado

0.34

 preparado

0.34

Recently

0.34

 तक

0.34

Activations Density 0.013%

identifiers or references that appear to be part of a programming or software context

hexadecimal identifiers

No Comments

No Known Activations