INDEX

Explanations

words related to language and communication, with a focus on specific names or terms

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

proletarian, proviral, prohibitive

np_acts-logits-general · gemini-2.5-flash-lite

New Auto-Interp

Top Features by Cosine Similarity

Comparing With GEMMA-2-9B-IT @ 9-gemmascope-res-131k

Configuration

google/gemma-scope-9b-it-res/layer_9/width_131k/average_l0_121

Prompts (Dashboard)

24,576 prompts, 128 tokens each

Dataset (Dashboard)

monology/pile-uncopyrighted

Features

131,072

Data Type

float32

Hook Name

blocks.9.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

 légitime

-0.32

 Policía

-0.32

 belleza

-0.32

 Nuss

-0.31

Saison

-0.31

 petugas

-0.31

atients

-0.31

 lhes

-0.31

ัง

-0.31

xbf

-0.31

POSITIVE LOGITS

 про

0.90

 Про

0.75

про

0.73

 ujednoznacz

0.73

Про

0.71

 przes

0.69

 ПРО

0.66

 contextLoads

0.64

 HasFactory

0.64

 przec

0.61

Activations Density 0.005%

words related to language and communication, with a focus on specific names or terms

proletarian, proviral, prohibitive

No Comments

No Known Activations