INDEX

Explanations

references to companies and their roles or statuses in a given context

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

companies and organizations

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

Географија

-0.27

 foule

-0.27

 automatiques

-0.26

 saurait

-0.25

 role

-0.24

г

-0.23

 problém

-0.23

 perioada

-0.23

ごろ

-0.22

 color

-0.22

POSITIVE LOGITS

ագրություններ

0.79

 يتيمه

0.70

 الحره

0.66

tagext

0.65

ContentLoaded

0.62

MLLoader

0.61

 comuniques

0.59

RenderAtEndOf

0.58

celotti

0.58

rungsseite

0.57

Activations Density 0.017%

references to companies and their roles or statuses in a given context

companies and organizations

No Comments

No Known Activations