INDEX

Explanations

discussions about options and advisories related to conditional situations or decisions

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

treatment and therapy options

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

#+#

-0.47

 himo

-0.45

styleType

-0.38

prefixer

-0.37

companying

-0.35

ruct

-0.35

oneofs

-0.34

 condens

-0.34

ện

-0.34

تفصیلات

-0.34

POSITIVE LOGITS

évaluateur

0.57

berdayakan

0.53

 juſ

0.48

 Majefty

0.47

astéroïdes

0.46

 bolsillos

0.46

Населення

0.45

UserScript

0.44

Autoritní

0.44

 Chriftian

0.44

Activations Density 0.078%

discussions about options and advisories related to conditional situations or decisions

treatment and therapy options

No Comments

No Known Activations