INDEX

Explanations

comparisons between different versions of movies or adaptations

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

adaptation, inaccurate, no relationship

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

 المعيارى

-0.71

ſchen

-0.69

pecabe

-0.67

RenderAtEndOf

-0.65

CloseOperation

-0.64

 westfalen

-0.63

guiente

-0.62

ſcher

-0.62

iſen

-0.62

ientras

-0.61

POSITIVE LOGITS

hyrchwyd

0.33

ंदीखरीदारी

0.31

 spin

0.30

 libre

0.28

 story

0.26

MLLoader

0.25

cast

0.23

 largely

0.23

محمد

0.23

 storyline

0.23

Activations Density 0.015%

comparisons between different versions of movies or adaptations

adaptation, inaccurate, no relationship

No Comments

No Known Activations