INDEX

Explanations

references to coding libraries or frameworks

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

R getResources

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

ロウィン

-0.82

httphttps

-0.79

الحياه

-0.78

ftagPool

-0.78

+:+

-0.76

AddTagHelper

-0.76

ſſung

-0.75

 المعيارى

-0.73

iſche

-0.70



-0.70

POSITIVE LOGITS

0.45

 désert

0.33

0.31

 doigt

0.29

0.28

 règne

0.28

0.27

Activations Density 0.000%

references to coding libraries or frameworks

R getResources

No Comments

No Known Activations