INDEX

Explanations

references to concepts of entropy and elliptical shapes

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

entropy change and difference

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

<h2>

-0.65

-0.47

an

-0.46

“

-0.46

er

-0.44

ان

-0.44

-0.43

<blockquote>

-0.42

POSITIVE LOGITS

majánló

0.87

 desmotivaciones

0.84

iſchen

0.81

 Entropy

0.81

<unused41>

0.78

<unused14>

0.78

<unused79>

0.78

Entropy

0.78

[@BOS@]

0.77

<unused8>

0.77

Activations Density 0.020%

references to concepts of entropy and elliptical shapes

entropy change and difference

No Comments

No Known Activations