INDEX

Explanations

key concepts related to dimensionality and computational complexity in machine learning and statistical modeling

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

large scale complexity

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

 heureux

-0.30

 conmigo

-0.30

 adaptado

-0.28

 morada

-0.27

 separados

-0.27

 contigo

-0.25

ęty

-0.25

 establecidos

-0.25

 servir

-0.25

 invitado

-0.25

POSITIVE LOGITS

 AssemblyCulture

0.85

 resourceCulture

0.77

تقاوى

0.76

UrlResolution

0.75

<unused3>

0.73

[@BOS@]

0.73

<unused28>

0.73

<unused43>

0.73

<unused8>

0.73

<unused14>

0.73

Activations Density 0.161%

key concepts related to dimensionality and computational complexity in machine learning and statistical modeling

large scale complexity

No Comments

No Known Activations