INDEX

Explanations

abbreviations, codes, or acronyms related to units or categories, particularly in contexts like finance or metrics

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

Krebs solution components

np_acts-logits-general · gemini-2.5-flash-lite

New Auto-Interp

Top Features by Cosine Similarity

Comparing With GEMMA-2-9B-IT @ 31-gemmascope-res-131k

Configuration

google/gemma-scope-9b-it-res/layer_31/width_131k/average_l0_109

Prompts (Dashboard)

24,576 prompts, 128 tokens each

Dataset (Dashboard)

monology/pile-uncopyrighted

Features

131,072

Data Type

float32

Hook Name

blocks.31.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

-0.40

 Holz

-0.38

<0xF0>

-0.37

 Bör

-0.37

={`${

-0.37

-0.36

 Grun

-0.35

po

-0.35

ر

-0.35

POSITIVE LOGITS

RegressionTest

0.90

 الحره

0.76

 poffible

0.73

Personensuche

0.72

 myſelf

0.72

 pleaſure

0.68

OGND

0.68

 deſt

0.67

دانشنامهٔ

0.67

 becauſe

0.67

Activations Density 0.005%

abbreviations, codes, or acronyms related to units or categories, particularly in contexts like finance or metrics

Krebs solution components

No Comments

No Known Activations