INDEX

Explanations

references to common factors and divisors, particularly in mathematical contexts

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

New Auto-Interp

Top Features by Cosine Similarity

Comparing With GEMMA-2-2B @ 20-gemmascope-res-16k

Configuration

google/gemma-scope-2b-pt-res/layer_20/width_16k/average_l0_71

Prompts (Dashboard)

36,864 prompts, 128 tokens each

Dataset (Dashboard)

monology/pile-uncopyrighted

Features

16,384

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

PlotsExplanationShow Test FieldDefault Test Text

IFrame

Link

Not in Any Lists

No Comments

Negative Logits

 يتيمه

-0.85

تقاوى

-0.79

 propOrder

-0.74

Rhestr

-0.73

脚注の使い方

-0.69

 quizá

-0.68

 صوتيه

-0.68

 chré

-0.67

 <=",

-0.67

principalColumn

-0.67

POSITIVE LOGITS

1.06

̵

0.78

 XNUMX

0.76

⁇

0.74

 analyzes

0.63

«

0.61

™

0.60

🇧

0.59

彼は

0.55

NUMX

0.55

Activations Density 0.115%

references to common factors and divisors, particularly in mathematical contexts

No Comments

No Known Activations