INDEX

Explanations

references to numerical data and statistics

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

New Auto-Interp

Top Features by Cosine Similarity

Comparing With GEMMA-2-2B @ 13-gemmascope-res-65k

Configuration

google/gemma-scope-2b-pt-res/layer_13/width_65k/average_l0_75

Prompts (Dashboard)

24,576 prompts, 128 tokens each

Dataset (Dashboard)

monology/pile-uncopyrighted

Features

65,536

Data Type

float32

Hook Name

blocks.13.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

 greateſt

-1.04

 myſelf

-1.03

MLLoader

-0.97

 ſche

-0.96

دانشنامهٔ

-0.95

 оригіналу

-0.93

ſelf

-0.92

 itſelf

-0.91

 Shakspeare

-0.91

CloseOperation

-0.90

POSITIVE LOGITS

0.66

0.52

«

0.47

He

0.46

0.44

Ben

0.44

Activations Density 0.119%

references to numerical data and statistics

No Comments

No Known Activations