INDEX

Explanations

quantitative references to large numbers

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

New Auto-Interp

Configuration

google/gemma-scope-9b-pt-res/layer_20/width_131k/average_l0_114

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

PlotsExplanationShow Test FieldDefault Test Text

IFrame

Link

Not in Any Lists

No Comments

Negative Logits

 Holt

-0.42

 tepi

-0.40

 modalités

-0.40

 attentes

-0.38

 critères

-0.38

 oscuro

-0.38

by

-0.38

ы

-0.36

bar

-0.36

POSITIVE LOGITS

 million

1.13

Million

1.11

 Million

1.10

VersionUID

1.05

 MILLION

1.04

 millions

1.02

 Millions

0.97

millions

0.96

 millón

0.96

 Millionen

0.94

Activations Density 0.020%

quantitative references to large numbers

No Comments

No Known Activations