INDEX

Explanations

specific nouns that denote quantities or divisions in various 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

 milliers

-0.86

everywhere

-0.84

 everywhere

-0.82

 muchos

-0.78

 centaines

-0.77

 muchas

-0.77

 многих

-0.76

thousands

-0.74

 duiz

-0.72

 muitas

-0.72

POSITIVE LOGITS

分别是

1.31

分別

1.15

分别

1.11

 respectively

1.08

 namely

1.04

 each

1.03

 apiece

0.99

respectively

0.91

一个是

0.89

 jeweils

0.87

Activations Density 0.730%

specific nouns that denote quantities or divisions in various contexts

No Comments

No Known Activations