INDEX

Explanations

references to languages and linguistics

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

New Auto-Interp

Top Features by Cosine Similarity

Comparing With GEMMA-2-2B @ 11-gemmascope-res-16k

Configuration

google/gemma-scope-2b-pt-res/layer_11/width_16k/average_l0_80

Prompts (Dashboard)

36,864 prompts, 128 tokens each

Dataset (Dashboard)

monology/pile-uncopyrighted

Features

16,384

Data Type

float32

Hook Name

blocks.11.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

 Mitar

-0.91

 himſelf

-0.90

 Supremo

-0.85

 eccell

-0.82

 themſelves

-0.82

 miſ

-0.80

 fubject

-0.80

 blest

-0.80

 Italij

-0.78

 poffible

-0.78

POSITIVE LOGITS

 languages

1.53

 language

1.44

 Language

1.37

Languages

1.29

Language

1.27

 Languages

1.25

languages

1.23

language

1.15

UAGES

1.13

 Lang

1.13

Activations Density 0.058%

references to languages and linguistics

No Comments

No Known Activations