INDEX

Explanations

references to processes and procedural language 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 @ 3-gemmascope-res-16k

Configuration

google/gemma-scope-2b-pt-res/layer_3/width_16k/average_l0_59

Prompts (Dashboard)

36,864 prompts, 128 tokens each

Dataset (Dashboard)

monology/pile-uncopyrighted

Features

16,384

Data Type

float32

Hook Name

blocks.3.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

 commerciali

-0.46

 pitä

-0.45

 löytyy

-0.44

 curfew

-0.44

 kiin

-0.43

 kunjungan

-0.43

 minat

-0.42

 közül

-0.42

unton

-0.41

 اختصاص

-0.40

POSITIVE LOGITS

 process

2.13

 processes

2.01

 Process

1.90

processes

1.83

 Processes

1.83

 PROCESS

1.80

process

1.74

Process

1.71

PROCESS

1.67

Проце

1.60

Activations Density 0.266%

references to processes and procedural language in various contexts

No Comments

No Known Activations