INDEX

Explanations

phrases and terms related to honor or being recognized

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

New Auto-Interp

Top Features by Cosine Similarity

Comparing With GEMMA-2-2B @ 8-gemmascope-res-16k

Configuration

google/gemma-scope-2b-pt-res/layer_8/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.8.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

 scolaire

-0.56

ulario

-0.53

BeginInit

-0.51

ArgsConstructor

-0.51

 bear

-0.49

[++

-0.48

[];

-0.48

XmlRootElement

-0.48

 virtuel

-0.47

 parseFrom

-0.47

POSITIVE LOGITS

 choix

0.65

SelectedValue

0.63

 selection

0.63

 choice

0.59

 Selection

0.59

pick

0.56

 انتخاب

0.55

Selection

0.55

choice

0.54

 pick

0.54

Activations Density 0.307%

phrases and terms related to honor or being recognized

No Comments

No Known Activations