INDEX

Explanations

phrases indicating the purpose or function of actions

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

 Administrativna

-0.94

ExtendWith

-0.74

NOPQRST

-0.72

SharedDtor

-0.69

OGND

-0.65

findpost

-0.65

bootstrapcdn

-0.65

 estimés

-0.64

WebVitals

-0.64

 kasarigan

-0.61

POSITIVE LOGITS

 việc

0.75

 sự

0.71

the

0.68

การ

0.64

 την

0.58

sự

0.51

 τη

0.50

信息的

0.49

isissez

0.48

studies

0.47

Activations Density 0.815%

phrases indicating the purpose or function of actions

No Comments

No Known Activations