INDEX

Explanations

mentions of various activities

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

New Auto-Interp

Top Features by Cosine Similarity

Comparing With GEMMA-2-9B-IT @ 20-gemmascope-res-16k

Configuration

google/gemma-scope-9b-it-res/layer_20/width_16k/average_l0_91

Prompts (Dashboard)

24,576 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

<bos>

-0.48

 automático

-0.48

setSearch

-0.46

Trusted

-0.46

 Cientí

-0.45

Dead

-0.44

 magnético

-0.44

følgelig

-0.44

 clás

-0.43

Océ

-0.42

POSITIVE LOGITS

 Facility

0.76

Facility

0.76

 Locality

0.75

 BaseActivity

0.73

bidity

0.69

 facility

0.69

ActivityCompat

0.68

 bari

0.67

facility

0.66

 eventType

0.66

Activations Density 0.082%

mentions of various activities

No Comments

No Known Activations