INDEX

Explanations

terms related to product features and quality descriptions

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

New Auto-Interp

Top Features by Cosine Similarity

Comparing With GEMMA-2-9B @ 34-gemmascope-res-16k

Configuration

google/gemma-scope-9b-pt-res/layer_34/width_16k/average_l0_114

Prompts (Dashboard)

24,576 prompts, 128 tokens each

Dataset (Dashboard)

monology/pile-uncopyrighted

Features

16,384

Data Type

float32

Hook Name

blocks.34.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

الدراسه

-0.55

 الرياضيه

-0.50

Personendaten

-0.50

يميديا

-0.47

 Roskov

-0.46

UnusedPrivate

-0.44

 Audiodateien

-0.42

BASEPATH

-0.42

Geplaatst

-0.41

elemField

-0.41

POSITIVE LOGITS

 aesthetically

2.23

 aesthetic

2.19

 aesthetics

2.13

aesthetic

1.96

 Aesthetic

1.94

Aesthetic

1.91

 estética

1.83

 Aesthetics

1.82

 beautiful

1.77

 esthétique

1.77

Activations Density 0.651%

terms related to product features and quality descriptions

No Comments

No Known Activations