INDEX

Explanations

positive experiences and descriptions in various contexts

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

New Auto-Interp

Top Features by Cosine Similarity

Comparing With GEMMA-2-9B @ 38-gemmascope-res-16k

Configuration

google/gemma-scope-9b-pt-res/layer_38/width_16k/average_l0_128

Prompts (Dashboard)

24,576 prompts, 128 tokens each

Dataset (Dashboard)

monology/pile-uncopyrighted

Features

16,384

Data Type

float32

Hook Name

blocks.38.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

 يتيمه

-0.76

astéroïdes

-0.70

LabelTagHelper

-0.69

queryInterface

-0.68

 informée

-0.68

setVerticalGroup

-0.66

 betweenstory

-0.65

AddTagHelper

-0.65

 nahilalakip

-0.63

ViewImports

-0.60

POSITIVE LOGITS

 behaviour

0.55

 Reasonable

0.54

 pelayanan

0.54

 nice

0.53

 reasonable

0.52

Nice

0.52

 👌

0.52

 good

0.51

👌

0.51

 Behaviour

0.50

Activations Density 0.046%

positive experiences and descriptions in various contexts

No Comments

No Known Activations