INDEX

Explanations

references to hills

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

New Auto-Interp

Top Features by Cosine Similarity

Comparing With GEMMA-2-2B @ 3-gemmascope-res-16k

Configuration

google/gemma-scope-2b-pt-res/layer_3/width_16k/average_l0_59

Prompts (Dashboard)

36,864 prompts, 128 tokens each

Dataset (Dashboard)

monology/pile-uncopyrighted

Features

16,384

Data Type

float32

Hook Name

blocks.3.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

 Bourgoin

-0.67

 Qantas

-0.67

 Evansville

-0.67

Botswana

-0.64

quard

-0.64

granate

-0.63

 Botswana

-0.62

ambique

-0.62

gnac

-0.61

emias

-0.61

POSITIVE LOGITS

 Hill

3.56

Hill

3.38

 hill

3.37

 HILL

3.16

hill

2.79

 hills

2.69

HILL

2.49

 Hills

2.33

Hills

2.04

hills

2.01

Activations Density 0.053%

references to hills

No Comments

No Known Activations