INDEX

Explanations

references to different types and contexts of "ball" along with associated activities

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

New Auto-Interp

Top Features by Cosine Similarity

Comparing With GEMMA-2-2B @ 5-gemmascope-res-16k

Configuration

google/gemma-scope-2b-pt-res/layer_5/width_16k/average_l0_68

Prompts (Dashboard)

36,864 prompts, 128 tokens each

Dataset (Dashboard)

monology/pile-uncopyrighted

Features

16,384

Data Type

float32

Hook Name

blocks.5.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

 Prieto

-0.71

helves

-0.71

 McCabe

-0.70

]';

-0.67

%")

-0.67

setPreferred

-0.64

%";

-0.64

><!--

-0.63

]";

-0.62

-0.61

POSITIVE LOGITS

 ball

2.76

 balls

2.61

 BALL

2.54

 Ball

2.53

ball

2.45

Ball

2.43

 Balls

2.39

Balls

2.22

balls

2.21

BALL

2.20

Activations Density 0.035%

references to different types and contexts of "ball" along with associated activities

No Comments

No Known Activations