INDEX

Explanations

mentions and variations of the word "banana."

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-131k

Configuration

google/gemma-scope-9b-it-res/layer_20/width_131k/average_l0_81

Prompts (Dashboard)

24,576 prompts, 128 tokens each

Dataset (Dashboard)

monology/pile-uncopyrighted

Features

131,072

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

IFrame

Link

Not in Any Lists

No Comments

Negative Logits

fjspx

-0.53

 giorgio

-0.48

 Савезне

-0.47

}{*}{

-0.45

ertale

-0.45

*~

-0.45

oult

-0.44

/*

-0.44

:]:

-0.43

')";

-0.43

POSITIVE LOGITS

 Banana

1.20

 banana

1.19

Banana

1.14

 Bananas

1.03

Bananas

1.02

banana

1.00

 bananas

0.98

 banane

0.94

 banan

0.84

🍌

0.73

Activations Density 0.002%

mentions and variations of the word "banana."

No Comments

No Known Activations