INDEX

Explanations

occurrences of the word "the."

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

New Auto-Interp

Top Features by Cosine Similarity

Comparing With GEMMA-2-2B @ 1-gemmascope-mlp-16k

Configuration

google/gemma-scope-2b-pt-mlp/layer_1/width_16k/average_l0_105

Prompts (Dashboard)

24,576 prompts, 128 tokens each

Dataset (Dashboard)

monology/pile-uncopyrighted

Features

16,384

Data Type

float32

Hook Name

blocks.1.hook_mlp_out

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

IntoConstraints

-0.66

the

-0.63

at

-0.59

 meny

-0.54

...

-0.52

 through

-0.51

…

-0.49

webElementXpaths

-0.49

igshid

-0.49

POSITIVE LOGITS

 itſelf

0.89

 photolibrary

0.85

 myſelf

0.81

 Theſe

0.78

 AssemblyCulture

0.77

$_"

0.75

 himſelf

0.72

^(@)

0.72

ApplicationTests

0.71

 Forumite

0.69

Activations Density 0.054%

occurrences of the word "the."

No Comments

No Known Activations