INDEX

Explanations

names of books and media productions

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

New Auto-Interp

Top Features by Cosine Similarity

Comparing With GEMMA-2-2B @ 4-gemmascope-res-16k

Configuration

google/gemma-scope-2b-pt-res/layer_4/width_16k/average_l0_124

Prompts (Dashboard)

36,864 prompts, 128 tokens each

Dataset (Dashboard)

monology/pile-uncopyrighted

Features

16,384

Data Type

float32

Hook Name

blocks.4.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

 Either

-0.49

 either

-0.47

Either

-0.45

 whatever

-0.44

classmethod

-0.42

 Whatever

-0.42

Whatever

-0.42

 Luckily

-0.41

 Tempest

-0.41

Storia

-0.40

POSITIVE LOGITS

 autorytatywna

0.91

出版年

0.80

SourceChecksum

0.80

 Roskov

0.73

IntoConstraints

0.72

Personendaten

0.71

 AssemblyCulture

0.70

 mergeFrom

0.69

StructEnd

0.68

SequentialGroup

0.68

Activations Density 2.236%

names of books and media productions

No Comments

No Known Activations