INDEX

Explanations

various instances of the word "od" and its forms in context

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

New Auto-Interp

Top Features by Cosine Similarity

Comparing With GEMMA-2-2B @ 23-gemmascope-res-16k

Configuration

google/gemma-scope-2b-pt-res/layer_23/width_16k/average_l0_75

Prompts (Dashboard)

36,864 prompts, 128 tokens each

Dataset (Dashboard)

monology/pile-uncopyrighted

Features

16,384

Data Type

float32

Hook Name

blocks.23.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

 referenties

-0.90

 Audiodateien

-0.82

bootstrapcdn

-0.80

mapStateToProps

-0.78

 pinulongan

-0.77

contentLoaded

-0.73

клопе

-0.72

__(/*!

-0.72

 հղումներ

-0.68

 désolés

-0.68

POSITIVE LOGITS

nature

0.49

UnusedPrivate

0.46

mui

0.42

肩

0.42

uses

0.42

 couverts

0.42

 aspect

0.41

属于

0.39

urlopen

0.39

属

0.38

Activations Density 0.074%

various instances of the word "od" and its forms in context

No Comments

No Known Activations