INDEX

Explanations

terms and concepts related to dimensions and measures

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

PlotsExplanationShow Test FieldDefault Test Text

IFrame

Link

Not in Any Lists

No Comments

Negative Logits

<h1>

-0.39

/*

-0.38

<strong>

-0.35

 OMITBAD

-0.34

AGM

-0.33

audiovisuel

-0.32

<h3>

-0.32

<h2>

-0.32

 invokingState

-0.31

POSITIVE LOGITS

 whoſe

0.74

 disambiguazione

0.67

 ब्रेकडाउन

0.65

 reaſon

0.65

 كومونز

0.63

 anſ

0.63

)++;

0.62

 Monfieur

0.61

 myſelf

0.61

يميديا

0.60

Activations Density 0.054%

terms and concepts related to dimensions and measures

No Comments

No Known Activations