INDEX

Explanations

terms referring to active ingredients and their properties in medicinal or chemical contexts

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

ftagPool

-0.56

WaitGroup

-0.53

errHandler

-0.47

SuppressMessage

-0.46

 fieltro

-0.45

 wireType

-0.44

 kurtka

-0.43

complexContent

-0.42

 bluza

-0.42

ReusableCell

-0.42

POSITIVE LOGITS

 ingredient

0.54

substance

0.53

ingredient

0.52

Personendaten

0.52

 ingredi

0.52

isContained

0.49

 ingredients

0.48

 substance

0.47

ActionCreators

0.47

 apiKey

0.46

Activations Density 0.039%

terms referring to active ingredients and their properties in medicinal or chemical contexts

No Comments

No Known Activations