INDEX

Explanations

words and terms related to medical treatments and drug classifications

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

 jScrollPane

-0.76

tagHelperRunner

-0.68

 lenker

-0.67

 HttpNotFound

-0.67

$_['

-0.64

 defaultstate

-0.60

 препратки

-0.59

CodeDom

-0.57

 Unclassified

-0.56

Efq

-0.56

POSITIVE LOGITS

DeleteBehavior

0.57

 Honig

0.54

 sporo

0.49

się

0.48

mallows

0.47

)‏

0.46

Literatuur

0.46

ddots

0.45

 Transfermarkt

0.45

กัน

0.44

Activations Density 4.021%

words and terms related to medical treatments and drug classifications

No Comments

No Known Activations