INDEX

Explanations

terms related to numbers and measurements indicating size

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

New Auto-Interp

Top Features by Cosine Similarity

Comparing With GEMMA-2-2B @ 21-gemmascope-res-16k

Configuration

google/gemma-scope-2b-pt-res/layer_21/width_16k/average_l0_70

Prompts (Dashboard)

36,864 prompts, 128 tokens each

Dataset (Dashboard)

monology/pile-uncopyrighted

Features

16,384

Data Type

float32

Hook Name

blocks.21.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

estacks

-0.70

<eos>

-0.61

 tartalomajánló

-0.60

expandindo

-0.58

primaryStage

-0.56

ItemBackground

-0.56

 mergeFrom

-0.53

例句

-0.52

++]=

-0.51

uitton

-0.51

POSITIVE LOGITS

</h1>

0.81

 الحره

0.81

</h4>

0.75

IntoConstraints

0.75

 wireType

0.71

 Saltar

0.70

httphttps

0.68

XmlAccessType

0.67

ioutil

0.65

 autorytatywna

0.64

Activations Density 2.310%

terms related to numbers and measurements indicating size

No Comments

No Known Activations