INDEX

Explanations

phrases indicating contrast or conditions in discourse

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

New Auto-Interp

Top Features by Cosine Similarity

Comparing With GEMMA-2-2B @ 11-gemmascope-res-16k

Configuration

google/gemma-scope-2b-pt-res/layer_11/width_16k/average_l0_80

Prompts (Dashboard)

36,864 prompts, 128 tokens each

Dataset (Dashboard)

monology/pile-uncopyrighted

Features

16,384

Data Type

float32

Hook Name

blocks.11.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

 Saltar

-0.59

 Jurí

-0.57

 fubject

-0.54

byz

-0.53

 PicClick

-0.53

ódó

-0.52

 Tomé

-0.52

Ivo

-0.52

 unprepared

-0.51

 ftate

-0.51

POSITIVE LOGITS

DataPropertyName

0.70

httphttps

0.65

Билгалдахарш

0.62

onViewCreated

0.61

 Paglinawan

0.61

fjspx

0.60

aarrggbb

0.60

 unknownFields

0.59

 зато

0.58

Hentet

0.57

Activations Density 0.243%

phrases indicating contrast or conditions in discourse

No Comments

No Known Activations