INDEX

Explanations

code structures, such as class and function definitions along with their parameters

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

New Auto-Interp

Top Features by Cosine Similarity

Comparing With GEMMA-2-2B @ 20-gemmascope-res-16k

Configuration

google/gemma-scope-2b-pt-res/layer_20/width_16k/average_l0_71

Prompts (Dashboard)

36,864 prompts, 128 tokens each

Dataset (Dashboard)

monology/pile-uncopyrighted

Features

16,384

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

 bienfaits

-0.61

 âmes

-0.60

 nuages

-0.60

 vœux

-0.60

 bénéfices

-0.59

 renseignements

-0.59

 pouvoirs

-0.58

 rigue

-0.58

 touristes

-0.57

 négociations

-0.56

POSITIVE LOGITS

 list

1.06

 listOf

1.06

 lista

0.98

listOf

0.95

Seznam

0.94

 các

0.93

ListOf

0.85

 Liste

0.84

 những

0.83

NameList

0.83

Activations Density 0.467%

code structures, such as class and function definitions along with their parameters

No Comments

No Known Activations