INDEX

Explanations

programming-related expressions and constructs

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

New Auto-Interp

Top Features by Cosine Similarity

Comparing With GEMMA-2-2B @ 12-gemmascope-res-65k

Configuration

google/gemma-scope-2b-pt-res/layer_12/width_65k/average_l0_72

Prompts (Dashboard)

24,576 prompts, 128 tokens each

Dataset (Dashboard)

monology/pile-uncopyrighted

Features

65,536

Data Type

float32

Hook Name

blocks.12.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

 estimés

-0.63

Dun

-0.47

 '\\;'

-0.44

 lievito

-0.44

increments

-0.44

RefNanny

-0.43

TypedDataSet

-0.43

онов

-0.42

httphttps

-0.41

 growth

-0.41

POSITIVE LOGITS

ercicio

0.62

省市镇

0.59

raborty

0.58

 propOrder

0.57

ſelves

0.57

ніципалі

0.56

 quirúrg

0.55

 clim

0.55

ihnachten

0.55

 vocale

0.54

Activations Density 0.069%

programming-related expressions and constructs

No Comments

No Known Activations