INDEX

Explanations

instances of scholarly assertion and data presentation

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

ngOnInit

-0.72

 termica

-0.69

Obrázky

-0.59

styleUrls

-0.58

 affichés

-0.58

 ब्रेकडाउन

-0.57

 davvero

-0.56

Really

-0.56

ibatis

-0.56

 quedando

-0.56

POSITIVE LOGITS

SourceChecksum

0.77

 investigated

0.73

 examined

0.71

 hypothe

0.66

 investigate

0.59

 Roskov

0.59

 examine

0.59

 investigates

0.58

 assessed

0.57

 evaluated

0.57

Activations Density 0.870%

instances of scholarly assertion and data presentation

No Comments

No Known Activations