INDEX

Explanations

phrases indicating belief or speculation

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

New Auto-Interp

Top Features by Cosine Similarity

Comparing With GEMMA-2-2B @ 4-gemmascope-res-16k

Configuration

google/gemma-scope-2b-pt-res/layer_4/width_16k/average_l0_124

Prompts (Dashboard)

36,864 prompts, 128 tokens each

Dataset (Dashboard)

monology/pile-uncopyrighted

Features

16,384

Data Type

float32

Hook Name

blocks.4.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

Последние

-0.78

 препратки

-0.72

بوابة

-0.69

存于互联网档案馆

-0.68

="#"><

-0.67

 समीक्षाओं

-0.67

 RISERVATA

-0.66

labelControl

-0.65

isome

-0.64

RegressionTest

-0.63

POSITIVE LOGITS

 might

0.67

might

0.66

 Might

0.63

 have

0.62

FormItem

0.60

Might

0.60

 Maynard

0.59

 должна

0.57

leih

0.56

wikk

0.56

Activations Density 0.383%

phrases indicating belief or speculation

No Comments

No Known Activations