INDEX

Explanations

expressions related to parentheses and nested structures

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

New Auto-Interp

Top Features by Cosine Similarity

Comparing With GEMMA-2-2B @ 15-gemmascope-res-16k

Configuration

google/gemma-scope-2b-pt-res/layer_15/width_16k/average_l0_78

Prompts (Dashboard)

36,864 prompts, 128 tokens each

Dataset (Dashboard)

monology/pile-uncopyrighted

Features

16,384

Data Type

float32

Hook Name

blocks.15.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

 igång

-0.73

 parteci

-0.72

 auroit

-0.72

 feroit

-0.68

 enfans

-0.66

 condamné

-0.65

 silenzio

-0.65

 avoient

-0.65

 circonst

-0.64

ThroughAttribute

-0.63

POSITIVE LOGITS

 समीक्षक

0.52

外部リンク

0.51

Begriffsklä

0.50

 ostavi

0.49

EDEFAULT

0.49

ôt

0.48

ям

0.48

OGND

0.48

ristas

0.48

fal

0.47

Activations Density 0.414%

expressions related to parentheses and nested structures

No Comments

No Known Activations