INDEX

Explanations

references to the second person and forms of the verb "to be."

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-16k

Configuration

google/gemma-scope-2b-pt-res/layer_12/width_16k/average_l0_82

Prompts (Dashboard)

36,864 prompts, 128 tokens each

Dataset (Dashboard)

monology/pile-uncopyrighted

Features

16,384

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

IFrame

Link

Not in Any Lists

No Comments

Negative Logits

Synonymes

-0.55

 alcanzó

-0.53

ursa

-0.52

mathit

-0.51

wykle

-0.50

 preced

-0.48

あった

-0.48

 démocr

-0.48

 titolo

-0.47

 SEDS

-0.47

POSITIVE LOGITS

 looking

0.83

 getting

0.82

ViewInit

0.79

 doing

0.77

 working

0.76

 trying

0.75

حوالہ

0.74

 playing

0.72

 going

0.72

stateProvider

0.71

Activations Density 0.333%

references to the second person and forms of the verb "to be."

No Comments

No Known Activations