INDEX

Explanations

phrases indicating abandonment and dependency

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

Following "to" and resulting states

np_acts-logits-general · gemini-2.0-flash

left to die or be without

np_acts-logits-general · gemini-2.5-flash-lite

New Auto-Interp

Top Features by Cosine Similarity

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

Configuration

google/gemma-scope-2b-pt-res/layer_16/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.16.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

 nakalista

-0.58

ScopeManager

-0.51

раздо

-0.49

"}")

-0.49

expandindo

-0.49

něž

-0.48

ctiva

-0.47

(©

-0.46

ecken

-0.46

gnation

-0.45

POSITIVE LOGITS

 fend

0.80

StructEnd

0.79

自助

0.62

AutoScale

0.60

Portail

0.56

 WebDriverWait

0.55

 désolés

0.55

 harden

0.55

 cherchés

0.55

 tự

0.55

Activations Density 0.256%

phrases indicating abandonment and dependency

Following "to" and resulting states

left to die or be without

No Comments

No Known Activations