INDEX

Explanations

negative sentiments or advice against certain actions

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

Follows a negative command

np_acts-logits-general · gemini-2.0-flash

don't bother or waste

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

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

IFrame

Link

Not in Any Lists

No Comments

Negative Logits

obtenir

-0.54

 correctly

-0.52

 Finds

-0.51

 gewinnen

-0.50

ureusement

-0.49

vég

-0.48

 realisation

-0.48

esgue

-0.47

kapan

-0.47

Acquire

-0.47

POSITIVE LOGITS

 bother

1.20

 waste

1.11

waste

0.98

 bothering

0.98

 wasting

0.97

 bothered

0.89

Waste

0.86

 worry

0.86

 WASTE

0.84

 Waste

0.84

Activations Density 0.441%

negative sentiments or advice against certain actions

Follows a negative command

don't bother or waste

No Comments

No Known Activations