INDEX

Explanations

terms related to negative impacts and consequences

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

Worsening, exacerbating, or damaging effects

np_acts-logits-general · gemini-2.0-flash

worsen, further, exacerbate

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

New Auto-Interp

Top Features by Cosine Similarity

Comparing With GEMMA-2-9B @ 24-gemmascope-res-16k

Configuration

google/gemma-scope-9b-pt-res/layer_24/width_16k/average_l0_114

Prompts (Dashboard)

24,576 prompts, 128 tokens each

Dataset (Dashboard)

monology/pile-uncopyrighted

Features

16,384

Data Type

float32

Hook Name

blocks.24.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

oby

-0.35

isamment

-0.34

 cucchiaio

-0.32

 dedication

-0.32

 quedado

-0.31

 maravilla

-0.30

 quyền

-0.30

entet

-0.29

 safeguards

-0.29

 importance

-0.29

POSITIVE LOGITS

 further

0.86

 worsen

0.84

further

0.83

 exacerbate

0.82

 worsening

0.80

 exacerb

0.79

 aggravate

0.74

 worsened

0.73

 FURTHER

0.72

Further

0.71

Activations Density 0.380%

terms related to negative impacts and consequences

Worsening, exacerbating, or damaging effects

worsen, further, exacerbate

No Comments

No Known Activations