INDEX

Explanations

words and phrases indicating relationships or associations between different concepts or entities

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

followed by "with" or "by"

np_acts-logits-general · gemini-2.0-flash

accompanied by or with

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

PlotsExplanationShow Test FieldDefault Test Text

IFrame

Link

Not in Any Lists

No Comments

Negative Logits

 mesmas

-0.67

 mesma

-0.55

⋮

-0.54

 mismas

-0.52

İstinadlar

-0.52

orang

-0.51

 priorité

-0.47

atoire

-0.46

 handelt

-0.45

 Empieza

-0.45

POSITIVE LOGITS

Extinguishing

0.82

 كومونز

0.75

 accompanied

0.72

 accompanying

0.71

 함께

0.70

adpleegd

0.68

<>());

0.67

 accompag

0.65

accompanied

0.65

 accompanies

0.65

Activations Density 0.305%

words and phrases indicating relationships or associations between different concepts or entities

followed by "with" or "by"

accompanied by or with

No Comments

No Known Activations