INDEX

Explanations

names and terms related to legal and governmental contexts

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

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

PlotsExplanationShow Test FieldDefault Test Text

IFrame

Link

Not in Any Lists

No Comments

Negative Logits

kloped

-0.70

 vraiment

-0.66

 BorderRadius

-0.64

 المعيارى

-0.62

 MonoBehaviour

-0.61

 digne

-0.61

Knows

-0.58

apimachinery

-0.58

expandindo

-0.57

Lähteet

-0.57

POSITIVE LOGITS

 previously

1.20

 sebelumnya

1.11

 earlier

1.09

previously

0.99

 Previously

0.99

 Earlier

0.97

earlier

0.94

Previously

0.94

Earlier

0.93

 ранее

0.90

Activations Density 0.337%

names and terms related to legal and governmental contexts

No Comments

No Known Activations