INDEX

Explanations

references to legal cases and court jurisdictions

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

New Auto-Interp

Top Features by Cosine Similarity

Comparing With GEMMA-2-2B @ 12-gemmascope-res-65k

Configuration

google/gemma-scope-2b-pt-res/layer_12/width_65k/average_l0_72

Prompts (Dashboard)

24,576 prompts, 128 tokens each

Dataset (Dashboard)

monology/pile-uncopyrighted

Features

65,536

Data Type

float32

Hook Name

blocks.12.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

tak

-0.58

-0.55

 rest

-0.53

app

-0.52

 little

-0.52

</em>

-0.52

пас

-0.52

 walk

-0.50

 comfort

-0.50

躇

-0.50

POSITIVE LOGITS

 nahilalakip

0.79

=$?

0.77

aarrggbb

0.74

Efq

0.73

 InputDecoration

0.73

 houſe

0.73

الدراسه

0.72

 Serap

0.72

SharedCtor

0.71

 ſon

0.71

Activations Density 0.096%

references to legal cases and court jurisdictions

No Comments

No Known Activations