INDEX

Explanations

references to expertise and authority in a professional context

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

AndEndTag

-0.71

 kasarigan

-0.67

IVEREF

-0.65

########.

-0.63

kí

-0.58

 Roskov

-0.56

applicant

-0.56

 borrower

-0.54

طلاق

-0.53

 poveznice

-0.53

POSITIVE LOGITS

 expert

2.59

 experts

2.43

expert

2.18

experts

2.00

 Expert

1.98

 specialists

1.98

Expert

1.96

 experto

1.94

 expertise

1.94

 specialist

1.93

Activations Density 0.340%

references to expertise and authority in a professional context

No Comments

No Known Activations