INDEX

Explanations

expressions of the word "kind" or related discussions about types or categories

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

Autoritní

-0.63

ViewFeatures

-0.59

UVWXYZ

-0.58

 TestBed

-0.58

 yaptığı

-0.57

freiheit

-0.52

 romanos

-0.51

Datuak

-0.51

rotum

-0.50

 kese

-0.49

POSITIVE LOGITS

 sorta

0.75

KIND

0.66

hearted

0.66

 Kind

0.65

 hearted

0.65

 KIND

0.63

 sort

0.61

 SORT

0.54

 kind

0.53

Kind

0.53

Activations Density 0.061%

expressions of the word "kind" or related discussions about types or categories

No Comments

No Known Activations