INDEX

Explanations

terms related to design and designers

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

New Auto-Interp

Top Features by Cosine Similarity

Comparing With GEMMA-2-2B @ 0-gemmascope-res-16k

Configuration

google/gemma-scope-2b-pt-res/layer_0/width_16k/average_l0_105

Prompts (Dashboard)

36,864 prompts, 128 tokens each

Dataset (Dashboard)

monology/pile-uncopyrighted

Features

16,384

Data Type

float32

Hook Name

blocks.0.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

 referenties

-0.61

 lassen

-0.53

 espèce

-0.53

findOne

-0.53

 komik

-0.52

 amendment

-0.51

optString

-0.51

conse

-0.50

 odds

-0.49

readLine

-0.49

POSITIVE LOGITS

 designer

0.74

 Tower

0.73

designer

0.72

ftagPool

0.71

 Wikimedijinoj

0.67

Designer

0.66

 HasFactory

0.66

 препратки

0.65

 TOWER

0.64

DeleteBehavior

0.63

Activations Density 0.184%

terms related to design and designers

No Comments

No Known Activations