INDEX

Explanations

phrases that indicate changes or events occurring in relation to other subjects

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

 TestBed

-0.76

 Roskov

-0.67

 '\\;'

-0.67

addPreferredGap

-0.64

 antaranya

-0.61

adpleegd

-0.61

DebuggerStep

-0.61

 dermed

-0.59

ChromeDriver

-0.58

تفصیلات

-0.57

POSITIVE LOGITS

ModelSerializer

0.62

 matters

0.59

 matter

0.58

 designing

0.57

 figuring

0.56

 AssemblyTitle

0.53

 mattered

0.52

 cutting

0.51

 capturing

0.49

ViewFeatures

0.49

Activations Density 0.052%

phrases that indicate changes or events occurring in relation to other subjects

No Comments

No Known Activations