INDEX

Explanations

phrases that indicate product comparisons and quality assessments

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

CloseOperation

-0.58

 TextInputType

-0.58

PyExc

-0.57

atience

-0.57

]='\

-0.56

PhysRevD

-0.55

 nonatomic

-0.55

oredCriteria

-0.55

paksa

-0.54

ChildIndex

-0.54

POSITIVE LOGITS

 excellent

1.89

 good

1.81

excellent

1.70

good

1.57

 superb

1.52

Excellent

1.48

 Excellent

1.47

 EXCELL

1.43

 fantastic

1.43

 excellente

1.40

Activations Density 0.418%

phrases that indicate product comparisons and quality assessments

No Comments

No Known Activations