INDEX

Explanations

comparative expressions and language that indicates comparisons between different subjects or items

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

Jeografia

-0.63

StateException

-0.57

SequentialGroup

-0.57

odeon

-0.57

 Wikispecies

-0.56

ikation

-0.56

llium

-0.54

ส์

-0.53

 compound

-0.52

colgroup

-0.52

POSITIVE LOGITS

 favorably

0.90

 favourably

0.83

 apples

0.78

TagHelper

0.73

 Comparisons

0.71

 disponibilités

0.68

 liken

0.67

 against

0.64

 sánh

0.64

 compares

0.63

Activations Density 0.106%

comparative expressions and language that indicates comparisons between different subjects or items

No Comments

No Known Activations