INDEX

Explanations

expressions signaling a change in context or emphasis, particularly at the beginning of a sentence or paragraph

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

 חיצוני

-0.78

יצוני

-0.78

 אחרים

-0.76

 houſe

-0.69

 برانيه

-0.69

 recev

-0.68

 மற்ற

-0.68

 vierge

-0.67

 culoare

-0.66

 چقدر

-0.66

POSITIVE LOGITS

 iprot

0.65

 cherchés

0.63

NIS

0.60

 typelib

0.60

FunctionFlags

0.58

באנגלית

0.57

ThroughAttribute

0.56

 IndexPath

0.55

HMO

0.54

SDM

0.54

Activations Density 0.001%

expressions signaling a change in context or emphasis, particularly at the beginning of a sentence or paragraph

No Comments

No Known Activations