INDEX

Explanations

statements expressing levels of concern or emphasis regarding a topic

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

New Auto-Interp

Top Features by Cosine Similarity

Comparing With GEMMA-2-2B @ 12-gemmascope-res-65k

Configuration

google/gemma-scope-2b-pt-res/layer_12/width_65k/average_l0_72

Prompts (Dashboard)

24,576 prompts, 128 tokens each

Dataset (Dashboard)

monology/pile-uncopyrighted

Features

65,536

Data Type

float32

Hook Name

blocks.12.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

zepam

-0.76

 cerevisiae

-0.69

 يتيمه

-0.66

bootstrapcdn

-0.66

﴿

-0.65

nachron

-0.64

Zuz

-0.63

Raz

-0.63

 Muhamma

-0.62

ptic

-0.62

POSITIVE LOGITS

oweit

0.77

saraba

0.75

 soweit

0.73

 sejauh

0.71

 calendriers

0.68

 hoof

0.67

InSection

0.67

ofar

0.63

 regards

0.62

HideInInspector

0.61

Activations Density 0.016%

statements expressing levels of concern or emphasis regarding a topic

No Comments

No Known Activations