INDEX

Explanations

programming-related terms or code execution processes

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

New Auto-Interp

Top Features by Cosine Similarity

Comparing With GEMMA-2-2B @ 5-gemmascope-res-16k

Configuration

google/gemma-scope-2b-pt-res/layer_5/width_16k/average_l0_68

Prompts (Dashboard)

36,864 prompts, 128 tokens each

Dataset (Dashboard)

monology/pile-uncopyrighted

Features

16,384

Data Type

float32

Hook Name

blocks.5.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

AntiForgeryToken

-0.71

 দেখ

-0.61

 Bring

-0.59

плек

-0.58

 bring

-0.58

Bring

-0.57

 knew

-0.56

saw

-0.56

 offers

-0.54

 доби

-0.53

POSITIVE LOGITS

ίνεται

0.89

 AssemblyVersion

0.68

ється

0.66

 Wird

0.66

ются

0.64

-------------</

0.64

 jScrollPane

0.63

醐

0.63

Wird

0.62

 ProtoMessage

0.61

Activations Density 0.071%

programming-related terms or code execution processes

No Comments

No Known Activations