INDEX

Explanations

programming-related terms and their attributes in code contexts

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

New Auto-Interp

Top Features by Cosine Similarity

Comparing With GEMMA-2-2B @ 23-gemmascope-res-16k

Configuration

google/gemma-scope-2b-pt-res/layer_23/width_16k/average_l0_75

Prompts (Dashboard)

36,864 prompts, 128 tokens each

Dataset (Dashboard)

monology/pile-uncopyrighted

Features

16,384

Data Type

float32

Hook Name

blocks.23.hook_resid_post

Hook Layer

Architecture

jumprelu

Context Size

1,024

Dataset

monology/pile-uncopyrighted

Activation Function

relu

Embeds

IFrame

Link

Not in Any Lists

No Comments

Negative Logits

-0.65

<<"

-0.58

>;

-0.56

fjspx

-0.54

=='

-0.53

↵↵

-0.53

Personensuche

-0.53

-0.52

<>();

-0.51

or

-0.50

POSITIVE LOGITS

 myſelf

0.82

ंदीखरीदारी

0.72

 оригіналу

0.71

 fhew

0.69

colhead

0.67

 JpaRepository

0.66

 Monfieur

0.65

 AssemblyTitle

0.64

 itſelf

0.64

principalColumn

0.63

Activations Density 0.165%

programming-related terms and their attributes in code contexts

No Comments

No Known Activations