INDEX

Explanations

patterns related to string manipulation and regular expressions

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

 Schofield

-0.53

 defaultstate

-0.52

EClass

-0.50

 Administrativna

-0.47

 Trauer

-0.47

quiao

-0.47

tabular

-0.46

ScopeManager

-0.46

 estekak

-0.45

TargetException

-0.45

POSITIVE LOGITS

 chars

0.87

 string

0.81

strlen

0.80

 characters

0.79

Character

0.79

 StringBuffer

0.78

 substring

0.76

 strlen

0.76

 Character

0.76

 strings

0.76

Activations Density 0.198%

patterns related to string manipulation and regular expressions

No Comments

No Known Activations