INDEX

Explanations

instances of different types of initiation markers or tags in the text

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

New Auto-Interp

Top Features by Cosine Similarity

Comparing With GEMMA-2-2B @ 13-gemmascope-res-65k

Configuration

google/gemma-scope-2b-pt-res/layer_13/width_65k/average_l0_75

Prompts (Dashboard)

24,576 prompts, 128 tokens each

Dataset (Dashboard)

monology/pile-uncopyrighted

Features

65,536

Data Type

float32

Hook Name

blocks.13.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

 CreateTagHelper

-0.83

ScopeManager

-0.67

sieren

-0.56

 createState

-0.46

 argparse

-0.46

ArgumentParser

-0.45

igem

-0.44

("")]

-0.43

ishments

-0.42

addChild

-0.42

POSITIVE LOGITS

SequentialGroup

0.75

 Celui

0.64

 aéreo

0.60

enseur

0.59

 Ceux

0.57

 relâche

0.57

क्रम

0.57

 sekali

0.57

 Bancroft

0.57

 judiciaire

0.56

Activations Density 0.035%

instances of different types of initiation markers or tags in the text

No Comments

No Known Activations

instances of different types of initiation markers or tags in the text

No Comments

No Known Activations