INDEX

Explanations

instances of the word "back" in various contexts, often indicating a return or reversal

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

BSA

-0.64

QRST

-0.64

 defaultstate

-0.63

velocityX

-0.60

SEB

-0.59

aarrggbb

-0.58

BSA

-0.58

 wireType

-0.58

 IBOutlet

-0.56

UVWXYZ

-0.56

POSITIVE LOGITS

Encyklopedia

0.56

 into

0.53

 betweenstory

0.51

 again

0.50

tikzpicture

0.49

 виправивши

0.48

لينكات

0.48

 विश्वसनीयता

0.46

ruptedException

0.45

Autoritní

0.45

Activations Density 0.047%

instances of the word "back" in various contexts, often indicating a return or reversal

No Comments

No Known Activations