INDEX

Explanations

phrases indicating scholarly analysis or critique

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

New Auto-Interp

Top Features by Cosine Similarity

Comparing With GEMMA-2-2B @ 15-gemmascope-res-16k

Configuration

google/gemma-scope-2b-pt-res/layer_15/width_16k/average_l0_78

Prompts (Dashboard)

36,864 prompts, 128 tokens each

Dataset (Dashboard)

monology/pile-uncopyrighted

Features

16,384

Data Type

float32

Hook Name

blocks.15.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

 kaarangay

-0.80

MigrationBuilder

-0.80

 виправивши

-0.73

CppCodeGen

-0.71

DockStyle

-0.70

ReusableCell

-0.68

ыгана

-0.67

richTextPanel

-0.67

findpost

-0.66

postsleuth

-0.65

POSITIVE LOGITS

 theor

0.47

 pens

0.45

órc

0.45

 argue

0.45

 blamed

0.44

tern

0.43

tség

0.43

ัก

0.43

 correctes

0.43

有人

0.42

Activations Density 0.369%

phrases indicating scholarly analysis or critique

No Comments

No Known Activations