INDEX

Explanations

references to specific brands or types of products

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

"Amb" followed by other letters

np_acts-logits-general · gemini-2.0-flash

Amb followed by specific endings

np_acts-logits-general · gemini-2.5-flash-lite

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

expandindo

-0.85

abestanden

-0.71

 Wikimedijinoj

-0.70

gameserver

-0.68

DebuggerNonUser

-0.68

="@+

-0.67

lacian

-0.66

 initComponents

-0.66

Билгалдахарш

-0.63

 ModelExpression

-0.62

POSITIVE LOGITS

Amb

0.86

Amb

0.79

amb

0.75

emb

0.62

 Ambro

0.60

amb

0.60

mb

0.59

Emb

0.59

EMB

0.57

 Ambrose

0.57

Activations Density 0.084%

references to specific brands or types of products

"Amb" followed by other letters

Amb followed by specific endings

No Comments

No Known Activations