INDEX

Explanations

expressions of recommendation or personal opinion regarding books

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

New Auto-Interp

Top Features by Cosine Similarity

Comparing With GEMMA-2-2B @ 4-gemmascope-mlp-65k

Configuration

google/gemma-scope-2b-pt-mlp/layer_4/width_65k/average_l0_66

Prompts (Dashboard)

24,576 prompts, 128 tokens each

Dataset (Dashboard)

monology/pile-uncopyrighted

Features

65,536

Data Type

float32

Hook Name

blocks.4.hook_mlp_out

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

TintMode

-0.76

istoitu

-0.67

Демографія

-0.66

 архивлан

-0.66

 Infórmanos

-0.65

 متعلقه

-0.64

LEncoder

-0.61

 województwie

-0.61

Datuak

-0.60

GeoNames

-0.60

POSITIVE LOGITS

rator

0.49

dô

0.48

keit

0.47

 utro

0.46

כות

0.45

ăng

0.45

öffentlich

0.45

 дело

0.45

 Wurzel

0.45

 temporaire

0.45

Activations Density 0.015%

expressions of recommendation or personal opinion regarding books

No Comments

No Known Activations