INDEX

Explanations

mentions of user interface elements or controls

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

New Auto-Interp

Top Features by Cosine Similarity

Comparing With GEMMA-2-2B @ 1-gemmascope-mlp-16k

Configuration

google/gemma-scope-2b-pt-mlp/layer_1/width_16k/average_l0_105

Prompts (Dashboard)

24,576 prompts, 128 tokens each

Dataset (Dashboard)

monology/pile-uncopyrighted

Features

16,384

Data Type

float32

Hook Name

blocks.1.hook_mlp_out

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

ui

-2.45

UI

-1.97

ui

-1.82

UI

-1.61

Ui

-1.27

Ui

-1.00

uis

-0.84

UIC

-0.83

uição

-0.82

UIS

-0.82

POSITIVE LOGITS

škas

0.49

hets

0.48

 plis

0.47

æus

0.45

hagen

0.45

нской

0.44

agang

0.44

appé

0.44

VertexAttrib

0.44

 GoogleFonts

0.43

Activations Density 0.124%

mentions of user interface elements or controls

No Comments

No Known Activations