INDEX

Explanations

suffixes -ization, -ation

np_max-act · gemini-2.0-flash

New Auto-Interp

Configuration

mwhanna/qwen3-4b-transcoders/layer_1.safetensors

Prompts (Dashboard)

16,384 prompts, 128 tokens each

Dataset (Dashboard)

monology/pile-uncopyrighted

Features

163,840

Data Type

float32

Hook Name

blocks.1.mlp.hook_in

Architecture

transcoder

Context Size

8,192

Dataset

monology/pile-uncopyrighted

Embeds

PlotsExplanationShow Test FieldDefault Test Text

IFrame

Link

Not in Any Lists

No Comments

Negative Logits

 tender

-0.28

(sprintf

-0.28

 caught

-0.27

olerance

-0.26

è¿ĩåħ³

-0.26

peg

-0.25

AE

-0.24

(ui

-0.24

 questions

-0.24

æĮĩåįĹ

-0.23

POSITIVE LOGITS

unkt

0.27

ä¸įæĪĲ

0.27

 Sachs

0.27

lickr

0.25

å½ĵæĹ¶çļĦ

0.25

VERAGE

0.25

riere

0.25

liest

0.25

IRST

0.24

ÑĢÐµÐ¼

0.24

Activations Density 21.686%

suffixes -ization, -ation

No Comments

No Known Activations