INDEX

Explanations

Emphasizing adverbs

np_max-act · gemini-2.0-flash

New Auto-Interp

Configuration

mwhanna/qwen3-4b-transcoders/layer_20.safetensors

Prompts (Dashboard)

16,384 prompts, 128 tokens each

Dataset (Dashboard)

monology/pile-uncopyrighted

Features

163,840

Data Type

float32

Hook Name

blocks.20.mlp.hook_in

Architecture

transcoder

Context Size

8,192

Dataset

monology/pile-uncopyrighted

Embeds

IFrame

Link

Not in Any Lists

No Comments

Negative Logits

 Bloss

-0.28

onium

-0.27

boro

-0.26

-away

-0.26

guid

-0.25

iplinary

-0.25

unity

-0.25

ini

-0.24

 miracle

-0.24

 Formatting

-0.23

POSITIVE LOGITS

 pant

0.31

å·¨

0.27

.linkedin

0.26

redd

0.25

éĵ¾

0.24

 credited

0.24

Sal

0.24

èİĺ

0.24

fib

0.24

sal

0.24

Activations Density 16.872%

Emphasizing adverbs

No Comments

No Known Activations