INDEX

Explanations

programming question context

np_max-act-logits · gemini-2.0-flash

New Auto-Interp

Configuration

mwhanna/qwen3-4b-transcoders/layer_31.safetensors

Prompts (Dashboard)

16,384 prompts, 128 tokens each

Dataset (Dashboard)

monology/pile-uncopyrighted

Features

163,840

Data Type

float32

Hook Name

blocks.31.mlp.hook_in

Architecture

transcoder

Context Size

8,192

Dataset

monology/pile-uncopyrighted

Embeds

IFrame

Link

Negative Logits

 sina

-0.27

uet

-0.27

 stand

-0.26

ÑĦÑĢ

-0.26

 steht

-0.26

gnu

-0.26

eel

-0.25

 stehen

-0.25

PB

-0.24

 deductions

-0.24

POSITIVE LOGITS

 looping

0.28

mod

0.27

çĽ¸

0.27

æĿŀ

0.27

lo

0.26

.Reporting

0.26

gh

0.26

flow

0.25

 loop

0.25

 TKey

0.25

Activations Density 0.897%

programming question context

No Known Activations