INDEX

Explanations

period

np_max-act · gemini-2.0-flash

technical or academic terminology, particularly in specialized domains like technology, medicine, and science. The neuron seems particularly tuned to identifying jargon, complex concepts, and domain-specific language that appears in formal or technical writing.

oai_token-act-pair · deepseek-v3 Triggered by @finnwit

New Auto-Interp

Configuration

mwhanna/qwen3-4b-transcoders/layer_14.safetensors

Prompts (Dashboard)

16,384 prompts, 128 tokens each

Dataset (Dashboard)

monology/pile-uncopyrighted

Features

163,840

Data Type

float32

Hook Name

blocks.14.mlp.hook_in

Architecture

transcoder

Context Size

8,192

Dataset

monology/pile-uncopyrighted

Embeds

IFrame

Link

Not in Any Lists

No Comments

Negative Logits

ÃŃsticas

-0.26

enha

-0.25

ëŀľ

-0.25

çĽ¸å¯¹

-0.24

åı¸æľº

-0.24

çĪ¸

-0.24

ffect

-0.23

 cannon

-0.23

è¿ĩçļĦ

-0.23

issing

-0.22

POSITIVE LOGITS

ivre

0.27

åŀĽ

0.26

æĤ¨çļĦåŃ©åŃĲ

0.24

 Chart

0.24

å½Ģ

0.23

éŀ£

0.23

 chart

0.23

 Schema

0.23

,**

0.23

 schematic

0.22

Activations Density 2.587%

period

technical or academic terminology, particularly in specialized domains like technology, medicine, and science. The neuron seems particularly tuned to identifying jargon, complex concepts, and domain-specific language that appears in formal or technical writing.

No Comments

No Known Activations

period

technical or academic terminology, particularly in specialized domains like technology, medicine, and science. The neuron seems particularly tuned to identifying jargon, complex concepts, and domain-specific language that appears in formal or technical writing.

No Comments

No Known Activations