The
Echo-DSRN(N)
(
Dual State Recurrent Neural Network
, short name:
Echo-DSRN
, also know as
echo
) is a novel architecture specifically designed to be a viable alternative for low-resource tasks that are currently being inefficiently handled by the excessive scale of Large Language Models (LLMs) 🌱
⚠️ Important Notice
This is a
research prototype
and
demo model
.
Not production-ready
Will hallucinate and give incorrect answers
Do not use for any real-world decisions
Intended for architecture experimentation only
What Works
Text generation is fluent
Memory usage is constant O(1)
Runs on CPUs, NPUs, GPUs (Tested on AMD's ROCm and Apple's MPS)
What Doesn't Work
Factual accuracy
Instruction following
Common sense reasoning
🏗️ Architecture Details
Property
Value
Model Type
echo_dsrn
Layers
8
Hidden Dim
512
Attention Heads
4
MLP Ratio
8.0
Vocab Size
32011
Hybrid Attention
True
RMSNorm
True
📊 Parameter Breakdown
Component
Parameters
% of Total
Total
114.69M (114,687,488)
100%
Embeddings
16.39M
14.29%
DSRN Blocks (Aggregate)
81.91M
71.42%
LM Head
16.39M
14.29%
🧩 Internal Block Structure (Per Layer)
Sub-Component
Parameters
Description
MLP (Feed-Forward)
4.20M
Upscaled hidden layers
DSRN Slow State
3.15M
Constant-time memory gates
GRU Fast State
1.58M
Recurrent fast path
Surprise Gating
264,192
Dynamic focus mechanism
Normalization
1,024
LayerNorm / RMSNorm
Supervised Fine-Tuning (SFTTrainer)
2 epochs on a single
AMD Ryzen AI Max+ 395
(128 GB RAM)
Evaluation
Tasks
Version
Filter
n-shot
Metric
Value
Stderr
arc_easy
1
none
0
acc
↑
0.4289
±
0.0102
none
0
acc_norm
↑
0.4078
±
0.0101
boolq
2
none
0
acc
↑
0.4064
±
0.0086
hellaswag
1
none
0
acc
↑
0.2692
±
0.0044
none
0
acc_norm
↑
0.2757
±
0.0045
piqa
1
none
0
acc
↑
0.5789
±
0.0115
none
0
acc_norm
↑
0.5637
±
0.0116
sciq
1
none
0
acc
↑
0.5980
±
0.0155
none
0
acc_norm
↑
0.5610
±
0.0157
winogrande
1
none
0
acc
↑
0.4957
±
0.0141
Tasks
Version
Filter
n-shot
Metric
Value
Stderr
arc_easy
1
none
5
acc
↑
0.3910
±
0.0100
none
5
acc_norm
↑
0.3645
±
0.0099
boolq
2
none
5
acc
↑
0.5098
±
0.0087
hellaswag
1
none
5
acc
↑
0.2717
±
0.0044
none
5
acc_norm
↑
0.2717
±
0.0044
piqa
1
none
5
acc
↑
0.5686
±
0.0116
none
5
acc_norm
↑
0.5642
±
0.0116
sciq
1
none
5
acc
↑
0.5570
±
0.0157
none
5
acc_norm
↑
0.4970
±
0.0158
winogrande
1
none
5
acc
↑
0.4933
±
0.0141
Runs of ethicalabs Echo-DSRN-114M-v0.1.2 on huggingface.co
197
Total runs
-4
24-hour runs
1
3-day runs
-7
7-day runs
-1.6K
30-day runs
More Information About Echo-DSRN-114M-v0.1.2 huggingface.co Model
Echo-DSRN-114M-v0.1.2 huggingface.co is an AI model on huggingface.co that provides Echo-DSRN-114M-v0.1.2's model effect (), which can be used instantly with this ethicalabs Echo-DSRN-114M-v0.1.2 model. huggingface.co supports a free trial of the Echo-DSRN-114M-v0.1.2 model, and also provides paid use of the Echo-DSRN-114M-v0.1.2. Support call Echo-DSRN-114M-v0.1.2 model through api, including Node.js, Python, http.
Echo-DSRN-114M-v0.1.2 huggingface.co is an online trial and call api platform, which integrates Echo-DSRN-114M-v0.1.2's modeling effects, including api services, and provides a free online trial of Echo-DSRN-114M-v0.1.2, you can try Echo-DSRN-114M-v0.1.2 online for free by clicking the link below.
ethicalabs Echo-DSRN-114M-v0.1.2 online free url in huggingface.co:
Echo-DSRN-114M-v0.1.2 is an open source model from GitHub that offers a free installation service, and any user can find Echo-DSRN-114M-v0.1.2 on GitHub to install. At the same time, huggingface.co provides the effect of Echo-DSRN-114M-v0.1.2 install, users can directly use Echo-DSRN-114M-v0.1.2 installed effect in huggingface.co for debugging and trial. It also supports api for free installation.
Echo-DSRN-114M-v0.1.2 install url in huggingface.co: