Downloads · 30 days
0
echo-mdlm-27m-ladder
lupodevelop/echo-mdlm-27m-ladder
echo-mdlm-27m-ladder is a fill-mask model from lupodevelop. Use it when you need the model to fill a missing word. It is set up for echo-1.58. The card lists the license as apache-2.0.
A bundle of 27M-parameter checkpoints (dmodel 448, 8 layers) that back the paper's mechanistic results: the recovery ablation, the multi-epoch repetition ladder (1 to 20 epochs on a fixed 75M-token pool), the techniqu…
Access
Public
Updated Jul 20, 2026
Repo size
1.6 GB
Likes
0
Public
Repo makeup
Click a slice to open those files.
.pt1.6 GB · 100%
At a glance
- Task
- Fill-Mask
- Library
- echo-1.58
- License
- apache-2.0
- Access
- Public
- Created
- Jul 20, 2026
- Updated
- Jul 20, 2026
- SHA
- 63a5c38d
Try a prompt
Model card
From the Hugging Face model README
Echo-1.58 27M Research Ladder
A bundle of 27M-parameter checkpoints (d_model 448, 8 layers) that back the paper's
mechanistic results: the recovery ablation, the multi-epoch repetition ladder (1 to 20
epochs on a fixed 75M-token pool), the technique arms, and the fresh-data baselines. Each
file is bfloat16, paired with its FP16 twin where relevant. These are provenance artifacts for
reproducing the tables and figures, not standalone models. See
https://github.com/lupodevelop/echo-1.58 for the code that consumes them and
results/RESULTS.md for the numbers.
Contents
fp16_e{1,5,10,20}.pt,tern_e{5,10,20}.pt,rec_e{5,10}.ptrepetition ladderrec_e5_lr3.pt,tern_e5_lr3.pt,rec_e5_vsched.pt,rec_e5_lr8.pt,rec_e5_rewarm.pttechnique and learning-rate arms
- Task
- Fill-Mask
- Library
- echo-1.58
- License
- apache-2.0
- Languages
- it
- Created
- Jul 20, 2026
- Updated
- Jul 20, 2026