ikimyaii/GatedDeltaNet-340M-unstructured-ELSA-70pct
AI model by ikimyaii
Older snapshot: check the original source for current information.
From the publisher
Model card & documentation
Source previewRead the publisher’s intended use, setup instructions, evaluations and limitations. The original model card is the source of truth.
m-a-p/340M-20B-GatedDeltaNet-pure-baseline, pruned to 70% sparsity with ELSA's official code (paper). This is a baseline for the EHWS models in this account. The sparsity is unstructured: one magnitude threshold per layer, so rows end up with different numbers of nonzeros. Every linear layer inside the blocks is pruned; embeddings and lmhead stay dense. Load it with flash-linear-attention installed (import fla) and AutoModelForCausalLM.frompretrained(..., trustremotecode=True). ELSA-official main.py, with the settings of ELSA's own HGRN-1.3B @80% run, unchanged for every model and sparsity: 4096 ADMM steps, z-step every 32 steps, batch 1 x 8 accumulation, lr 2e-4, lambda 0.01, bf16, 32,768 C4 sequences of 2,048 tokens,…
Read the full model card ↗ · Preview checked 2026-10-28T08:03:51.979Z
Inside the original model card — Document outline
Headings are captured from the source. Links open the publisher’s document, not a locally hosted copy.
Links from the model card
References supplied by the publisher, not independently verified endorsements. Check the destination before downloading files or entering credentials.
Documentation belongs to its respective authors. A listing is not a grant of reuse or training rights. Confirm the document’s own terms at the source.
Model overview
ikimyaii/GatedDeltaNet-340M-unstructured-ELSA-70pct is a task-unspecified model repository published by ikimyaii on Hugging Face. A library was not reported.
This page summarizes Hub metadata. For intended use, training data, evaluation results and limitations, consult the original model card.
Model facts
- Task
- Not reported
- Library
- Not reported
- Recent downloads (30 days)
- 0
- Cumulative likes
- 0
- Hugging Face trending score
- 0
- Reported safetensors parameters
- 399,531,200
- Architecture
- GatedDeltaNetForCausalLM
- License
- Not reported — check the model card
- Access
- Not gated by Hugging Face
- Created
- 2026-10-28T06:29:21.000Z
- Last modified
- 2026-10-28T06:29:31.000Z
Compatibility and lineage
Language coverage was not reported.
Reported base models: m-a-p/340M-20B-GatedDeltaNet-pure-baseline
Parameter count is not a RAM/VRAM requirement. Check precision, quantization, context length and runtime compatibility in the model card; no hardware or API-cost claim is inferred here.
Use and evaluate
Check license terms and access requirements first. Review the model card’s documented loading instructions, evaluations and safety limitations. Benchmark scores are not imported by this directory, and popularity is not an accuracy ranking.
No model weights are downloaded or executed by AltAPIs. Never enable remote model code without reviewing it.
Source and freshness
Source: Hugging Face Hub. Metadata observed 2026-10-28T06:47:50.589Z. Daily imports are snapshots, not real-time monitoring.
Popularity and source listings do not establish security, suitability, licensing rights or benchmark performance.