AmareshHebbar/Stretch32K_TinyLlama_1.1B_PI
text-generation model by AmareshHebbar
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.
TinyLlama/TinyLlamav1.1 was trained with a 2048 token window. This model reads 16384 tokens, a 8x stretch, trained on a single GPU. It is one of six models in the Stretch32K collection, all trained on the same PG19 chunks with a different context extension method and evaluated on the same tests. Position Interpolation. Every position is divided by the stretch factor, so the longest new position lands inside the range the model was trained on. Then the whole model is finetuned on long PG19 books so it adapts to the denser positions. Training data: AmareshHebbar/Stretch32KDataTinyLlama1.1B. Memory tricks: PyTorch FlashAttention kernels, gradient checkpointing, chunked loss over the vocabulary, and packed sequences with no padding. Retrieval feeds a model a few chunks; it cannot reason across a whole book or contract in one pass. The base model rows below show what happens past 2048 tokens with no training, and with the popular training…
Read the full model card ↗ · Preview checked 2026-11-16T07:19:13.638Z
Inside the original model card — Document outline
- Stretch32K PI (TinyLlama 1.1B)
- Method
- Why finetune
- Results
- Perplexity by context length
- Retrieval by context length (this model)
- Training hyperparameters
- Usage
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. Reported project/model license: apache-2.0. A listing is not a grant of reuse or training rights. Confirm the document’s own terms at the source.
Model overview
AmareshHebbar/Stretch32K_TinyLlama_1.1B_PI is a text-generation model repository published by AmareshHebbar on Hugging Face. The source reports the transformers library.
This page summarizes Hub metadata. For intended use, training data, evaluation results and limitations, consult the original model card.
Model facts
- Task
- text-generation
- Library
- transformers
- Recent downloads (30 days)
- 0
- Cumulative likes
- 0
- Hugging Face trending score
- 0
- Reported safetensors parameters
- 1,100,048,384
- Architecture
- LlamaForCausalLM
- License
- apache-2.0
- Access
- Not gated by Hugging Face
- Created
- 2026-11-16T06:55:45.000Z
- Last modified
- 2026-11-16T07:00:12.000Z
Compatibility and lineage
Reported languages: en
Reported base models: TinyLlama/TinyLlama_v1.1
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 2026-11-16T07:00:14.548Z. Daily imports are snapshots, not real-time monitoring.
Popularity and source listings do not establish security, suitability, licensing rights or benchmark performance.