Open-weight model
ppt-pythia-160m-uniform250-previous_mse_restore_attention_warmup2000-seed1024-stage2
by Qing Yao qing-yao/ppt-pythia-160m-uniform250-previous_mse_restore_attention_warmup2000-seed1024-stage2
ppt-pythia-160m-uniform250-previous_mse_restore_attention_warmup2000-seed1024-stage2 is an open-weight model from Qing Yao. It has 162M parameters and a 2,048-token context. At 16-bit it needs about 0.4 GB of GPU memory, which fits on 1x MI300X from $1.85 an hour, at the lowest prices in the SAVRN Index.
Runs On
What it takes to serve ppt-pythia-160m-uniform250-previous_mse_restore_attention_warmup2000-seed1024-stage2 (162M parameters): the memory its weights need at each precision, and the cheapest way to rent enough data-center GPUs to hold them.
| Precision | Weights | Memory needed | Cheapest setup | Per hour | Also fits |
|---|---|---|---|---|---|
| 16-bit | 0.3 GB | 0.4 GB | 1x MI300X (192 GB) Vultr |
$1.85 | 1x H100 $1.99 · 1x MI325X $2.00 |
| 8-bit | 0.2 GB | 0.2 GB | 1x MI300X (192 GB) Vultr |
$1.85 | 1x H100 $1.99 · 1x MI325X $2.00 |
| 4-bit | 0.1 GB | 0.1 GB | 1x MI300X (192 GB) Vultr |
$1.85 | 1x H100 $1.99 · 1x MI325X $2.00 |
Memory is the weights at that precision plus 20% for the runtime and a short context; a long context needs more. Prices are the lowest on-demand hourly rates in the SAVRN Index, read Oct 19, 2026.
Model Card
The publisher has not written a card for this model.
Configuration
- Architecture
- GPTNeoXForCausalLM
- Context length (tokens)
- 2,048
- Layers
- 12
- Hidden size
- 768
- Feed-forward size
- 3,072
- Attention heads
- 12
- Vocabulary size
- 50,304
- Model type
- gpt_neox
Identity and Version
- Repository
- qing-yao/ppt-pythia-160m-uniform250-previous_mse_restore_attention_warmup2000-seed1024-stage2
- Publisher
- Qing Yao
- Task
- Not stated by the source
- Modality
- Other
- Library
- Not stated by the source
- Parameters
- 162M parameters
- Languages
- Not stated by the source
- Revision
- 76e868e39e846f420ea5f8597061deea5d46f4f2
- First published
- 2026-09-30
- Last updated
- 2026-09-30
Files and Weights
39 files, 3.3 GB in total. The weights are 17 files totalling 3.2 GB in bin, pt, pth, safetensors.
Every file
| File | Type | Size | SHA-256 |
|---|---|---|---|
| checkpoint-2000/model.safetensors | Weights | 324.7 MB | 4c74cecec13b |
| checkpoint-2000/optimizer.pt | Weights | 649.4 MB | 1972a5a6ac98 |
| checkpoint-2000/rng_state.pth | Weights | 14.6 KB | 276504d728eb |
| checkpoint-2000/scheduler.pt | Weights | 1.5 KB | 37397595a5b2 |
| checkpoint-2000/training_args.bin | Weights | 5.4 KB | 63751a1e09ac |
| checkpoint-4000/model.safetensors | Weights | 324.7 MB | 59c5c1be892b |
| checkpoint-4000/optimizer.pt | Weights | 649.4 MB | 445a6226c0eb |
| checkpoint-4000/rng_state.pth | Weights | 14.6 KB | c5a3cb7f15e7 |
| checkpoint-4000/scheduler.pt | Weights | 1.5 KB | 4c21814734cf |
| checkpoint-4000/training_args.bin | Weights | 5.4 KB | 63751a1e09ac |
| checkpoint-6000/model.safetensors | Weights | 324.7 MB | 5023858ec646 |
| checkpoint-6000/optimizer.pt | Weights | 649.4 MB | 6a6630e97a8b |
| checkpoint-6000/rng_state.pth | Weights | 14.6 KB | 870e59542ad1 |
| checkpoint-6000/scheduler.pt | Weights | 1.5 KB | 5e80ab03ee12 |
| checkpoint-6000/training_args.bin | Weights | 5.4 KB | 63751a1e09ac |
| model.safetensors | Weights | 324.7 MB | 5023858ec646 |
| training_args.bin | Weights | 5.4 KB | 63751a1e09ac |
| checkpoint-2000/config.json | Configuration | 781 B | — |
| checkpoint-2000/generation_config.json | Configuration | 225 B | — |
| checkpoint-2000/trainer_state.json | Configuration | 15.2 KB | — |
| checkpoint-4000/config.json | Configuration | 781 B | — |
| checkpoint-4000/generation_config.json | Configuration | 225 B | — |
| checkpoint-4000/trainer_state.json | Configuration | 29.9 KB | — |
| checkpoint-6000/config.json | Configuration | 781 B | — |
| checkpoint-6000/generation_config.json | Configuration | 225 B | — |
| checkpoint-6000/trainer_state.json | Configuration | 44.6 KB | — |
| config.json | Configuration | 781 B | — |
| experiment.json | Configuration | 12.4 KB | — |
| generation_config.json | Configuration | 225 B | — |
| trainer_state.json | Configuration | 44.6 KB | — |
| .gitattributes | Repository | 1.5 KB | — |
| checkpoint-2000/tokenizer.json | Tokenizer | 3.6 MB | — |
| checkpoint-2000/tokenizer_config.json | Tokenizer | 349 B | — |
| checkpoint-4000/tokenizer.json | Tokenizer | 3.6 MB | — |
| checkpoint-4000/tokenizer_config.json | Tokenizer | 349 B | — |
| checkpoint-6000/tokenizer.json | Tokenizer | 3.6 MB | — |
| checkpoint-6000/tokenizer_config.json | Tokenizer | 349 B | — |
| tokenizer.json | Tokenizer | 3.6 MB | — |
| tokenizer_config.json | Tokenizer | 349 B | — |
License and Download
- License
- Not stated by the source
- Access
- Open weights, no gate
- Download size
- 3.2 GB
Released by Qing Yao through its official repository on Hugging Face.
Memory Requirements
| Precision | Weights in memory |
|---|---|
| As published | 3.2 GB |
| 16-bit | 0.3 GB |
| 8-bit | 0.2 GB |
| 4-bit | 0.1 GB |
Weights only, from the published parameter count; the key-value cache and runtime add to this.
Questions About ppt-pythia-160m-uniform250-previous_mse_restore_attention_warmup2000-seed1024-stage2
How much GPU memory does ppt-pythia-160m-uniform250-previous_mse_restore_attention_warmup2000-seed1024-stage2 need?
About 0.4 GB at 16-bit and 0.1 GB at 4-bit: the weights (162M parameters) plus a working margin. A long context needs more.
What is the cheapest GPU to run ppt-pythia-160m-uniform250-previous_mse_restore_attention_warmup2000-seed1024-stage2 on?
At 16-bit, 1x MI300X from $1.85 an hour; at 4-bit, 1x MI300X from $1.85 an hour, at the lowest on-demand prices the SAVRN Index lists.
What is ppt-pythia-160m-uniform250-previous_mse_restore_attention_warmup2000-seed1024-stage2's context length?
2,048 tokens, from the maximum position embeddings in its published configuration.