Open-weight model
rexgradient-psim-meditron7b
by Zachary Horvitz zhorvitz/rexgradient-psim-meditron7b
rexgradient-psim-meditron7b is a model from Zachary Horvitz (access requested at publisher). Its published files total 2.6 GB.
Private. Trained on ReXGradient-160K reports, which are distributed under a data use agreement (rajpurkarlab/ReXGradient-160K). Do not make public or share outside that agreement: the adapters can reproduce training text.
Model Card
Private. Trained on ReXGradient-160K reports, which are distributed under a data use agreement (rajpurkarlab/ReXGradient-160K). Do not make public or share outside that agreement: the adapters can reproduce training text. Each subfolder is a LoRA adapter (rank 128, alpha 256, bf16) on epfl-llm/meditron-7b that samples whole chest-X-ray reports unconditionally from a lone BOS, serialized as Training data is a deterministic fraction of the ReXGradient train split: reports with sha256(id) mod K == 0 (K = 10 or 40). The fractions are nested (the 1/40 rows are a subset of the 1/10 rows), and all rows with sha256(id) mod 5!= 0 are never seen by either model. Canonical psim recipe: lr 5e-5 cosine…
Excerpt from the card by Zachary Horvitz.
Identity and Version
- Repository
- zhorvitz/rexgradient-psim-meditron7b
- Publisher
- Zachary Horvitz
- Task
- Not stated by the source
- Modality
- Other
- Library
- peft
- Parameters
- Not stated by the source
- Languages
- Not stated by the source
- Revision
- c8fa4032bf433cf504278565079a7876ffea7f1b
- First published
- 2026-09-24
- Last updated
- 2026-09-24
Files and Weights
20 files, 2.6 GB in total. The weights are 2 files totalling 2.6 GB in safetensors.
Every file
| File | Type | Size | SHA-256 |
|---|---|---|---|
| 10th/adapter_model.safetensors | Weights | 1.3 GB | — |
| 40th/adapter_model.safetensors | Weights | 1.3 GB | — |
| 10th/adapter_config.json | Configuration | 1.1 KB | — |
| 10th/added_tokens.json | Configuration | 344 B | — |
| 10th/data_manifest.json | Configuration | 1.2 KB | — |
| 10th/special_tokens_map.json | Configuration | 1.2 KB | — |
| 40th/adapter_config.json | Configuration | 1.1 KB | — |
| 40th/added_tokens.json | Configuration | 344 B | — |
| 40th/data_manifest.json | Configuration | 1.2 KB | — |
| 40th/special_tokens_map.json | Configuration | 1.2 KB | — |
| README.md | Documentation | 2.2 KB | — |
| 10th/git_commit.txt | Other | 41 B | — |
| 40th/git_commit.txt | Other | 41 B | — |
| .gitattributes | Repository | 1.5 KB | — |
| 10th/tokenizer.json | Tokenizer | 3.6 MB | — |
| 10th/tokenizer.model | Tokenizer | 499.7 KB | — |
| 10th/tokenizer_config.json | Tokenizer | 4.2 KB | — |
| 40th/tokenizer.json | Tokenizer | 3.6 MB | — |
| 40th/tokenizer.model | Tokenizer | 499.7 KB | — |
| 40th/tokenizer_config.json | Tokenizer | 4.2 KB | — |
License and Download
- License
- Not stated by the source
- Access
- Access requested at publisher
- Download size
- 2.6 GB
Zachary Horvitz grants access through its official repository on Hugging Face.
Built From
- Adapter of epfl-llm/meditron-7b
- Derived from epfl-llm/meditron-7b
Memory Requirements
| Precision | Weights in memory |
|---|---|
| As published | 2.6 GB |
Weights only, from the published parameter count; the key-value cache and runtime add to this.