cmedina/agenx-lora-training
1
0
Fork 0
Code Issues Pull Requests Actions Packages Projects Releases Wiki Activity

docs: rename config and docs to reflect Ornith-35B model

Browse Source
This commit is contained in:
Christian Medina
2026-06-30 20:41:53 -04:00
parent 6f31fa0197
commit 6bd6662f5f
3 changed files with 7 additions and 7 deletions
Download Patch File Download Diff File Expand all files Collapse all files

8
train-on-this-server.sh
View File

@@ -70,8 +70,8 @@ fi
echo ""
echo "=== Starting Training ==="
echo "Config: training/configs/llama2-7b-lora.yaml"
echo "Output: training/output/llama2-7b-lora/"
echo "Config: training/configs/ornith-35b-lora.yaml"
echo "Output: training/output/ornith-35b-lora/"
echo ""
echo "Training will take 6-24 hours depending on GPU."
echo "Press Ctrl+C to stop (model will be saved at checkpoint)."
@@ -79,8 +79,8 @@ echo ""
# Use torchrun for distributed training (2 GPUs)
# Add --check-only to validate without training:
# torchrun --nproc_per_node=2 training/scripts/train.py --config training/configs/llama2-7b-lora.yaml --check-only
torchrun --nproc_per_node=2 training/scripts/train.py --config training/configs/llama2-7b-lora.yaml
# torchrun --nproc_per_node=2 training/scripts/train.py --config training/configs/ornith-35b-lora.yaml --check-only
torchrun --nproc_per_node=2 training/scripts/train.py --config training/configs/ornith-35b-lora.yaml
echo ""
echo "=== Training Complete ==="

6
training/README.md
View File

@@ -53,13 +53,13 @@ This splits the 20k dataset into train/test sets (95/5).
### 2. Train Model
```bash
python scripts/train.py --config configs/llama2-7b-lora.yaml
python scripts/train.py --config configs/ornith-35b-lora.yaml
```
Or with custom parameters:
```bash
python scripts/train.py --config configs/llama2-7b-lora.yaml --epochs 5 --batch-size 8
python scripts/train.py --config configs/ornith-35b-lora.yaml --epochs 5 --batch-size 8
```
Training takes approximately 6-24 hours depending on GPU.
@@ -68,7 +68,7 @@ Training takes approximately 6-24 hours depending on GPU.
```bash
python scripts/inference.py \
--model output/llama2-7b-lora \
--model output/ornith-35b-lora \
--task "Fix parser crash on malformed JSON" \
--files src/parser.cpp \
--tests-run --test-count 294 \

0
training/configs/llama2-7b-lora.yaml → training/configs/ornith-35b-lora.yaml
View File