Upload step-150 LoRA adapter from opus-4b-dsl-mixed run
Browse files- README.md +37 -0
- adapter_config.json +31 -0
- adapter_model.safetensors +3 -0
README.md
ADDED
|
@@ -0,0 +1,37 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
---
|
| 2 |
+
base_model: Qwen/Qwen3.5-4B
|
| 3 |
+
library_name: peft
|
| 4 |
+
tags:
|
| 5 |
+
- tinker
|
| 6 |
+
- peft
|
| 7 |
+
- lora
|
| 8 |
+
- opus-magnum
|
| 9 |
+
---
|
| 10 |
+
|
| 11 |
+
# opus-4b-dsl-mixed-step150-2026-04-29
|
| 12 |
+
|
| 13 |
+
LoRA adapter (rank 32) trained with RL on a custom Opus-Magnum-style motion-planning task using the **dsl** answer representation. Snapshot at training step 150 / 300.
|
| 14 |
+
|
| 15 |
+
## Source training run
|
| 16 |
+
|
| 17 |
+
- wandb: [opus-4b-dsl-mixed-2026-04-29 (mqhz79iy)](https://wandb.ai/websim/opus-task/runs/mqhz79iy)
|
| 18 |
+
- tinker checkpoint: `tinker://addea7a1-fbe5-59b8-a467-5de3736c3404:train:0/sampler_weights/000150`
|
| 19 |
+
- distances: 1, 2, 3, 4
|
| 20 |
+
- task types: move, transmute, bond
|
| 21 |
+
- hard_task_fraction: 0.15 (cap on bond + d=4 share of train pool)
|
| 22 |
+
- learning rate: 1e-5
|
| 23 |
+
- group size: 8, groups per batch: 16
|
| 24 |
+
- renderer: qwen3_5_disable_thinking
|
| 25 |
+
|
| 26 |
+
## Usage
|
| 27 |
+
|
| 28 |
+
```python
|
| 29 |
+
from peft import PeftModel
|
| 30 |
+
from transformers import AutoModelForCausalLM, AutoTokenizer
|
| 31 |
+
|
| 32 |
+
base = "Qwen/Qwen3.5-4B"
|
| 33 |
+
adapter = "maxbittker/opus-4b-dsl-mixed-step150-2026-04-29"
|
| 34 |
+
tok = AutoTokenizer.from_pretrained(base)
|
| 35 |
+
model = AutoModelForCausalLM.from_pretrained(base, device_map="auto")
|
| 36 |
+
model = PeftModel.from_pretrained(model, adapter)
|
| 37 |
+
```
|
adapter_config.json
ADDED
|
@@ -0,0 +1,31 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
{
|
| 2 |
+
"alpha_pattern": {},
|
| 3 |
+
"auto_mapping": null,
|
| 4 |
+
"base_model_name_or_path": "Qwen/Qwen3.5-4B",
|
| 5 |
+
"bias": "none",
|
| 6 |
+
"corda_config": null,
|
| 7 |
+
"eva_config": null,
|
| 8 |
+
"exclude_modules": null,
|
| 9 |
+
"fan_in_fan_out": false,
|
| 10 |
+
"inference_mode": false,
|
| 11 |
+
"init_lora_weights": true,
|
| 12 |
+
"layer_replication": null,
|
| 13 |
+
"layers_pattern": null,
|
| 14 |
+
"layers_to_transform": null,
|
| 15 |
+
"loftq_config": {},
|
| 16 |
+
"lora_alpha": 32,
|
| 17 |
+
"lora_bias": false,
|
| 18 |
+
"lora_dropout": 0,
|
| 19 |
+
"megatron_config": null,
|
| 20 |
+
"megatron_core": "megatron.core",
|
| 21 |
+
"modules_to_save": null,
|
| 22 |
+
"peft_type": "LORA",
|
| 23 |
+
"r": 32,
|
| 24 |
+
"rank_pattern": {},
|
| 25 |
+
"revision": null,
|
| 26 |
+
"target_modules": "all-linear",
|
| 27 |
+
"task_type": "CAUSAL_LM",
|
| 28 |
+
"trainable_token_indices": null,
|
| 29 |
+
"use_dora": false,
|
| 30 |
+
"use_rslora": false
|
| 31 |
+
}
|
adapter_model.safetensors
ADDED
|
@@ -0,0 +1,3 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
version https://git-lfs.github.com/spec/v1
|
| 2 |
+
oid sha256:b93a9589efad46e44a4508a11dba6018e1bcda621a6ffca7e6837b5964cddab1
|
| 3 |
+
size 291703264
|