PondLLM Qwen3.5 2B communication adapter
This is a local experimental LoRA adapter for Qwen/Qwen3.5-2B-Base. It produces bounded JSON
actions for the PondLLM simulator. It is not a general chat model.
The V3 curriculum retains the balanced action examples and adds paired communication cases:
- signal a locally visible food coordinate when a nearby organism cannot see it;
- suppress a redundant signal when the recipient already sees the food;
- move toward food described in received memory;
- preserve the control behavior when that memory is absent.
Training
- Records: 20,006 (19,806 train, 200 evaluation)
- Method: 4-bit QLoRA SFT
- LoRA: rank 8, alpha 16, dropout 0.05
- Maximum length: 1,024
- Effective batch size: 32
- Learning rate:
2e-4 - Epochs: 2
- Seed: 7
- Runtime on RTX 5070 Ti: 13,118.7 seconds
- Aggregate train loss: 0.05582
Exact parameters and framework versions are in run_manifest.json.
Evaluation
On 600 disjoint deterministic communication cases:
- syntax valid: 100.00%
- legal actions: 100.00%
- useful-signal rate: 99.33%
- redundant-signal rate: 0.00%
- payload-coordinate accuracy: 99.33%
- recipient-informed exact accuracy: 100.00%
- recipient-control exact accuracy: 100.00%
On the older 600-case fixed-stratified action set, syntax and legality were 100.00%, macro action accuracy was 70.50%, and exact accuracy was 58.17%.
See the night-four communication report for comparisons, limitations, hashes, and live-world results.
Limitations
This adapter imitates a specified communication policy. It does not demonstrate emergent language, subjective experience, or improved evolutionary fitness. Its JSON schema is specific to PondLLM. The base model is required to use the adapter.
- Downloads last month
- 4
Model tree for Xnizzorg/pondllm-qwen3.5-2b-base-communication-sft
Base model
Qwen/Qwen3.5-2B-Base
from peft import PeftModel from transformers import AutoModelForCausalLM base_model = AutoModelForCausalLM.from_pretrained("Qwen/Qwen3.5-2B-Base") model = PeftModel.from_pretrained(base_model, "Xnizzorg/pondllm-qwen3.5-2b-base-communication-sft")