Update README.md

README.md

@@ -69,6 +69,8 @@ base_model:
 
 # AndriLawrence/Qwen-3B-Intent-Microplan-v2
 
+“Local-first 3B model for VR / game companions that outputs strict {dialog, intent, microplan} JSON from a CONTEXT event.”
+
 **English-only** finetune of **Qwen2.5-3B-Instruct** for **intent + microplan–driven NPC dialog**.
 The model reads a structured **CONTEXT JSON** (environment, relationship, mood, signals) and produces:
 
@@ -95,7 +97,7 @@ The model reads a structured **CONTEXT JSON** (environment, relationship, mood,
 ## 📦 Assets
 
   * **LoRA adapters (PEFT, SFT)** → `checkpoints/adapter_final`
-  * **Merged FP16** → `merged/sft-fp16`
+  * **Merged FP16** → `./`
   * **GGUF quants (llama.cpp / llama-cpp-python)** → `gguf/sft-q6_k.gguf`, `gguf/sft-q4_k_m.gguf`
   * **GGUF Style Fine-tune (Example)** → `gguf/rin_style.gguf` (See fine-tuning section)
 
@@ -182,7 +184,7 @@ They balance creativity with JSON stability for Rin:
 
 ```json
 {
-  "temperature": 0.92,
+  "temperature": 0.65,
   "top_p": 0.90,
   "top_k": 40,
   "repetition_penalty": 1.05,
@@ -284,7 +286,7 @@ print(tok.decode(out[0], skip_special_tokens=True))
 ```python
 from transformers import AutoTokenizer, AutoModelForCausalLM
 
-MODEL = "AndriLawrence/Qwen-3B-Intent-Microplan-v2/merged/sft-fp16"
+MODEL = "AndriLawrence/Qwen-3B-Intent-Microplan-v2/"
 
 tok = AutoTokenizer.from_pretrained(MODEL, use_fast=True, trust_remote_code=True)
 model = AutoModelForCausalLM.from_pretrained(