Upload fine-tuned Llama 3.1 8B QLoRA model

README.md

@@ -1,64 +1,65 @@
+
 ---
-            library_name: peft
-            base_model: meta-llama/Llama-3.1-8B
-            tags:
-            - llama
-            - lora
-            - qlora
-            - fine-tuned
-            license: llama3.1
-            language:
-            - en
-            pipeline_tag: text-generation
-            ---
+    library_name: peft
+    base_model: meta-llama/Llama-3.1-8B
+    tags:
+    - llama
+    - lora
+    - qlora
+    - fine-tuned
+    license: llama3.1
+    language:
+    - en
+    pipeline_tag: text-generation
+    ---
 
-            # Llama 3.1 8B - Robot Task Planning (QLoRA Fine-tuned)
+    # Llama 3.1 8B - Robot Task Planning (QLoRA Fine-tuned)
 
-            This model is a fine-tuned version of [meta-llama/Llama-3.1-8B](https://huggingface.co/meta-llama/Llama-3.1-8B) specialized for **robot task planning** using QLoRA (4-bit quantization + LoRA).
+    This model is a fine-tuned version of [meta-llama/Llama-3.1-8B](https://huggingface.co/meta-llama/Llama-3.1-8B) specialized for **robot task planning** using QLoRA (4-bit quantization + LoRA).
 
-            The model converts natural language robot commands into structured task sequences for construction robots including excavators and dump trucks.
+    The model converts natural language robot commands into structured task sequences for construction robots including excavators and dump trucks.
 
-            ## Model Details
+    ## Model Details
 
-            - **Base Model**: meta-llama/Llama-3.1-8B
-            - **Fine-tuning Method**: QLoRA (4-bit quantization + LoRA)
-            - **LoRA Rank**: 16
-            - **LoRA Alpha**: 32
-            - **Target Modules**: q_proj, k_proj, v_proj, o_proj, gate_proj, up_proj, down_proj
+    - **Base Model**: meta-llama/Llama-3.1-8B
+    - **Fine-tuning Method**: QLoRA (4-bit quantization + LoRA)
+    - **LoRA Rank**: 16
+    - **LoRA Alpha**: 32
+    - **Target Modules**: q_proj, k_proj, v_proj, o_proj, gate_proj, up_proj, down_proj
 
-            ## Usage
+    ## Usage
 
-            ```python
-            from transformers import AutoTokenizer, AutoModelForCausalLM
-            from peft import PeftModel
+    ```python
+    from transformers import AutoTokenizer, AutoModelForCausalLM
+    from peft import PeftModel
 
-            # Load tokenizer and base model
-            tokenizer = AutoTokenizer.from_pretrained("meta-llama/Llama-3.1-8B")
-            base_model = AutoModelForCausalLM.from_pretrained("meta-llama/Llama-3.1-8B")
+    # Load tokenizer and base model
+    tokenizer = AutoTokenizer.from_pretrained("meta-llama/Llama-3.1-8B")
+    base_model = AutoModelForCausalLM.from_pretrained("meta-llama/Llama-3.1-8B")
 
-            # Load LoRA adapter
-            model = PeftModel.from_pretrained(base_model, "YongdongWang/llama-3.1-8b-dart-qlora")
+    # Load LoRA adapter
+    model = PeftModel.from_pretrained(base_model, "YongdongWang/llama-3.1-8b-dart-qlora")
 
-            # Generate robot task sequence
-            command = "Deploy Excavator 1 to Soil Area 1 for excavation"
-            inputs = tokenizer(command, return_tensors="pt")
-            outputs = model.generate(**inputs, max_new_tokens=5120)
-            response = tokenizer.decode(outputs[0], skip_special_tokens=True)
-            print(response)
+    # Generate robot task sequence
+    command = "Deploy Excavator 1 to Soil Area 1 for excavation"
+    inputs = tokenizer(command, return_tensors="pt")
+    outputs = model.generate(**inputs, max_new_tokens=5120)
+    response = tokenizer.decode(outputs[0], skip_special_tokens=True)
+    print(response)
 
-            Training Details
+    Training Details
 
-            Training Data: DART LLM Tasks - Robot command and task planning dataset
-            Domain: Construction robotics (excavators, dump trucks, soil/rock areas)
-            Training Epochs: 5
-            Batch Size: 16 (with gradient accumulation)
-            Learning Rate: 2e-4
-            Optimizer: paged_adamw_8bit
+    Training Data: DART LLM Tasks - Robot command and task planning dataset
+    Domain: Construction robotics (excavators, dump trucks, soil/rock areas)
+    Training Epochs: 5
+    Batch Size: 16 (with gradient accumulation)
+    Learning Rate: 2e-4
+    Optimizer: paged_adamw_8bit
 
-            Capabilities
+    Capabilities
 
-            Multi-robot coordination: Handle multiple excavators and dump trucks
-            Task dependencies: Generate proper task sequences with dependencies
-            Spatial reasoning: Understand soil areas, rock areas, puddles, and navigation
-            Action planning: Convert commands to structured JSON task definitions
-            
+    Multi-robot coordination: Handle multiple excavators and dump trucks
+    Task dependencies: Generate proper task sequences with dependencies
+    Spatial reasoning: Understand soil areas, rock areas, puddles, and navigation
+    Action planning: Convert commands to structured JSON task definitions
+    

adapter_config.json

@@ -24,13 +24,13 @@
   "rank_pattern": {},
   "revision": null,
   "target_modules": [
-    "v_proj",
-    "down_proj",
-    "up_proj",
-    "q_proj",
     "k_proj",
+    "down_proj",
+    "gate_proj",
+    "v_proj",
     "o_proj",
-    "gate_proj"
+    "up_proj",
+    "q_proj"
   ],
   "task_type": "CAUSAL_LM",
   "trainable_token_indices": null,

adapter_model.safetensors

@@ -1,3 +1,3 @@
 version https://git-lfs.github.com/spec/v1
-oid sha256:a2c34c3f7912c764c06876b19e3358996b1df3fd836e8639590935332bcdb878
+oid sha256:686aa6705c3b268bfc62872bc6a20ee3f78af943123a851082ea5a8beecf764b
 size 167832240

checkpoint-24/adapter_config.json

@@ -24,13 +24,13 @@
   "rank_pattern": {},
   "revision": null,
   "target_modules": [
-    "v_proj",
-    "down_proj",
-    "up_proj",
-    "q_proj",
     "k_proj",
+    "down_proj",
+    "gate_proj",
+    "v_proj",
     "o_proj",
-    "gate_proj"
+    "up_proj",
+    "q_proj"
   ],
   "task_type": "CAUSAL_LM",
   "trainable_token_indices": null,

checkpoint-24/adapter_model.safetensors

@@ -1,3 +1,3 @@
 version https://git-lfs.github.com/spec/v1
-oid sha256:1badd577fd317617bda766d393842d69930dd84a7c9cbf5635c0f3723724fbe8
+oid sha256:ff80ada358f7d45551b795de60e3448a3be3b37ca434a09529d49259e740e104
 size 167832240

checkpoint-24/optimizer.pt

@@ -1,3 +1,3 @@
 version https://git-lfs.github.com/spec/v1
-oid sha256:ce411169d94121fa1d28e23b2471e35ac5577673cc20d674bf41714359bb5b13
+oid sha256:6b5ab0f12eed125e6b567965fc0df61f90b689c01a7cf9d5b32828e25f5b9211
 size 85728532

checkpoint-24/tokenizer.json

@@ -1,3 +1,3 @@
 version https://git-lfs.github.com/spec/v1
-oid sha256:881912e4e7f25194ad8e82fd5f12292ba4a70376303f608aa270fb4bde3bc9b7
-size 17210189
+oid sha256:6b9e4e7fb171f92fd137b777cc2714bf87d11576700a1dcd7a399e7bbe39537b
+size 17209920

checkpoint-24/trainer_state.json

@@ -1,6 +1,6 @@
 {
   "best_global_step": 24,
-  "best_metric": 0.15350762009620667,
+  "best_metric": 0.026073265820741653,
   "best_model_checkpoint": "./outputs/llama3.1-8b-lora-qlora-dart-llm/checkpoint-24",
   "epoch": 4.0,
   "eval_steps": 500,
@@ -11,62 +11,62 @@
   "log_history": [
     {
       "epoch": 0.8791208791208791,
-      "grad_norm": 0.5172705054283142,
+      "grad_norm": 0.5106796026229858,
       "learning_rate": 0.00019594929736144976,
-      "loss": 0.9618,
+      "loss": 0.6176,
       "step": 5
     },
     {
       "epoch": 1.0,
-      "eval_loss": 0.5014150142669678,
-      "eval_runtime": 2.0191,
-      "eval_samples_per_second": 5.448,
-      "eval_steps_per_second": 5.448,
+      "eval_loss": 0.15218934416770935,
+      "eval_runtime": 2.7083,
+      "eval_samples_per_second": 4.062,
+      "eval_steps_per_second": 4.062,
       "step": 6
     },
     {
       "epoch": 1.7032967032967035,
-      "grad_norm": 0.44574207067489624,
+      "grad_norm": 0.4140273630619049,
       "learning_rate": 0.00015406408174555976,
-      "loss": 0.3429,
+      "loss": 0.1118,
       "step": 10
     },
     {
       "epoch": 2.0,
-      "eval_loss": 0.21539340913295746,
-      "eval_runtime": 2.0264,
-      "eval_samples_per_second": 5.428,
-      "eval_steps_per_second": 5.428,
+      "eval_loss": 0.04035872593522072,
+      "eval_runtime": 2.7366,
+      "eval_samples_per_second": 4.02,
+      "eval_steps_per_second": 4.02,
       "step": 12
     },
     {
       "epoch": 2.5274725274725274,
-      "grad_norm": 0.338784396648407,
+      "grad_norm": 0.19475381076335907,
       "learning_rate": 8.57685161726715e-05,
-      "loss": 0.1674,
+      "loss": 0.0205,
       "step": 15
     },
     {
       "epoch": 3.0,
-      "eval_loss": 0.16001786291599274,
-      "eval_runtime": 2.029,
-      "eval_samples_per_second": 5.421,
-      "eval_steps_per_second": 5.421,
+      "eval_loss": 0.029018037021160126,
+      "eval_runtime": 2.6384,
+      "eval_samples_per_second": 4.169,
+      "eval_steps_per_second": 4.169,
       "step": 18
     },
     {
       "epoch": 3.3516483516483517,
-      "grad_norm": 0.2452668696641922,
+      "grad_norm": 0.0900561586022377,
       "learning_rate": 2.4425042564574184e-05,
-      "loss": 0.1408,
+      "loss": 0.0216,
       "step": 20
     },
     {
       "epoch": 4.0,
-      "eval_loss": 0.15350762009620667,
-      "eval_runtime": 2.0182,
-      "eval_samples_per_second": 5.451,
-      "eval_steps_per_second": 5.451,
+      "eval_loss": 0.026073265820741653,
+      "eval_runtime": 2.7519,
+      "eval_samples_per_second": 3.997,
+      "eval_steps_per_second": 3.997,
       "step": 24
     }
   ],

checkpoint-24/training_args.bin

@@ -1,3 +1,3 @@
 version https://git-lfs.github.com/spec/v1
-oid sha256:2846e9375e9180e7cfcd8c59ec58088e8cc4b41268da419e806212bdaeecdb21
+oid sha256:5c790f6569290a09d32f8a5b1e2904334b6f58a1c2321c61a41455b1cc5658b8
 size 5432

checkpoint-25/adapter_config.json

@@ -24,13 +24,13 @@
   "rank_pattern": {},
   "revision": null,
   "target_modules": [
-    "v_proj",
-    "down_proj",
-    "up_proj",
-    "q_proj",
     "k_proj",
+    "down_proj",
+    "gate_proj",
+    "v_proj",
     "o_proj",
-    "gate_proj"
+    "up_proj",
+    "q_proj"
   ],
   "task_type": "CAUSAL_LM",
   "trainable_token_indices": null,

checkpoint-25/adapter_model.safetensors

@@ -1,3 +1,3 @@
 version https://git-lfs.github.com/spec/v1
-oid sha256:a2c34c3f7912c764c06876b19e3358996b1df3fd836e8639590935332bcdb878
+oid sha256:686aa6705c3b268bfc62872bc6a20ee3f78af943123a851082ea5a8beecf764b
 size 167832240

checkpoint-25/optimizer.pt

@@ -1,3 +1,3 @@
 version https://git-lfs.github.com/spec/v1
-oid sha256:b390f6d1a3d168cedf3457a776111498b89dfb304d7edd9f7047b94305887e1e
+oid sha256:0b2d76bcc892f6aa50acb4b2f920ced51c29d48a2e262a4e75f95ad24696d858
 size 85728532

checkpoint-25/tokenizer.json

@@ -1,3 +1,3 @@
 version https://git-lfs.github.com/spec/v1
-oid sha256:881912e4e7f25194ad8e82fd5f12292ba4a70376303f608aa270fb4bde3bc9b7
-size 17210189
+oid sha256:6b9e4e7fb171f92fd137b777cc2714bf87d11576700a1dcd7a399e7bbe39537b
+size 17209920

checkpoint-25/trainer_state.json

@@ -1,6 +1,6 @@
 {
   "best_global_step": 25,
-  "best_metric": 0.15348903834819794,
+  "best_metric": 0.02604079246520996,
   "best_model_checkpoint": "./outputs/llama3.1-8b-lora-qlora-dart-llm/checkpoint-25",
   "epoch": 4.175824175824176,
   "eval_steps": 500,
@@ -11,77 +11,77 @@
   "log_history": [
     {
       "epoch": 0.8791208791208791,
-      "grad_norm": 0.5172705054283142,
+      "grad_norm": 0.5106796026229858,
       "learning_rate": 0.00019594929736144976,
-      "loss": 0.9618,
+      "loss": 0.6176,
       "step": 5
     },
     {
       "epoch": 1.0,
-      "eval_loss": 0.5014150142669678,
-      "eval_runtime": 2.0191,
-      "eval_samples_per_second": 5.448,
-      "eval_steps_per_second": 5.448,
+      "eval_loss": 0.15218934416770935,
+      "eval_runtime": 2.7083,
+      "eval_samples_per_second": 4.062,
+      "eval_steps_per_second": 4.062,
       "step": 6
     },
     {
       "epoch": 1.7032967032967035,
-      "grad_norm": 0.44574207067489624,
+      "grad_norm": 0.4140273630619049,
       "learning_rate": 0.00015406408174555976,
-      "loss": 0.3429,
+      "loss": 0.1118,
       "step": 10
     },
     {
       "epoch": 2.0,
-      "eval_loss": 0.21539340913295746,
-      "eval_runtime": 2.0264,
-      "eval_samples_per_second": 5.428,
-      "eval_steps_per_second": 5.428,
+      "eval_loss": 0.04035872593522072,
+      "eval_runtime": 2.7366,
+      "eval_samples_per_second": 4.02,
+      "eval_steps_per_second": 4.02,
       "step": 12
     },
     {
       "epoch": 2.5274725274725274,
-      "grad_norm": 0.338784396648407,
+      "grad_norm": 0.19475381076335907,
       "learning_rate": 8.57685161726715e-05,
-      "loss": 0.1674,
+      "loss": 0.0205,
       "step": 15
     },
     {
       "epoch": 3.0,
-      "eval_loss": 0.16001786291599274,
-      "eval_runtime": 2.029,
-      "eval_samples_per_second": 5.421,
-      "eval_steps_per_second": 5.421,
+      "eval_loss": 0.029018037021160126,
+      "eval_runtime": 2.6384,
+      "eval_samples_per_second": 4.169,
+      "eval_steps_per_second": 4.169,
       "step": 18
     },
     {
       "epoch": 3.3516483516483517,
-      "grad_norm": 0.2452668696641922,
+      "grad_norm": 0.0900561586022377,
       "learning_rate": 2.4425042564574184e-05,
-      "loss": 0.1408,
+      "loss": 0.0216,
       "step": 20
     },
     {
       "epoch": 4.0,
-      "eval_loss": 0.15350762009620667,
-      "eval_runtime": 2.0182,
-      "eval_samples_per_second": 5.451,
-      "eval_steps_per_second": 5.451,
+      "eval_loss": 0.026073265820741653,
+      "eval_runtime": 2.7519,
+      "eval_samples_per_second": 3.997,
+      "eval_steps_per_second": 3.997,
       "step": 24
     },
     {
       "epoch": 4.175824175824176,
-      "grad_norm": 0.26511526107788086,
+      "grad_norm": 0.09460670500993729,
       "learning_rate": 0.0,
-      "loss": 0.1287,
+      "loss": 0.0157,
       "step": 25
     },
     {
       "epoch": 4.175824175824176,
-      "eval_loss": 0.15348903834819794,
-      "eval_runtime": 2.0275,
-      "eval_samples_per_second": 5.425,
-      "eval_steps_per_second": 5.425,
+      "eval_loss": 0.02604079246520996,
+      "eval_runtime": 2.8313,
+      "eval_samples_per_second": 3.885,
+      "eval_steps_per_second": 3.885,
       "step": 25
     }
   ],

checkpoint-25/training_args.bin

@@ -1,3 +1,3 @@
 version https://git-lfs.github.com/spec/v1
-oid sha256:2846e9375e9180e7cfcd8c59ec58088e8cc4b41268da419e806212bdaeecdb21
+oid sha256:5c790f6569290a09d32f8a5b1e2904334b6f58a1c2321c61a41455b1cc5658b8
 size 5432

tokenizer.json

@@ -1,3 +1,3 @@
 version https://git-lfs.github.com/spec/v1
-oid sha256:881912e4e7f25194ad8e82fd5f12292ba4a70376303f608aa270fb4bde3bc9b7
-size 17210189
+oid sha256:6b9e4e7fb171f92fd137b777cc2714bf87d11576700a1dcd7a399e7bbe39537b
+size 17209920

training_args.bin

@@ -1,3 +1,3 @@
 version https://git-lfs.github.com/spec/v1
-oid sha256:2846e9375e9180e7cfcd8c59ec58088e8cc4b41268da419e806212bdaeecdb21
+oid sha256:5c790f6569290a09d32f8a5b1e2904334b6f58a1c2321c61a41455b1cc5658b8
 size 5432