Upload Llama 3.1 8B fine-tuned for lead response prediction (Trained with Unsloth)

README.md

@@ -4,6 +4,7 @@ tags:
 - peft
 - lora
 - llama
+- unsloth
 ---
 
 # Lead Response Prediction - Llama 3.1 8B LoRA

adapter_config.json

@@ -3,11 +3,11 @@
   "alpha_pattern": {},
   "arrow_config": null,
   "auto_mapping": {
-    "base_model_class": "LlamaForCausalLM",
-    "parent_library": "transformers.models.llama.modeling_llama",
+    "base_model_class": "Qwen2ForCausalLM",
+    "parent_library": "transformers.models.qwen2.modeling_qwen2",
     "unsloth_fixed": true
   },
-  "base_model_name_or_path": "unsloth/meta-llama-3.1-8b-bnb-4bit",
+  "base_model_name_or_path": "unsloth/qwen2.5-14b-instruct-bnb-4bit",
   "bias": "none",
   "corda_config": null,
   "ensure_weight_tying": false,
@@ -20,7 +20,7 @@
   "layers_pattern": null,
   "layers_to_transform": null,
   "loftq_config": {},
-  "lora_alpha": 32,
+  "lora_alpha": 64,
   "lora_bias": false,
   "lora_dropout": 0.05,
   "megatron_config": null,
@@ -29,16 +29,16 @@
   "peft_type": "LORA",
   "peft_version": "0.18.0",
   "qalora_group_size": 16,
-  "r": 16,
+  "r": 32,
   "rank_pattern": {},
   "revision": null,
   "target_modules": [
     "k_proj",
-    "q_proj",
+    "v_proj",
     "up_proj",
-    "down_proj",
     "o_proj",
-    "v_proj",
+    "down_proj",
+    "q_proj",
     "gate_proj"
   ],
   "target_parameters": null,

adapter_model.safetensors

@@ -1,3 +1,3 @@
 version https://git-lfs.github.com/spec/v1
-oid sha256:eb75825349a3497c9da190d78972b78ef765e759baee24638d29f396fc1695ae
-size 167832240
+oid sha256:7f723ba7955cfa64d70a96114a7286559b5eee1e6b188810c15a09c1dfe766a0
+size 550593184