End of training

README.md

@@ -54,7 +54,7 @@ hub_model_id: nblinh/060f0e2e-cae3-4521-9339-56052a5dd075
 hub_repo: null
 hub_strategy: end
 hub_token: null
-learning_rate: 0.0003
+learning_rate: 0.0002
 load_in_4bit: true
 load_in_8bit: true
 local_rank: null
@@ -66,7 +66,7 @@ lora_model_dir: null
 lora_r: 16
 lora_target_linear: true
 lr_scheduler: cosine
-max_steps: 100
+max_steps: 50
 micro_batch_size: 1
 mlflow_experiment_name: /tmp/817b142d60816d15_train_data.json
 model_type: AutoModelForCausalLM
@@ -124,20 +124,20 @@ More information needed
 ### Training hyperparameters
 
 The following hyperparameters were used during training:
-- learning_rate: 0.0003
+- learning_rate: 0.0002
 - train_batch_size: 1
 - eval_batch_size: 1
 - seed: 42
 - optimizer: Use OptimizerNames.ADAMW_BNB with betas=(0.9,0.999) and epsilon=1e-08 and optimizer_args=No additional optimizer arguments
 - lr_scheduler_type: cosine
 - lr_scheduler_warmup_steps: 10
-- training_steps: 100
+- training_steps: 50
 
 ### Training results
 
 | Training Loss | Epoch  | Step | Validation Loss |
 |:-------------:|:------:|:----:|:---------------:|
-| 3.8603        | 0.0075 | 100  | nan             |
+| 2.0198        | 0.0037 | 50   | nan             |
 
 
 ### Framework versions

adapter_config.json

@@ -21,12 +21,12 @@
   "revision": null,
   "target_modules": [
     "down_proj",
-    "q_proj",
-    "k_proj",
     "v_proj",
+    "q_proj",
+    "o_proj",
     "gate_proj",
     "up_proj",
-    "o_proj"
+    "k_proj"
   ],
   "task_type": "CAUSAL_LM",
   "use_dora": false,

adapter_model.bin

@@ -1,3 +1,3 @@
 version https://git-lfs.github.com/spec/v1
-oid sha256:55a7a371923799dfb479f15db9bc9936b14153755d8d2599963fabc72ebfb5d8
+oid sha256:f07738aac9deb2af5ddd72bc1e4c3bd6dbccb1b4288d46db2b281b632c32ccab
 size 50573530

adapter_model.safetensors

@@ -1,3 +1,3 @@
 version https://git-lfs.github.com/spec/v1
-oid sha256:3b3c2cfcefc15179c7b188d8b6376372328fae2adef122b6637578ffe823d5c4
+oid sha256:05268f85a1de9d4a88013f613b51dd02eb2486c1570457f2bada9598621084ef
 size 50503544

training_args.bin

@@ -1,3 +1,3 @@
 version https://git-lfs.github.com/spec/v1
-oid sha256:6646ee2a3ac11ab9edb08efefa3fca6bc41f6253544852927f5315ad8ec79c8e
+oid sha256:005d026224e9adacece530a316fd13914a07530e0a48b021ab93434c5d493cd2
 size 6776