End of training

README.md

@@ -54,7 +54,7 @@ hub_model_id: nblinh/d80920cd-4e4c-4cb6-a037-cf7ca55e4ee2
 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/936e6ba20e1ad52a_train_data.json
 model_type: AutoModelForCausalLM
@@ -103,7 +103,7 @@ xformers_attention: true
 
 This model is a fine-tuned version of [unsloth/gemma-2-9b-it](https://huggingface.co/unsloth/gemma-2-9b-it) on the None dataset.
 It achieves the following results on the evaluation set:
-- Loss: 0.4479
+- Loss: 0.5153
 
 ## Model description
 
@@ -122,20 +122,20 @@ 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 |
 |:-------------:|:------:|:----:|:---------------:|
-| 0.5182        | 0.0027 | 100  | 0.4479          |
+| 0.4012        | 0.0014 | 50   | 0.5153          |
 
 
 ### Framework versions

adapter_config.json

@@ -22,11 +22,11 @@
   "target_modules": [
     "q_proj",
     "gate_proj",
-    "v_proj",
     "k_proj",
+    "v_proj",
     "up_proj",
-    "down_proj",
-    "o_proj"
+    "o_proj",
+    "down_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:c037215e8a833423c6912ada9478f8dbe0bebecab64364bc1261043bd6cfa70b
+oid sha256:43cc28155245c61ca3a73977dab9aa4dd92edfff58c27dfc7c5521349b0c2e36
 size 216284090

adapter_model.safetensors

@@ -1,3 +1,3 @@
 version https://git-lfs.github.com/spec/v1
-oid sha256:ca98c6429ae3c9a7e613a245835d230762e5ec6a6fd2f3be61b08bc2995a94d6
+oid sha256:19b3e3e14de22e98f8aaec47cbba421e35155396620cc051b59775fb7fb8b9a2
 size 216151256

training_args.bin

@@ -1,3 +1,3 @@
 version https://git-lfs.github.com/spec/v1
-oid sha256:dad1921e6de536571b848e582b720d63099b89e54e302d6c1603e24f786b9874
+oid sha256:ed29e80e01f9099b21aad851f24b8e2fc6b8e682c427937facc5d243ff1e5e0f
 size 6776