End of training

README.md

@@ -54,7 +54,7 @@ hub_model_id: nblinh/10b868e8-f1c8-4d9a-8998-c5b2f3d45ffb
 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/d1b69aefdc9411a5_train_data.json
 model_type: AutoModelForCausalLM
@@ -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 |
 |:-------------:|:------:|:----:|:---------------:|
-| 0.0           | 0.0001 | 100  | nan             |
+| 0.0           | 0.0001 | 50   | nan             |
 
 
 ### Framework versions

adapter_config.json

@@ -20,13 +20,13 @@
   "rank_pattern": {},
   "revision": null,
   "target_modules": [
-    "v_proj",
-    "gate_proj",
-    "o_proj",
-    "q_proj",
     "up_proj",
+    "q_proj",
+    "down_proj",
+    "o_proj",
     "k_proj",
-    "down_proj"
+    "v_proj",
+    "gate_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:bbabf323eaf3b5067c6afa011ba195b560252ee9cb3d857cc0ebbfb5f52d308a
+oid sha256:361a6ca304b84715d120fda4c9f3405a69da2629915d2f4ace36e4a3a89b8867
 size 34895178

adapter_model.safetensors

@@ -1,3 +1,3 @@
 version https://git-lfs.github.com/spec/v1
-oid sha256:ac5ee24e019f3a92d7a3fc0102d9912b13e9a1e0182ee3d4de63c6c154408315
+oid sha256:c77f52f6f669e9b208a303e264af27692b1506796679948e97d39270c66ef520
 size 34793120

training_args.bin

@@ -1,3 +1,3 @@
 version https://git-lfs.github.com/spec/v1
-oid sha256:23b9cff7f83b98f4f01416e220ba3d5e200be5ed4f9b6d3d1fe7960198cd72c4
+oid sha256:30697c45250140264a030d42c8af187071c6ad5289cb7c6a1994545c8ab70b97
 size 6776