End of training

README.md

@@ -64,7 +64,7 @@ lora_model_dir: null
 lora_r: 8
 lora_target_linear: true
 lr_scheduler: cosine
-max_steps: 50
+max_steps: 1
 micro_batch_size: 2
 mlflow_experiment_name: /tmp/b86be1d045906067_train_data.json
 model_type: AutoModelForCausalLM
@@ -89,7 +89,7 @@ wandb_name: 94a6cad8-c2cc-425e-aa67-ad52a965d356
 wandb_project: Gradients-On-Demand
 wandb_run: your_name
 wandb_runid: 94a6cad8-c2cc-425e-aa67-ad52a965d356
-warmup_steps: 10
+warmup_steps: 1
 weight_decay: 0.0
 xformers_attention: null
 
@@ -100,8 +100,6 @@ xformers_attention: null
 # 94a6cad8-c2cc-425e-aa67-ad52a965d356
 
 This model is a fine-tuned version of [jhflow/mistral7b-lora-multi-turn-v2](https://huggingface.co/jhflow/mistral7b-lora-multi-turn-v2) on the None dataset.
-It achieves the following results on the evaluation set:
-- Loss: 0.6012
 
 ## Model description
 
@@ -128,19 +126,14 @@ The following hyperparameters were used during training:
 - total_train_batch_size: 8
 - 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: 50
+- lr_scheduler_warmup_steps: 2
+- training_steps: 1
 
 ### Training results
 
 | Training Loss | Epoch  | Step | Validation Loss |
 |:-------------:|:------:|:----:|:---------------:|
 | No log        | 0.0039 | 1    | 0.8528          |
-| 3.1245        | 0.0390 | 10   | 0.6828          |
-| 2.4105        | 0.0780 | 20   | 0.6242          |
-| 2.5341        | 0.1170 | 30   | 0.6087          |
-| 2.3267        | 0.1559 | 40   | 0.6020          |
-| 2.543         | 0.1949 | 50   | 0.6012          |
 
 
 ### Framework versions

adapter_config.json

@@ -20,13 +20,13 @@
   "rank_pattern": {},
   "revision": null,
   "target_modules": [
-    "o_proj",
     "up_proj",
-    "gate_proj",
     "k_proj",
+    "gate_proj",
     "down_proj",
-    "v_proj",
-    "q_proj"
+    "q_proj",
+    "o_proj",
+    "v_proj"
   ],
   "task_type": "CAUSAL_LM",
   "use_dora": false,

adapter_model.bin

@@ -1,3 +1,3 @@
 version https://git-lfs.github.com/spec/v1
-oid sha256:84dfe489ca26c0721704592b90a0ae07474ff8d09daa76de681cf172b5def105
+oid sha256:31e9e4f307563c755eed62a8071792676961f5dffe12c442a98952b6394bc89a
 size 84047370

adapter_model.safetensors

@@ -1,3 +1,3 @@
 version https://git-lfs.github.com/spec/v1
-oid sha256:292eadaa61de5dd99e6bf9ec43a5dde3d7c676fce4bff127477199d42e858834
+oid sha256:6ef85cd79f7d6a1c7af7c3b5d2da59f4d707d0c6fd2f556b38c62de65f8ba9cf
 size 83945296

training_args.bin

@@ -1,3 +1,3 @@
 version https://git-lfs.github.com/spec/v1
-oid sha256:68f2d8d172f297e9444c68fa0d9a882ea3108657ab700be8fcd9ce3587988d09
+oid sha256:ede6b6fdee903a3074da3a171c93ab5343e1258adcee4206296f452294c79e95
 size 6776