End of training

README.md

@@ -6,7 +6,7 @@ tags:
 - axolotl
 - generated_from_trainer
 model-index:
-- name: b1494398-57b9-4061-b394-da82c151d1ae
+- name: 73bda874-d4bf-4b2d-bdc6-18b3a9a8e0c4
   results: []
 ---
 
@@ -19,15 +19,15 @@ should probably proofread and complete it, then remove this comment. -->
 axolotl version: `0.4.1`
 ```yaml
 adapter: lora
-base_model: Qwen/Qwen2.5-14B-Instruct
+base_model: Qwen/Qwen2.5-7B-Instruct
 bf16: auto
 dataset_prepared_path: null
 datasets:
 - data_files:
-  - 8f359061ef840864_train_data.json
+  - d185db34f9ed4537_train_data.json
   ds_type: json
   format: custom
-  path: /root/G.O.D-test/core/data/8f359061ef840864_train_data.json
+  path: /root/G.O.D-test/core/data/d185db34f9ed4537_train_data.json
   type:
     field_input: tools
     field_instruction: func_name
@@ -49,7 +49,7 @@ fsdp_config: null
 gradient_accumulation_steps: 4
 gradient_checkpointing: false
 group_by_length: false
-hub_model_id: souging/b1494398-57b9-4061-b394-da82c151d1ae
+hub_model_id: souging/73bda874-d4bf-4b2d-bdc6-18b3a9a8e0c4
 hub_repo: null
 hub_strategy: checkpoint
 hub_token: null
@@ -58,16 +58,16 @@ load_in_4bit: false
 load_in_8bit: false
 local_rank: null
 logging_steps: 1
-lora_alpha: 64
+lora_alpha: 48
 lora_dropout: 0.05
 lora_fan_in_fan_out: null
 lora_model_dir: null
-lora_r: 32
+lora_r: 24
 lora_target_linear: true
 lr_scheduler: cosine
-max_steps: 100
-micro_batch_size: 1
-mlflow_experiment_name: /tmp/8f359061ef840864_train_data.json
+max_steps: 80
+micro_batch_size: 2
+mlflow_experiment_name: /tmp/d185db34f9ed4537_train_data.json
 model_type: AutoModelForCausalLM
 num_epochs: 4
 optimizer: adamw_bnb_8bit
@@ -86,10 +86,10 @@ trust_remote_code: true
 val_set_size: 0.05
 wandb_entity: null
 wandb_mode: online
-wandb_name: 1d376839-b404-43d5-9ab3-103297cfbe18
+wandb_name: 72ac60aa-9b01-44e4-9ac9-0307450f38fa
 wandb_project: Gradients-On-Demand
 wandb_run: your_name
-wandb_runid: 1d376839-b404-43d5-9ab3-103297cfbe18
+wandb_runid: 72ac60aa-9b01-44e4-9ac9-0307450f38fa
 warmup_steps: 100
 weight_decay: 0.01
 xformers_attention: null
@@ -98,11 +98,11 @@ xformers_attention: null
 
 </details><br>
 
-# b1494398-57b9-4061-b394-da82c151d1ae
+# 73bda874-d4bf-4b2d-bdc6-18b3a9a8e0c4
 
-This model is a fine-tuned version of [Qwen/Qwen2.5-14B-Instruct](https://huggingface.co/Qwen/Qwen2.5-14B-Instruct) on the None dataset.
+This model is a fine-tuned version of [Qwen/Qwen2.5-7B-Instruct](https://huggingface.co/Qwen/Qwen2.5-7B-Instruct) on the None dataset.
 It achieves the following results on the evaluation set:
-- Loss: 0.0005
+- Loss: 0.0009
 
 ## Model description
 
@@ -122,24 +122,24 @@ More information needed
 
 The following hyperparameters were used during training:
 - learning_rate: 0.0002
-- train_batch_size: 1
-- eval_batch_size: 1
+- train_batch_size: 2
+- eval_batch_size: 2
 - seed: 42
 - distributed_type: multi-GPU
 - num_devices: 8
 - gradient_accumulation_steps: 4
-- total_train_batch_size: 32
-- total_eval_batch_size: 8
+- total_train_batch_size: 64
+- total_eval_batch_size: 16
 - 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: 100
-- training_steps: 100
+- training_steps: 80
 
 ### Training results
 
 | Training Loss | Epoch  | Step | Validation Loss |
 |:-------------:|:------:|:----:|:---------------:|
-| 0.0           | 0.0837 | 100  | 0.0005          |
+| 0.0108        | 0.1339 | 80   | 0.0009          |
 
 
 ### Framework versions

adapter_model.bin

@@ -1,3 +1,3 @@
 version https://git-lfs.github.com/spec/v1
-oid sha256:62041656417109919d46c4854b59ea30564eec442f094b308939325d9342f9e7
+oid sha256:27963a21504e3c4c8c888b17040eb0f16b7fbc72dde7d93560cc401d6c2d9aec
 size 242362666