End of training

README.md

@@ -24,18 +24,18 @@ base_model_config: unsloth/Llama-3.2-3B-Instruct
 bf16: auto
 chat_template: llama3
 dataset_prepared_path: null
-dataset_type: instruct
 datasets:
-- ds_type: json
+- data_files:
+  - ds_example.json
+  ds_type: json
   path: data/ds_example.json
-  type: alpaca
+  type: instruct
 debug: null
 deepspeed: null
 early_stopping_patience: null
 eval_max_new_tokens: 128
 eval_table_size: null
 evals_per_epoch: 4
-file_format: json
 flash_attention: true
 fp16: null
 fsdp: null
@@ -59,7 +59,7 @@ lora_model_dir: null
 lora_r: 32
 lora_target_linear: true
 lr_scheduler: cosine
-max_steps: 10
+max_steps: 5
 micro_batch_size: 2
 mlflow_experiment_name: miner_id_24
 mlflow_tracking_uri: http://94.156.8.49:5000
@@ -94,7 +94,7 @@ xformers_attention: null
 
 This model is a fine-tuned version of [unsloth/Llama-3.2-3B-Instruct](https://huggingface.co/unsloth/Llama-3.2-3B-Instruct) on the None dataset.
 It achieves the following results on the evaluation set:
-- Loss: 5.0689
+- Loss: 11.2689
 
 ## Model description
 
@@ -128,8 +128,8 @@ The following hyperparameters were used during training:
 
 | Training Loss | Epoch | Step | Validation Loss |
 |:-------------:|:-----:|:----:|:---------------:|
-| 5.4988        | 0.8   | 1    | 5.0751          |
-| 5.2725        | 1.6   | 2    | 5.0689          |
+| 13.3042       | 0.8   | 1    | 11.5243         |
+| 13.1934       | 1.6   | 2    | 11.2689         |
 
 
 ### Framework versions

adapter_model.bin

@@ -1,3 +1,3 @@
 version https://git-lfs.github.com/spec/v1
-oid sha256:f63cc530a8bbed71ec5c9e53b4a02a21d00fd2376226c3f48343ec58132cbbe5
+oid sha256:bb4842d4922c5f75424db453ffc16bec1e112d1aed9337c4f5a87df1c18fab6e
 size 982663982