End of training

README.md

@@ -3,9 +3,10 @@ library_name: peft
 license: apache-2.0
 base_model: Qwen/Qwen2.5-7B-Instruct
 tags:
+- axolotl
 - generated_from_trainer
 model-index:
-- name: miner_id_24
+- name: e6ccd50b-c55b-400e-bcc7-bf043937b944
   results: []
 ---
 
@@ -40,10 +41,10 @@ datasets:
     system_prompt: ''
 debug: null
 deepspeed: null
-early_stopping_patience: 
+early_stopping_patience: 3
 early_stopping_threshold: 0.0001
 eval_max_new_tokens: 128
-eval_steps: 
+eval_steps: 120
 eval_strategy: null
 flash_attention: true
 fp16: null
@@ -52,7 +53,7 @@ fsdp_config: null
 gradient_accumulation_steps: 2
 gradient_checkpointing: false
 group_by_length: false
-hub_model_id: 
+hub_model_id: mrferr3t/e6ccd50b-c55b-400e-bcc7-bf043937b944
 hub_repo: null
 hub_strategy: checkpoint
 hub_token: null
@@ -60,7 +61,7 @@ learning_rate: 0.0004
 load_in_4bit: false
 load_in_8bit: false
 local_rank: null
-logging_steps: 
+logging_steps: 120
 lora_alpha: 16
 lora_dropout: 0.05
 lora_fan_in_fan_out: null
@@ -68,7 +69,7 @@ lora_model_dir: null
 lora_r: 8
 lora_target_linear: true
 lr_scheduler: cosine
-max_steps: 1
+max_steps: 
 micro_batch_size: 32
 mlflow_experiment_name: /tmp/0258260fff514215_train_data.json
 model_type: AutoModelForCausalLM
@@ -79,7 +80,7 @@ pad_to_sequence_len: true
 resume_from_checkpoint: /workspace/hub_repo/last-checkpoint
 s2_attention: null
 sample_packing: false
-save_steps: 
+save_steps: 120
 saves_per_epoch: 0
 sequence_len: 512
 strict: false
@@ -87,9 +88,9 @@ tf32: false
 tokenizer_type: AutoTokenizer
 train_on_inputs: false
 trust_remote_code: true
-val_set_size: 0.0
+val_set_size: 0.05
 wandb_entity: null
-wandb_mode: disabled
+wandb_mode: 
 wandb_name: 7ce57913-98f3-40da-a864-33f8df60155d
 wandb_project: Gradients-On-Demand
 wandb_run: your_name
@@ -102,11 +103,11 @@ xformers_attention: null
 
 </details><br>
 
-# miner_id_24
+# e6ccd50b-c55b-400e-bcc7-bf043937b944
 
 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: 0.1909
+- Loss: 0.2069
 
 ## Model description
 
@@ -134,7 +135,7 @@ The following hyperparameters were used during training:
 - 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: 1
+- num_epochs: 100
 
 ### Training results
 
@@ -148,6 +149,9 @@ The following hyperparameters were used during training:
 | 0.1677        | 0.5503 | 260  | 0.2076          |
 | 0.1963        | 0.6603 | 312  | 0.2016          |
 | 0.192         | 0.7704 | 364  | 0.1909          |
+| 0.1694        | 0.8804 | 416  | 0.2012          |
+| 0.1806        | 0.9905 | 468  | 0.1981          |
+| 0.1239        | 1.1005 | 520  | 0.2069          |
 
 
 ### Framework versions

adapter_model.bin

@@ -1,3 +1,3 @@
 version https://git-lfs.github.com/spec/v1
-oid sha256:56892c2e428e2dd943f947b12324750662f045fbdd73ae86cb320c6eb5b55993
+oid sha256:642e30f44f2b8e08a04736eac1344d4177cb85e1b55018234ddb7cf8924c0d91
 size 80881450