End of training

README.md

@@ -44,11 +44,11 @@ eval_max_new_tokens: 128
 eval_table_size: null
 evals_per_epoch: 4
 flash_attention: true
-fp16: true
+fp16: null
 fsdp: null
 fsdp_config: null
 gradient_accumulation_steps: 4
-gradient_checkpointing: true
+gradient_checkpointing: false
 group_by_length: false
 hub_model_id: dimasik87/de33f694-50e1-4884-8566-140c2e9ee8b8
 hub_repo: null
@@ -59,30 +59,30 @@ load_in_4bit: false
 load_in_8bit: false
 local_rank: null
 logging_steps: 1
-lora_alpha: 16
-lora_dropout: 0.1
+lora_alpha: 32
+lora_dropout: 0.05
 lora_fan_in_fan_out: null
 lora_model_dir: null
-lora_r: 8
+lora_r: 16
 lora_target_linear: true
 lr_scheduler: cosine
 max_memory:
   0: 70GiB
-max_steps: 25
+max_steps: 50
 micro_batch_size: 1
 mlflow_experiment_name: /tmp/823d37b8588de3b9_train_data.json
 model_type: AutoModelForCausalLM
-num_epochs: 3
+num_epochs: 4
 optimizer: adamw_torch
 output_dir: miner_id_24
 pad_to_sequence_len: true
 resume_from_checkpoint: null
 s2_attention: null
 sample_packing: false
-saves_per_epoch: 3
+saves_per_epoch: 4
 sequence_len: 2028
 strict: false
-tf32: false
+tf32: true
 tokenizer_type: AutoTokenizer
 train_on_inputs: false
 trust_remote_code: true
@@ -105,7 +105,7 @@ xformers_attention: null
 
 This model is a fine-tuned version of [NousResearch/Hermes-2-Pro-Llama-3-8B](https://huggingface.co/NousResearch/Hermes-2-Pro-Llama-3-8B) on the None dataset.
 It achieves the following results on the evaluation set:
-- Loss: 0.2460
+- Loss: 0.2001
 
 ## Model description
 
@@ -133,21 +133,25 @@ The following hyperparameters were used during training:
 - optimizer: Use OptimizerNames.ADAMW_TORCH 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: 25
+- training_steps: 50
 
 ### Training results
 
 | Training Loss | Epoch  | Step | Validation Loss |
 |:-------------:|:------:|:----:|:---------------:|
 | 0.9466        | 0.0001 | 1    | 1.1514          |
-| 0.9876        | 0.0003 | 3    | 1.1419          |
-| 1.311         | 0.0005 | 6    | 0.9818          |
-| 0.8303        | 0.0008 | 9    | 0.6860          |
-| 0.4477        | 0.0011 | 12   | 0.4301          |
-| 0.2958        | 0.0014 | 15   | 0.2870          |
-| 0.2894        | 0.0016 | 18   | 0.2600          |
-| 0.1838        | 0.0019 | 21   | 0.2497          |
-| 0.2865        | 0.0022 | 24   | 0.2460          |
+| 0.886         | 0.0004 | 4    | 1.0636          |
+| 0.6905        | 0.0007 | 8    | 0.6157          |
+| 0.3233        | 0.0011 | 12   | 0.2960          |
+| 0.1823        | 0.0015 | 16   | 0.2611          |
+| 0.1479        | 0.0018 | 20   | 0.2346          |
+| 0.2695        | 0.0022 | 24   | 0.2255          |
+| 0.1645        | 0.0025 | 28   | 0.2141          |
+| 0.1804        | 0.0029 | 32   | 0.2088          |
+| 0.2318        | 0.0033 | 36   | 0.2063          |
+| 0.1563        | 0.0036 | 40   | 0.2020          |
+| 0.1512        | 0.0040 | 44   | 0.2004          |
+| 0.1529        | 0.0044 | 48   | 0.2001          |
 
 
 ### Framework versions

adapter_model.bin

@@ -1,3 +1,3 @@
 version https://git-lfs.github.com/spec/v1
-oid sha256:bf89dd606ee553bc0bfa09d864ad3b6eb8a44bb971145c53c61722e26da412f3
-size 84047370
+oid sha256:cc348215c8f42868c0b42c19ff8cb067fb3846b0e10cf94d0cc54cdf760d4a64
+size 167934026