End of training

README.md

@@ -47,7 +47,7 @@ flash_attention: true
 fp16: null
 fsdp: null
 fsdp_config: null
-gradient_accumulation_steps: 16
+gradient_accumulation_steps: 4
 gradient_checkpointing: false
 group_by_length: false
 hub_model_id: tuanna08go/9d3d6c1c-01e3-47b5-8f25-e2bdb653d048
@@ -58,7 +58,7 @@ learning_rate: 0.0001
 load_in_4bit: false
 load_in_8bit: false
 local_rank: null
-logging_steps: 10
+logging_steps: 5
 lora_alpha: 16
 lora_dropout: 0.05
 lora_fan_in_fan_out: null
@@ -66,8 +66,8 @@ lora_model_dir: null
 lora_r: 8
 lora_target_linear: true
 lr_scheduler: cosine
-max_steps: 50
-micro_batch_size: 8
+max_steps: 1
+micro_batch_size: 2
 mlflow_experiment_name: /tmp/511bc7520306da8a_train_data.json
 model_type: AutoModelForCausalLM
 num_epochs: 1
@@ -91,7 +91,7 @@ wandb_name: 9d3d6c1c-01e3-47b5-8f25-e2bdb653d048
 wandb_project: Gradients-On-Demand
 wandb_run: your_name
 wandb_runid: 9d3d6c1c-01e3-47b5-8f25-e2bdb653d048
-warmup_steps: 2
+warmup_steps: 1
 weight_decay: 0.0
 xformers_attention: null
 
@@ -102,8 +102,6 @@ xformers_attention: null
 # 9d3d6c1c-01e3-47b5-8f25-e2bdb653d048
 
 This model is a fine-tuned version of [TinyLlama/TinyLlama-1.1B-Chat-v0.6](https://huggingface.co/TinyLlama/TinyLlama-1.1B-Chat-v0.6) on the None dataset.
-It achieves the following results on the evaluation set:
-- Loss: 0.5400
 
 ## Model description
 
@@ -123,26 +121,21 @@ More information needed
 
 The following hyperparameters were used during training:
 - learning_rate: 0.0001
-- train_batch_size: 8
-- eval_batch_size: 8
+- train_batch_size: 2
+- eval_batch_size: 2
 - seed: 42
-- gradient_accumulation_steps: 16
-- total_train_batch_size: 128
+- gradient_accumulation_steps: 4
+- 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: 2
-- training_steps: 50
+- training_steps: 1
 
 ### Training results
 
 | Training Loss | Epoch  | Step | Validation Loss |
 |:-------------:|:------:|:----:|:---------------:|
-| No log        | 0.0051 | 1    | 0.6309          |
-| 0.6061        | 0.0511 | 10   | 0.6027          |
-| 0.5841        | 0.1021 | 20   | 0.5628          |
-| 0.5459        | 0.1532 | 30   | 0.5474          |
-| 0.5444        | 0.2042 | 40   | 0.5410          |
-| 0.5309        | 0.2553 | 50   | 0.5400          |
+| No log        | 0.0003 | 1    | 0.6470          |
 
 
 ### Framework versions

adapter_model.bin

@@ -1,3 +1,3 @@
 version https://git-lfs.github.com/spec/v1
-oid sha256:ad33962941d36aca7856cc2eaea138b62c367eff8fe3b2e5eb227d967e1eb70e
+oid sha256:7cba69cde24d4551d6b26b2ac2b5f35952116c01e2f2a78747faf149932837a0
 size 25342042