End of training

README.md

@@ -65,7 +65,7 @@ lora_model_dir: null
 lora_r: 8
 lora_target_linear: true
 lr_scheduler: cosine
-max_steps: 1
+max_steps: 50
 micro_batch_size: 2
 mlflow_experiment_name: /tmp/ec0b90fdc2afbfeb_train_data.json
 model_type: AutoModelForCausalLM
@@ -92,7 +92,7 @@ wandb_name: c3a8521c-ca19-4205-a4e5-ce6d8bd32662
 wandb_project: Gradients-On-Demand
 wandb_run: your_name
 wandb_runid: c3a8521c-ca19-4205-a4e5-ce6d8bd32662
-warmup_steps: 1
+warmup_steps: 10
 weight_decay: 0.0
 xformers_attention: null
 
@@ -103,6 +103,8 @@ xformers_attention: null
 # 669c5dda-014f-3f6e-d206-040178c15ce8
 
 This model is a fine-tuned version of [EleutherAI/pythia-70m-deduped](https://huggingface.co/EleutherAI/pythia-70m-deduped) on the None dataset.
+It achieves the following results on the evaluation set:
+- Loss: 20.0296
 
 ## Model description
 
@@ -129,14 +131,19 @@ The following hyperparameters were used during training:
 - 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: 1
+- lr_scheduler_warmup_steps: 10
+- training_steps: 50
 
 ### Training results
 
 | Training Loss | Epoch  | Step | Validation Loss |
 |:-------------:|:------:|:----:|:---------------:|
 | No log        | 0.0003 | 1    | 20.3402         |
+| 82.5967       | 0.0026 | 10   | 20.3063         |
+| 93.181        | 0.0051 | 20   | 20.2140         |
+| 79.6657       | 0.0077 | 30   | 20.0940         |
+| 86.0579       | 0.0102 | 40   | 20.0395         |
+| 69.3403       | 0.0128 | 50   | 20.0296         |
 
 
 ### Framework versions

adapter_model.bin

@@ -1,3 +1,3 @@
 version https://git-lfs.github.com/spec/v1
-oid sha256:9fbb8407e4c51867c449929a1bf8c103dd6de5c9422e8097abdc4685e53c78c4
+oid sha256:d38f8719bcba281d52fbfb26838951f6ce2a6f3e229f6144dc7ec40a58678742
 size 1590526