End of training

README.md

@@ -43,13 +43,13 @@ deepspeed: null
 device_map: auto
 do_eval: true
 num_epochs: 3
-load_best_model_at_end: true
+# load_best_model_at_end: true
 # early_stopping_patience: 1
 eval_batch_size: 4
 eval_max_new_tokens: 128
-# eval_steps: 200
+
 eval_table_size: null
-evals_per_epoch: true
+evals_per_epoch: null
 flash_attention: true
 fp16: false
 fsdp: null
@@ -121,7 +121,7 @@ xformers_attention: null
 
 This model is a fine-tuned version of [unsloth/SmolLM-360M](https://huggingface.co/unsloth/SmolLM-360M) on the None dataset.
 It achieves the following results on the evaluation set:
-- Loss: 0.1570
+- Loss: 0.1265
 
 ## Model description
 
@@ -155,10 +155,8 @@ The following hyperparameters were used during training:
 
 | Training Loss | Epoch  | Step | Validation Loss |
 |:-------------:|:------:|:----:|:---------------:|
-| No log        | 0.0027 | 1    | 3.2056          |
-| 0.5417        | 0.5450 | 200  | 0.5135          |
-| 0.2789        | 1.0899 | 400  | 0.2482          |
-| 0.1355        | 1.6349 | 600  | 0.1570          |
+| 0.3036        | 1.0    | 367  | 0.2539          |
+| 0.1304        | 1.6349 | 600  | 0.1265          |
 
 
 ### Framework versions

adapter_model.bin

@@ -1,3 +1,3 @@
 version https://git-lfs.github.com/spec/v1
-oid sha256:f4aea9b9dfa07a360175051f874c5d0f8ca67a668f84871c4df0aa0aff00872a
+oid sha256:4ae09074903f80cf3f473a3ee6c121cbab981d41ed76db72fd3bc46c9784764d
 size 69629450