AI-4-Health/HPP-FINETUNED-Meta-Llama-3-8B-Instruct

README.md

@@ -18,12 +18,7 @@ should probably proofread and complete it, then remove this comment. -->
 
 This model is a fine-tuned version of [meta-llama/Meta-Llama-3-8B-Instruct](https://huggingface.co/meta-llama/Meta-Llama-3-8B-Instruct) on an unknown dataset.
 It achieves the following results on the evaluation set:
-- eval_loss: 1.4540
-- eval_runtime: 9.5935
-- eval_samples_per_second: 11.153
-- eval_steps_per_second: 1.459
-- epoch: 1.6649
-- step: 200
+- Loss: 1.4403
 
 ## Model description
 
@@ -51,9 +46,30 @@ The following hyperparameters were used during training:
 - optimizer: Adam with betas=(0.9,0.999) and epsilon=1e-08
 - lr_scheduler_type: cosine
 - lr_scheduler_warmup_ratio: 0.05
-- num_epochs: 3
+- num_epochs: 20
 - mixed_precision_training: Native AMP
 
+### Training results
+
+| Training Loss | Epoch  | Step | Validation Loss |
+|:-------------:|:------:|:----:|:---------------:|
+| 1.8026        | 0.4162 | 50   | 1.7101          |
+| 1.5158        | 0.8325 | 100  | 1.5503          |
+| 1.437         | 1.2487 | 150  | 1.4967          |
+| 1.3731        | 1.6649 | 200  | 1.4697          |
+| 1.4355        | 2.0812 | 250  | 1.4455          |
+| 1.3037        | 2.4974 | 300  | 1.4299          |
+| 1.4995        | 2.9136 | 350  | 1.4190          |
+| 1.5689        | 3.3299 | 400  | 1.4199          |
+| 1.5008        | 3.7461 | 450  | 1.4122          |
+| 1.357         | 4.1623 | 500  | 1.4182          |
+| 1.3323        | 4.5786 | 550  | 1.4174          |
+| 1.1464        | 4.9948 | 600  | 1.4071          |
+| 1.3099        | 5.4110 | 650  | 1.4232          |
+| 1.2026        | 5.8273 | 700  | 1.4183          |
+| 1.2336        | 6.2435 | 750  | 1.4403          |
+
+
 ### Framework versions
 
 - PEFT 0.11.1

adapter_model.safetensors

@@ -1,3 +1,3 @@
 version https://git-lfs.github.com/spec/v1
-oid sha256:6501ee19efb13db3066b3ca3111726396e9848de0f3f2bd951c6365b9f3ee49f
+oid sha256:c3d1e12a70fd36173ced19ee8461d02a1be7de1c5b43fdf0c7ec3594fc00da61
 size 27280152

runs/Jun14_00-49-10_jupiter/events.out.tfevents.1718297363.jupiter.387952.0

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:2fa1888cf5751f99b7d91213dec6bceae1c5c56c1b13c30a5905fcd8ee60fc13
+size 5637

runs/Jun14_00-52-54_jupiter/events.out.tfevents.1718297587.jupiter.387952.1

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:25ba077ac170b1f248d8104bff8e619eb5b485ffdf76ec44965f4f0fab2f8bab
+size 167788

tokenizer.json

@@ -1,11 +1,6 @@
 {
   "version": "1.0",
-  "truncation": {
-    "direction": "Right",
-    "max_length": 512,
-    "strategy": "LongestFirst",
-    "stride": 0
-  },
+  "truncation": null,
   "padding": null,
   "added_tokens": [
     {

training_args.bin

@@ -1,3 +1,3 @@
 version https://git-lfs.github.com/spec/v1
-oid sha256:02e896f485718fcc80ba95f42b802371c912c071b6c001915e5ead5020bf9fe6
-size 4923
+oid sha256:51c6f0012d8dbcf6c404cdd940bb60a5816ad0c5830ce0b969d6ae893eb88541
+size 4859