Model save

README.md

@@ -17,15 +17,15 @@ should probably proofread and complete it, then remove this comment. -->
 
 This model is a fine-tuned version of [seanmemery/MLP-FinLLM-7b-it](https://huggingface.co/seanmemery/MLP-FinLLM-7b-it) on the None dataset.
 It achieves the following results on the evaluation set:
-- Loss: 0.8668
-- Rewards/chosen: -1.0781
-- Rewards/rejected: -1.7109
-- Rewards/accuracies: 0.3306
-- Rewards/margins: 0.6289
-- Logps/rejected: -24.375
-- Logps/chosen: -31.375
-- Logits/rejected: -5.9062
-- Logits/chosen: -5.9375
+- Loss: 0.7571
+- Rewards/chosen: -1.4766
+- Rewards/rejected: -2.5
+- Rewards/accuracies: 0.4050
+- Rewards/margins: 1.0234
+- Logps/rejected: -27.125
+- Logps/chosen: -33.0
+- Logits/rejected: -5.9688
+- Logits/chosen: -6.0
 
 ## Model description
 
@@ -49,15 +49,14 @@ The following hyperparameters were used during training:
 - eval_batch_size: 1
 - seed: 42
 - optimizer: Adam with betas=(0.9,0.999) and epsilon=1e-08
-- lr_scheduler_type: cosine
-- num_epochs: 2
+- lr_scheduler_type: constant
+- num_epochs: 1
 
 ### Training results
 
 | Training Loss | Epoch | Step | Validation Loss | Rewards/chosen | Rewards/rejected | Rewards/accuracies | Rewards/margins | Logps/rejected | Logps/chosen | Logits/rejected | Logits/chosen |
 |:-------------:|:-----:|:----:|:---------------:|:--------------:|:----------------:|:------------------:|:---------------:|:--------------:|:------------:|:---------------:|:-------------:|
-| 1.4609        | 0.69  | 50   | 0.9233          | -2.0156        | -3.2344          | 0.3967             | 1.2188          | -26.375        | -32.5        | -5.9062         | -5.9375       |
-| 0.6914        | 1.39  | 100  | 0.8668          | -1.0781        | -1.7109          | 0.3306             | 0.6289          | -24.375        | -31.375      | -5.9062         | -5.9375       |
+| 1.0625        | 0.69  | 50   | 0.7571          | -1.4766        | -2.5             | 0.4050             | 1.0234          | -27.125        | -33.0        | -5.9688         | -6.0          |
 
 
 ### Framework versions

model-00001-of-00003.safetensors

@@ -1,3 +1,3 @@
 version https://git-lfs.github.com/spec/v1
-oid sha256:1760e028de8723b49a7963a8519b285247b411430355cb960ad44b85bfe8873a
+oid sha256:02e35d706a18f11cc48392745e5998e6c7fbfd788a43a3b79ce50b0a3f71f9be
 size 4938985352

model-00002-of-00003.safetensors

@@ -1,3 +1,3 @@
 version https://git-lfs.github.com/spec/v1
-oid sha256:f2b389ad2d4baaa5c2698e2509ef9c8c62a237ee38d2396ed388f26fca0db017
+oid sha256:58c0dcbc02514c639636ee9b1abe47f820c47ceb493348d491652e99b9ec1590
 size 4947390880

model-00003-of-00003.safetensors

@@ -1,3 +1,3 @@
 version https://git-lfs.github.com/spec/v1
-oid sha256:ec0f27ffa94147ee99f2744f8263bdd290dcc268c0e28373044d052c5f65974f
+oid sha256:89f326ba82ee94182a33addd7f5e6bf0c8ff14c3a2f48b15ee30b45d79edab0f
 size 3590488816

training_args.bin

@@ -1,3 +1,3 @@
 version https://git-lfs.github.com/spec/v1
-oid sha256:c3d46ff920a432f36d282b88331fd116f253f6a2dfae08faa6be636f76f1ccac
+oid sha256:eff9f548125fa959e5be9ebf7600bcd2f94c5dda451be5df73bda9ad16180e4f
 size 4984