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: 4.4273
-- Rewards/chosen: -22.75
-- Rewards/rejected: -22.125
-- Rewards/accuracies: 0.4959
-- Rewards/margins: -0.7070
-- Logps/rejected: -242.0
-- Logps/chosen: -255.0
-- Logits/rejected: -6.3125
-- Logits/chosen: -6.3125
+- Loss: 12.5069
+- Rewards/chosen: -209.0
+- Rewards/rejected: -213.0
+- Rewards/accuracies: 0.5207
+- Rewards/margins: 4.625
+- Logps/rejected: -448.0
+- Logps/chosen: -444.0
+- Logits/rejected: -8.5
+- Logits/chosen: -8.5
 
 ## Model description
 
@@ -56,16 +56,16 @@ The following hyperparameters were used during training:
 
 | Training Loss | Epoch | Step | Validation Loss | Rewards/chosen | Rewards/rejected | Rewards/accuracies | Rewards/margins | Logps/rejected | Logps/chosen | Logits/rejected | Logits/chosen |
 |:-------------:|:-----:|:----:|:---------------:|:--------------:|:----------------:|:------------------:|:---------------:|:--------------:|:------------:|:---------------:|:-------------:|
-| 3.9219        | 0.35  | 50   | 4.7722          | -21.25         | -20.5            | 0.4876             | -0.8359         | -226.0         | -239.0       | -5.0938         | -5.0938       |
-| 0.7031        | 0.7   | 100  | 4.5288          | -22.0          | -21.375          | 0.4959             | -0.6445         | -236.0         | -247.0       | -5.7188         | -5.7188       |
-| 3.3281        | 1.05  | 150  | 4.4110          | -22.875        | -22.25           | 0.4959             | -0.6289         | -244.0         | -256.0       | -5.9062         | -5.9062       |
-| 2.5469        | 1.4   | 200  | 4.4287          | -23.0          | -22.375          | 0.5041             | -0.6406         | -245.0         | -256.0       | -6.25           | -6.25         |
-| 9.25          | 1.75  | 250  | 4.4273          | -22.75         | -22.125          | 0.4959             | -0.7070         | -242.0         | -255.0       | -6.3125         | -6.3125       |
+| 16.875        | 0.35  | 50   | 22.8195         | -91.5          | -85.5            | 0.3967             | -6.1875         | -193.0         | -210.0       | -7.8438         | -7.8438       |
+| 2.2031        | 0.7   | 100  | 21.6111         | -101.0         | -96.5            | 0.4793             | -4.625          | -215.0         | -229.0       | -8.4375         | -8.4375       |
+| 16.125        | 1.05  | 150  | 21.3741         | -141.0         | -138.0           | 0.4215             | -3.25           | -298.0         | -310.0       | -9.0            | -9.0          |
+| 11.9375       | 1.4   | 200  | 23.0505         | -155.0         | -153.0           | 0.4711             | -1.2188         | -328.0         | -336.0       | -7.375          | -7.4062       |
+| 29.25         | 1.75  | 250  | 12.5069         | -209.0         | -213.0           | 0.5207             | 4.625           | -448.0         | -444.0       | -8.5            | -8.5          |
 
 
 ### Framework versions
 
 - Transformers 4.38.2
-- Pytorch 2.2.1
+- Pytorch 2.1.2+cu121
 - Datasets 2.18.0
 - Tokenizers 0.15.2

model-00001-of-00003.safetensors

@@ -1,3 +1,3 @@
 version https://git-lfs.github.com/spec/v1
-oid sha256:22f2fa4116f32ffc856bc0f201dd422458e85348f68aeb74838be6ad7eda4eeb
+oid sha256:13c4052551ec5bba0863fe3fff45afb4e98399593c104d658837ca5f05fdda24
 size 4938985352

model-00002-of-00003.safetensors

@@ -1,3 +1,3 @@
 version https://git-lfs.github.com/spec/v1
-oid sha256:ae563f863f14e45d71c175776471c0e51f386d503157e66e2174f976faa8b945
+oid sha256:d7c2511b0bcedfd754b8bea79bbc334cb5b434c3447372ca875e8984ff1088e1
 size 4947390880

model-00003-of-00003.safetensors

@@ -1,3 +1,3 @@
 version https://git-lfs.github.com/spec/v1
-oid sha256:e2372efc29c60b336032ea5d32b4aff90e28f469f39548560b8251800ee91f5a
+oid sha256:edb4b918f5508929a5ddf6509fe53325324551fc632dc2835040f6223b730d82
 size 3590488816

training_args.bin

@@ -1,3 +1,3 @@
 version https://git-lfs.github.com/spec/v1
-oid sha256:d8aee546a55aef76020ac1c41db26452f2fc349c8f93ccbf1177df0fc1832e16
+oid sha256:492c6aa5b19cd8f124170e6532df0c084e005e631cf63aa20ca652843bc06b96
 size 4984