Model save

README.md

@@ -13,6 +13,8 @@ should probably proofread and complete it, then remove this comment. -->
 # NeoLLM
 
 This model is a fine-tuned version of [](https://huggingface.co/) on an unknown dataset.
+It achieves the following results on the evaluation set:
+- Loss: 3.6730
 
 ## Model description
 
@@ -42,6 +44,20 @@ The following hyperparameters were used during training:
 
 ### Training results
 
+| Training Loss | Epoch | Step  | Validation Loss |
+|:-------------:|:-----:|:-----:|:---------------:|
+| 4.0108        | 0.08  | 5000  | 3.9854          |
+| 3.8815        | 0.16  | 10000 | 3.8573          |
+| 3.8252        | 0.24  | 15000 | 3.8052          |
+| 3.829         | 0.32  | 20000 | 3.7989          |
+| 3.8178        | 0.4   | 25000 | 3.7986          |
+| 3.7996        | 0.48  | 30000 | 3.7762          |
+| 3.7922        | 0.56  | 35000 | 3.7717          |
+| 3.7963        | 0.64  | 40000 | 3.7611          |
+| 3.7796        | 0.72  | 45000 | 3.7490          |
+| 3.784         | 0.8   | 50000 | 3.7419          |
+| 3.7602        | 0.88  | 55000 | 3.7324          |
+| 3.7128        | 0.96  | 60000 | 3.6730          |
 
 
 ### Framework versions

config.json

@@ -47,7 +47,6 @@
   "rms_norm_eps": 1e-06,
   "rope_scaling": null,
   "rope_theta": 10000.0,
-  "specific_dropout_rate": 0.05,
   "transformers_version": "4.57.0.dev0",
   "vocab_size": 151665
 }

model.safetensors

@@ -1,3 +1,3 @@
 version https://git-lfs.github.com/spec/v1
-oid sha256:02449f4ed1467742a7961a07c0582a1388573a0c883850b70bb80f4e19877a47
+oid sha256:0b97f0384852a23a0f8bd7ecf0e18bb324fe5e94ee5ab1b2fa9f7e253c04a994
 size 250512472

training_args.bin

@@ -1,3 +1,3 @@
 version https://git-lfs.github.com/spec/v1
-oid sha256:0522da83b92155eb5f7c9e53c8974c559dd3045c057f1f5643682a983ac760be
+oid sha256:be3cb2c72ecf0ac7ca6c18a8f9e1b649c26b3527ac1cbc370db5a4bf15d6a803
 size 5585