Model save

README.md

@@ -16,7 +16,7 @@ should probably proofread and complete it, then remove this comment. -->
 
 This model is a fine-tuned version of [meta-llama/Llama-3.2-11B-Vision](https://huggingface.co/meta-llama/Llama-3.2-11B-Vision) on the None dataset.
 It achieves the following results on the evaluation set:
-- Loss: 1.4192
+- Loss: 1.3433
 
 ## Model description
 
@@ -48,17 +48,21 @@ The following hyperparameters were used during training:
 
 | Training Loss | Epoch | Step | Validation Loss |
 |:-------------:|:-----:|:----:|:---------------:|
-| 0.2326        | 0.1   | 5    | 1.4239          |
-| 1.0087        | 0.2   | 10   | 1.5654          |
-| 0.5429        | 0.3   | 15   | 1.5371          |
-| 0.7429        | 0.4   | 20   | 1.4577          |
-| 0.7219        | 0.5   | 25   | 1.4189          |
-| 0.5514        | 0.6   | 30   | 1.4007          |
-| 0.6459        | 0.7   | 35   | 1.4267          |
-| 0.6111        | 0.8   | 40   | 1.4510          |
-| 0.9172        | 0.9   | 45   | 1.4461          |
-| 0.6547        | 1.0   | 50   | 1.4141          |
-| 0.3312        | 1.1   | 55   | 1.4192          |
+| 1.8654        | 0.1   | 5    | 1.6057          |
+| 1.6048        | 0.2   | 10   | 1.5329          |
+| 2.0881        | 0.3   | 15   | 1.4732          |
+| 1.3989        | 0.4   | 20   | 1.4411          |
+| 1.4862        | 0.5   | 25   | 1.4208          |
+| 1.2947        | 0.6   | 30   | 1.4054          |
+| 1.6434        | 0.7   | 35   | 1.3892          |
+| 1.3301        | 0.8   | 40   | 1.3763          |
+| 1.417         | 0.9   | 45   | 1.3651          |
+| 1.4909        | 1.0   | 50   | 1.3581          |
+| 1.0997        | 1.1   | 55   | 1.3538          |
+| 1.3593        | 1.2   | 60   | 1.3560          |
+| 1.3547        | 1.3   | 65   | 1.3510          |
+| 1.0256        | 1.4   | 70   | 1.3471          |
+| 1.1541        | 1.5   | 75   | 1.3433          |
 
 
 ### Framework versions

adapter_config.json

@@ -27,13 +27,13 @@
   "rank_pattern": {},
   "revision": null,
   "target_modules": [
+    "up_proj",
     "q_proj",
+    "k_proj",
+    "v_proj",
     "o_proj",
     "down_proj",
-    "gate_proj",
-    "v_proj",
-    "k_proj",
-    "up_proj"
+    "gate_proj"
   ],
   "task_type": null,
   "trainable_token_indices": null,

adapter_model.safetensors

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:09c6748f9b2a03de32bad6b1df404145a88b3e25e07e4ff3c87bff022015a5bd
+size 125866776

training_args.bin

@@ -1,3 +1,3 @@
 version https://git-lfs.github.com/spec/v1
-oid sha256:d830417e5e88dea6b2618be599512d12c04898860228dbebf9eadeb54696769a
+oid sha256:442c111c70c8f9509bb524116f511c5aac5579911da2cba3480f0b635a4719b6
 size 5507