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: 1.3441
+- Loss: 1.3393
 
 ## Model description
 
@@ -48,18 +48,28 @@ The following hyperparameters were used during training:
 
 | Training Loss | Epoch | Step | Validation Loss |
 |:-------------:|:-----:|:----:|:---------------:|
-| 1.0743        | 0.1   | 5    | 1.3437          |
-| 1.135         | 0.2   | 10   | 1.3584          |
-| 1.071         | 0.3   | 15   | 1.3599          |
-| 0.9413        | 0.4   | 20   | 1.3664          |
-| 1.0465        | 0.5   | 25   | 1.3513          |
-| 0.8599        | 0.6   | 30   | 1.3441          |
+| 1.873         | 0.1   | 5    | 1.6084          |
+| 1.6055        | 0.2   | 10   | 1.5360          |
+| 2.0931        | 0.3   | 15   | 1.4717          |
+| 1.3995        | 0.4   | 20   | 1.4390          |
+| 1.4799        | 0.5   | 25   | 1.4170          |
+| 1.2999        | 0.6   | 30   | 1.3999          |
+| 1.6356        | 0.7   | 35   | 1.3855          |
+| 1.3253        | 0.8   | 40   | 1.3713          |
+| 1.4132        | 0.9   | 45   | 1.3630          |
+| 1.4824        | 1.0   | 50   | 1.3557          |
+| 1.0958        | 1.1   | 55   | 1.3478          |
+| 1.3437        | 1.2   | 60   | 1.3490          |
+| 1.3517        | 1.3   | 65   | 1.3459          |
+| 1.0089        | 1.4   | 70   | 1.3415          |
+| 1.1556        | 1.5   | 75   | 1.3369          |
+| 1.3741        | 1.6   | 80   | 1.3393          |
 
 
 ### Framework versions
 
-- PEFT 0.15.1
-- Transformers 4.48.3
+- PEFT 0.10.0
+- Transformers 4.47.0
 - Pytorch 2.7.0a0+7c8ec84dab.nv25.03
 - Datasets 3.5.0
 - Tokenizers 0.21.1

adapter_config.json

@@ -6,9 +6,6 @@
   },
   "base_model_name_or_path": "meta-llama/Llama-3.2-11B-Vision",
   "bias": "none",
-  "corda_config": null,
-  "eva_config": null,
-  "exclude_modules": null,
   "fan_in_fan_out": false,
   "inference_mode": true,
   "init_lora_weights": "gaussian",
@@ -17,7 +14,6 @@
   "layers_to_transform": null,
   "loftq_config": {},
   "lora_alpha": 16,
-  "lora_bias": false,
   "lora_dropout": 0.1,
   "megatron_config": null,
   "megatron_core": "megatron.core",
@@ -27,20 +23,19 @@
   "rank_pattern": {},
   "revision": null,
   "target_modules": [
-    "mlp.gate_proj",
+    "cross_attn.o_proj",
     "mlp.up_proj",
     "self_attn.k_proj",
-    "self_attn.q_proj",
     "cross_attn.q_proj",
-    "self_attn.v_proj",
+    "self_attn.o_proj",
     "mlp.down_proj",
-    "cross_attn.v_proj",
-    "cross_attn.o_proj",
+    "self_attn.q_proj",
+    "self_attn.v_proj",
     "cross_attn.k_proj",
-    "self_attn.o_proj"
+    "cross_attn.v_proj",
+    "mlp.gate_proj"
   ],
   "task_type": null,
-  "trainable_token_indices": null,
   "use_dora": true,
   "use_rslora": false
 }

adapter_model.bin

@@ -1,3 +1,3 @@
 version https://git-lfs.github.com/spec/v1
-oid sha256:dbee23a185f12ba2d34645d89c5c61cce2dc223dbf9b2322661dde65d09de3e9
-size 126165813
+oid sha256:8c1e8333039ac6a46a8a0ebb67e019666d2cbec5332287dc5762948d4bff362e
+size 63333173