End of training

README.md

@@ -12,11 +12,12 @@ model-index:
 should probably proofread and complete it, then remove this comment. -->
 
 [<img src="https://raw.githubusercontent.com/wandb/assets/main/wandb-github-badge-28.svg" alt="Visualize in Weights & Biases" width="200" height="32"/>](https://wandb.ai/emnamaghrebi-epfl/idefics2-8B-ft-dataset/runs/i6dz9zuf)
+[<img src="https://raw.githubusercontent.com/wandb/assets/main/wandb-github-badge-28.svg" alt="Visualize in Weights & Biases" width="200" height="32"/>](https://wandb.ai/emnamaghrebi-epfl/idefics2-8B-ft-dataset/runs/sciuimwo)
 # idefics2-8b-manuals-ft-v4
 
 This model is a fine-tuned version of [HuggingFaceM4/idefics2-8b](https://huggingface.co/HuggingFaceM4/idefics2-8b) on an unknown dataset.
 It achieves the following results on the evaluation set:
-- Loss: 0.1312
+- Loss: 5.7217
 
 ## Model description
 
@@ -51,9 +52,9 @@ The following hyperparameters were used during training:
 
 | Training Loss | Epoch  | Step | Validation Loss |
 |:-------------:|:------:|:----:|:---------------:|
-| 8.0773        | 1.1299 | 25   | 0.6877          |
-| 0.3475        | 2.2599 | 50   | 0.1887          |
-| 0.1647        | 3.3898 | 75   | 0.1312          |
+| 15.1103       | 1.1299 | 25   | 14.9539         |
+| 13.8953       | 2.2599 | 50   | 12.0379         |
+| 8.7424        | 3.3898 | 75   | 5.7217          |
 
 
 ### Framework versions

adapter_config.json

@@ -10,7 +10,7 @@
   "layers_pattern": null,
   "layers_to_transform": null,
   "loftq_config": {},
-  "lora_alpha": 32,
+  "lora_alpha": 64,
   "lora_dropout": 0.1,
   "megatron_config": null,
   "megatron_core": "megatron.core",
@@ -19,7 +19,7 @@
   "r": 64,
   "rank_pattern": {},
   "revision": null,
-  "target_modules": ".*(text_model|modality_projection|perceiver_resampler).*(down_proj|gate_proj|up_proj|k_proj|q_proj|v_proj|o_proj).*$",
+  "target_modules": ".*(modality_projection|perceiver_resampler).*(k_proj|q_proj|v_proj).*$",
   "task_type": null,
   "use_dora": false,
   "use_rslora": false

adapter_model.safetensors

@@ -1,3 +1,3 @@
 version https://git-lfs.github.com/spec/v1
-oid sha256:3fb7f2057c99555a51fe5cd8c567e2074fcdafa2abd5729efee6a32593ceb6ca
-size 746528304
+oid sha256:9a9d5c0b60938cc7f723069477b9643af616c7d986d8a6d03123494906a3fe13
+size 11209608

training_args.bin

@@ -1,3 +1,3 @@
 version https://git-lfs.github.com/spec/v1
-oid sha256:1a401f4ac56bb712a238b846f3e8de3ababd9a13d6643b0c95762adbcbb5ef94
+oid sha256:851879ab1b20cd37c91ab4c6b87202529d9a58dacae5ac3d034412afedc526e0
 size 5240