AdamStormed22/granite-vision-weed-vqa-small

README.md

@@ -34,10 +34,10 @@ This model was trained with SFT.
 
 ### Framework versions
 
-- TRL: 0.15.2
-- Transformers: 4.50.0.dev0
+- TRL: 0.16.0
+- Transformers: 4.51.0.dev0
 - Pytorch: 2.6.0+cu124
-- Datasets: 3.4.1
+- Datasets: 3.5.0
 - Tokenizers: 0.21.1
 
 ## Citations

adapter_config.json

@@ -6,6 +6,7 @@
   },
   "base_model_name_or_path": "ibm-granite/granite-vision-3.1-2b-preview",
   "bias": "none",
+  "corda_config": null,
   "eva_config": null,
   "exclude_modules": null,
   "fan_in_fan_out": false,
@@ -26,132 +27,133 @@
   "rank_pattern": {},
   "revision": null,
   "target_modules": [
-    "model.layers.1.self_attn.k_proj",
-    "model.layers.6.self_attn.q_proj",
-    "model.layers.18.self_attn.k_proj",
-    "model.layers.7.self_attn.q_proj",
-    "model.layers.12.self_attn.k_proj",
-    "model.layers.11.self_attn.k_proj",
-    "29.self_attn.v_proj",
-    "model.layers.14.self_attn.v_proj",
-    "27.self_attn.q_proj",
-    "model.layers.3.self_attn.q_proj",
-    "model.layers.8.self_attn.q_proj",
-    "model.layers.20.self_attn.q_proj",
-    "model.layers.0.self_attn.q_proj",
-    "model.layers.17.self_attn.q_proj",
-    "37.self_attn.k_proj",
-    "model.layers.26.self_attn.q_proj",
-    "31.self_attn.v_proj",
-    "o_proj",
-    "model.layers.22.self_attn.q_proj",
+    "model.layers.13.self_attn.q_proj",
+    "27.self_attn.k_proj",
+    "gate_proj",
+    "model.layers.10.self_attn.q_proj",
+    "37.self_attn.q_proj",
+    "28.self_attn.k_proj",
+    "30.self_attn.q_proj",
     "model.layers.5.self_attn.v_proj",
-    "32.self_attn.q_proj",
-    "model.layers.21.self_attn.v_proj",
-    "model.layers.24.self_attn.k_proj",
-    "33.self_attn.v_proj",
-    "28.self_attn.v_proj",
-    "30.self_attn.v_proj",
-    "model.layers.9.self_attn.k_proj",
-    "29.self_attn.q_proj",
-    "model.layers.3.self_attn.v_proj",
-    "model.layers.10.self_attn.k_proj",
-    "model.layers.17.self_attn.k_proj",
     "35.self_attn.k_proj",
-    "33.self_attn.q_proj",
-    "39.self_attn.v_proj",
-    "32.self_attn.k_proj",
-    "31.self_attn.q_proj",
-    "model.layers.4.self_attn.v_proj",
-    "model.layers.19.self_attn.v_proj",
-    "36.self_attn.v_proj",
-    "model.layers.16.self_attn.k_proj",
-    "model.layers.26.self_attn.v_proj",
-    "model.layers.6.self_attn.k_proj",
-    "model.layers.17.self_attn.v_proj",
-    "34.self_attn.k_proj",
+    "model.layers.1.self_attn.v_proj",
+    "model.layers.6.self_attn.v_proj",
     "model.layers.25.self_attn.q_proj",
-    "down_proj",
+    "30.self_attn.k_proj",
+    "model.layers.17.self_attn.q_proj",
+    "39.self_attn.k_proj",
+    "model.layers.8.self_attn.q_proj",
+    "model.layers.14.self_attn.v_proj",
+    "31.self_attn.v_proj",
+    "28.self_attn.q_proj",
+    "model.layers.20.self_attn.v_proj",
     "model.layers.23.self_attn.q_proj",
-    "model.layers.19.self_attn.q_proj",
-    "model.layers.14.self_attn.k_proj",
-    "model.layers.7.self_attn.k_proj",
-    "model.layers.20.self_attn.k_proj",
-    "model.layers.4.self_attn.k_proj",
-    "model.layers.0.self_attn.v_proj",
-    "29.self_attn.k_proj",
-    "model.layers.8.self_attn.k_proj",
+    "model.layers.21.self_attn.v_proj",
+    "model.layers.2.self_attn.q_proj",
     "up_proj",
-    "model.layers.2.self_attn.k_proj",
-    "model.layers.22.self_attn.v_proj",
-    "model.layers.16.self_attn.q_proj",
-    "model.layers.25.self_attn.k_proj",
-    "model.layers.23.self_attn.v_proj",
-    "model.layers.12.self_attn.q_proj",
-    "model.layers.18.self_attn.v_proj",
-    "model.layers.4.self_attn.q_proj",
+    "model.layers.9.self_attn.q_proj",
+    "model.layers.0.self_attn.q_proj",
+    "model.layers.7.self_attn.v_proj",
+    "37.self_attn.v_proj",
+    "model.layers.1.self_attn.q_proj",
+    "model.layers.25.self_attn.v_proj",
+    "model.layers.4.self_attn.k_proj",
+    "model.layers.7.self_attn.k_proj",
+    "model.layers.13.self_attn.k_proj",
     "model.layers.22.self_attn.k_proj",
     "33.self_attn.k_proj",
-    "model.layers.1.self_attn.v_proj",
-    "model.layers.10.self_attn.q_proj",
-    "27.self_attn.k_proj",
-    "model.layers.9.self_attn.q_proj",
-    "35.self_attn.v_proj",
-    "model.layers.9.self_attn.v_proj",
+    "model.layers.9.self_attn.k_proj",
+    "model.layers.16.self_attn.v_proj",
+    "model.layers.21.self_attn.k_proj",
+    "31.self_attn.q_proj",
     "model.layers.15.self_attn.q_proj",
-    "model.layers.15.self_attn.v_proj",
-    "38.self_attn.k_proj",
-    "model.layers.19.self_attn.k_proj",
-    "model.layers.6.self_attn.v_proj",
-    "34.self_attn.v_proj",
-    "36.self_attn.q_proj",
-    "35.self_attn.q_proj",
-    "39.self_attn.k_proj",
-    "28.self_attn.k_proj",
-    "28.self_attn.q_proj",
-    "model.layers.13.self_attn.q_proj",
-    "model.layers.2.self_attn.v_proj",
-    "27.self_attn.v_proj",
-    "model.layers.24.self_attn.v_proj",
+    "model.layers.14.self_attn.k_proj",
+    "model.layers.7.self_attn.q_proj",
+    "model.layers.13.self_attn.v_proj",
+    "29.self_attn.v_proj",
+    "model.layers.4.self_attn.q_proj",
     "model.layers.24.self_attn.q_proj",
-    "model.layers.11.self_attn.q_proj",
-    "model.layers.12.self_attn.v_proj",
-    "model.layers.7.self_attn.v_proj",
-    "34.self_attn.q_proj",
-    "model.layers.23.self_attn.k_proj",
-    "39.self_attn.q_proj",
-    "model.layers.21.self_attn.k_proj",
-    "model.layers.20.self_attn.v_proj",
-    "36.self_attn.k_proj",
     "model.layers.21.self_attn.q_proj",
-    "model.layers.11.self_attn.v_proj",
-    "model.layers.14.self_attn.q_proj",
-    "gate_proj",
-    "model.layers.18.self_attn.q_proj",
-    "model.layers.10.self_attn.v_proj",
-    "model.layers.0.self_attn.k_proj",
-    "31.self_attn.k_proj",
-    "30.self_attn.k_proj",
-    "model.layers.1.self_attn.q_proj",
-    "37.self_attn.q_proj",
-    "model.layers.26.self_attn.k_proj",
-    "model.layers.13.self_attn.v_proj",
+    "38.self_attn.v_proj",
+    "model.layers.23.self_attn.k_proj",
+    "down_proj",
+    "model.layers.11.self_attn.k_proj",
+    "model.layers.19.self_attn.q_proj",
     "model.layers.3.self_attn.k_proj",
-    "model.layers.25.self_attn.v_proj",
-    "30.self_attn.q_proj",
     "model.layers.8.self_attn.v_proj",
-    "model.layers.13.self_attn.k_proj",
-    "38.self_attn.v_proj",
+    "model.layers.18.self_attn.q_proj",
+    "35.self_attn.v_proj",
+    "model.layers.18.self_attn.k_proj",
+    "36.self_attn.v_proj",
+    "32.self_attn.k_proj",
+    "model.layers.16.self_attn.q_proj",
+    "model.layers.16.self_attn.k_proj",
+    "model.layers.3.self_attn.q_proj",
+    "model.layers.23.self_attn.v_proj",
+    "model.layers.15.self_attn.v_proj",
+    "o_proj",
     "model.layers.5.self_attn.q_proj",
-    "37.self_attn.v_proj",
+    "30.self_attn.v_proj",
+    "model.layers.1.self_attn.k_proj",
+    "model.layers.11.self_attn.v_proj",
+    "39.self_attn.q_proj",
+    "model.layers.6.self_attn.q_proj",
+    "model.layers.20.self_attn.k_proj",
+    "model.layers.26.self_attn.k_proj",
     "model.layers.15.self_attn.k_proj",
+    "model.layers.4.self_attn.v_proj",
+    "28.self_attn.v_proj",
+    "model.layers.18.self_attn.v_proj",
+    "model.layers.12.self_attn.v_proj",
+    "34.self_attn.k_proj",
+    "model.layers.26.self_attn.q_proj",
+    "35.self_attn.q_proj",
+    "model.layers.12.self_attn.k_proj",
+    "model.layers.22.self_attn.v_proj",
+    "model.layers.0.self_attn.k_proj",
+    "model.layers.11.self_attn.q_proj",
+    "32.self_attn.v_proj",
+    "29.self_attn.q_proj",
+    "model.layers.0.self_attn.v_proj",
+    "27.self_attn.q_proj",
     "model.layers.5.self_attn.k_proj",
+    "model.layers.6.self_attn.k_proj",
+    "32.self_attn.q_proj",
+    "model.layers.2.self_attn.v_proj",
+    "33.self_attn.v_proj",
+    "model.layers.9.self_attn.v_proj",
+    "model.layers.20.self_attn.q_proj",
+    "37.self_attn.k_proj",
+    "33.self_attn.q_proj",
+    "34.self_attn.q_proj",
+    "model.layers.19.self_attn.k_proj",
+    "model.layers.26.self_attn.v_proj",
+    "model.layers.8.self_attn.k_proj",
+    "34.self_attn.v_proj",
+    "model.layers.24.self_attn.k_proj",
+    "model.layers.14.self_attn.q_proj",
+    "model.layers.2.self_attn.k_proj",
     "38.self_attn.q_proj",
-    "model.layers.2.self_attn.q_proj",
-    "32.self_attn.v_proj",
-    "model.layers.16.self_attn.v_proj"
+    "36.self_attn.q_proj",
+    "model.layers.22.self_attn.q_proj",
+    "36.self_attn.k_proj",
+    "model.layers.24.self_attn.v_proj",
+    "model.layers.12.self_attn.q_proj",
+    "39.self_attn.v_proj",
+    "38.self_attn.k_proj",
+    "27.self_attn.v_proj",
+    "model.layers.3.self_attn.v_proj",
+    "29.self_attn.k_proj",
+    "model.layers.10.self_attn.k_proj",
+    "model.layers.17.self_attn.k_proj",
+    "model.layers.17.self_attn.v_proj",
+    "model.layers.19.self_attn.v_proj",
+    "model.layers.10.self_attn.v_proj",
+    "model.layers.25.self_attn.k_proj",
+    "31.self_attn.k_proj"
   ],
   "task_type": null,
+  "trainable_token_indices": null,
   "use_dora": true,
   "use_rslora": false
 }

adapter_model.safetensors

@@ -1,3 +1,3 @@
 version https://git-lfs.github.com/spec/v1
-oid sha256:aaa6c72aa084fb2b58ec7b4a9339a880965c7628702d23409d08e6de04b58a3b
+oid sha256:edd5964576ba1b34aae2b58e73100c4467715ef05de2ae114f7b6428723ad374
 size 60254184

training_args.bin

@@ -1,3 +1,3 @@
 version https://git-lfs.github.com/spec/v1
-oid sha256:9b9e06e56620676184be8a62ed95757e452bf1d2641dc6af7e84e4f458120eb4
-size 5560
+oid sha256:02cb2b993a1111e13f7077c6e3080aad7109b8b38778a7b4e05efacf20dd46e4
+size 5624