Model save

README.md

@@ -4,9 +4,9 @@ library_name: transformers
 model_name: dispatchr-grpo
 tags:
 - generated_from_trainer
-- hf_jobs
-- grpo
 - trl
+- grpo
+- hf_jobs
 licence: license
 ---
 
@@ -28,7 +28,7 @@ print(output["generated_text"])
 
 ## Training procedure
 
-[<img src="https://raw.githubusercontent.com/wandb/assets/main/wandb-github-badge-28.svg" alt="Visualize in Weights & Biases" width="150" height="24"/>](https://wandb.ai/ggtejas/dispatchr-grpo/runs/459qeioi) 
+[<img src="https://raw.githubusercontent.com/wandb/assets/main/wandb-github-badge-28.svg" alt="Visualize in Weights & Biases" width="150" height="24"/>](https://wandb.ai/ggtejas/dispatchr-grpo/runs/6sez5w7m) 
 
 
 

adapter_config.json

@@ -30,10 +30,10 @@
   "rank_pattern": {},
   "revision": null,
   "target_modules": [
-    "q_proj",
-    "k_proj",
+    "o_proj",
     "v_proj",
-    "o_proj"
+    "q_proj",
+    "k_proj"
   ],
   "target_parameters": null,
   "task_type": "CAUSAL_LM",

adapter_model.safetensors

@@ -1,3 +1,3 @@
 version https://git-lfs.github.com/spec/v1
-oid sha256:e180fd4eb68d3a9a950cd63ea03484e33930ac078eb8874f1b5a7583df235c65
+oid sha256:2325e085d84dcd0291d43ba318b64506cfabe6dc06c5a243af93a7ce0c006f55
 size 47224624

metrics.jsonl

@@ -1,150 +1,75 @@
-{"batch": 1, "reward": null, "loss": 0.1223}
-{"batch": 2, "reward": null, "loss": -0.1174}
+{"batch": 1, "reward": null, "loss": 0.0}
+{"batch": 2, "reward": null, "loss": 0.0}
 {"batch": 3, "reward": null, "loss": 0.0}
-{"batch": 4, "reward": null, "loss": -0.0111}
-{"batch": 5, "reward": null, "loss": -0.132}
+{"batch": 4, "reward": null, "loss": 0.0}
+{"batch": 5, "reward": null, "loss": 0.0}
 {"batch": 6, "reward": null, "loss": 0.0}
-{"batch": 7, "reward": null, "loss": -0.0338}
-{"batch": 8, "reward": null, "loss": -0.0746}
-{"batch": 9, "reward": null, "loss": -0.0216}
+{"batch": 7, "reward": null, "loss": 0.0}
+{"batch": 8, "reward": null, "loss": 0.0}
+{"batch": 9, "reward": null, "loss": 0.0}
 {"batch": 10, "reward": null, "loss": 0.0}
-{"batch": 11, "reward": null, "loss": -0.0258}
-{"batch": 12, "reward": null, "loss": -0.0447}
-{"batch": 13, "reward": null, "loss": 0.0001}
-{"batch": 14, "reward": null, "loss": 0.1142}
-{"batch": 15, "reward": null, "loss": 0.2805}
-{"batch": 16, "reward": null, "loss": 0.0047}
-{"batch": 17, "reward": null, "loss": -0.0791}
-{"batch": 18, "reward": null, "loss": 0.0315}
-{"batch": 19, "reward": null, "loss": -0.0028}
-{"batch": 20, "reward": null, "loss": -0.1793}
-{"batch": 21, "reward": null, "loss": -0.1458}
-{"batch": 22, "reward": null, "loss": -0.3561}
-{"batch": 23, "reward": null, "loss": -0.1178}
-{"batch": 24, "reward": null, "loss": 0.0312}
-{"batch": 25, "reward": null, "loss": 0.0001}
-{"batch": 26, "reward": null, "loss": -0.2916}
-{"batch": 27, "reward": null, "loss": -0.0419}
-{"batch": 28, "reward": null, "loss": -0.1319}
-{"batch": 29, "reward": null, "loss": -0.0119}
-{"batch": 30, "reward": null, "loss": 0.0179}
-{"batch": 31, "reward": null, "loss": 0.0001}
+{"batch": 11, "reward": null, "loss": 0.5388}
+{"batch": 12, "reward": null, "loss": 0.0}
+{"batch": 13, "reward": null, "loss": 0.0}
+{"batch": 14, "reward": null, "loss": 0.0}
+{"batch": 15, "reward": null, "loss": 0.0}
+{"batch": 16, "reward": null, "loss": 0.0}
+{"batch": 17, "reward": null, "loss": 0.0}
+{"batch": 18, "reward": null, "loss": -1.0771}
+{"batch": 19, "reward": null, "loss": 0.0}
+{"batch": 20, "reward": null, "loss": 0.0}
+{"batch": 21, "reward": null, "loss": 0.0}
+{"batch": 22, "reward": null, "loss": 0.0}
+{"batch": 23, "reward": null, "loss": 0.0}
+{"batch": 24, "reward": null, "loss": 0.0}
+{"batch": 25, "reward": null, "loss": 0.1115}
+{"batch": 26, "reward": null, "loss": 0.0}
+{"batch": 27, "reward": null, "loss": 0.0001}
+{"batch": 28, "reward": null, "loss": 0.0}
+{"batch": 29, "reward": null, "loss": 0.1189}
+{"batch": 30, "reward": null, "loss": 0.0}
+{"batch": 31, "reward": null, "loss": 0.0}
 {"batch": 32, "reward": null, "loss": 0.0}
-{"batch": 33, "reward": null, "loss": -0.2818}
-{"batch": 34, "reward": null, "loss": 0.0001}
-{"batch": 35, "reward": null, "loss": 0.0811}
-{"batch": 36, "reward": null, "loss": 0.0384}
-{"batch": 37, "reward": null, "loss": -0.1263}
-{"batch": 38, "reward": null, "loss": 0.0001}
-{"batch": 39, "reward": null, "loss": 0.0003}
-{"batch": 40, "reward": null, "loss": -0.0356}
-{"batch": 41, "reward": null, "loss": 0.0276}
-{"batch": 42, "reward": null, "loss": 0.0001}
-{"batch": 43, "reward": null, "loss": 0.187}
-{"batch": 44, "reward": null, "loss": -0.0529}
-{"batch": 45, "reward": null, "loss": -0.0405}
-{"batch": 46, "reward": null, "loss": -0.1146}
-{"batch": 47, "reward": null, "loss": -0.1078}
-{"batch": 48, "reward": null, "loss": -0.0399}
-{"batch": 49, "reward": null, "loss": 0.1319}
-{"batch": 50, "reward": null, "loss": 0.0003}
-{"batch": 51, "reward": null, "loss": -0.0739}
-{"batch": 52, "reward": null, "loss": -0.0082}
-{"batch": 53, "reward": null, "loss": -0.1284}
-{"batch": 54, "reward": null, "loss": -0.0489}
-{"batch": 55, "reward": null, "loss": 0.0009}
-{"batch": 56, "reward": null, "loss": -0.0061}
-{"batch": 57, "reward": null, "loss": -0.0458}
-{"batch": 58, "reward": null, "loss": -0.118}
-{"batch": 59, "reward": null, "loss": 0.0003}
-{"batch": 60, "reward": null, "loss": 0.0003}
-{"batch": 61, "reward": null, "loss": -0.0654}
-{"batch": 62, "reward": null, "loss": 0.0491}
-{"batch": 63, "reward": null, "loss": 0.0826}
-{"batch": 64, "reward": null, "loss": 0.0006}
-{"batch": 65, "reward": null, "loss": 0.0892}
-{"batch": 66, "reward": null, "loss": 0.0003}
-{"batch": 67, "reward": null, "loss": 0.0105}
-{"batch": 68, "reward": null, "loss": -0.0459}
-{"batch": 69, "reward": null, "loss": -0.0484}
-{"batch": 70, "reward": null, "loss": 0.0006}
-{"batch": 71, "reward": null, "loss": 0.1133}
-{"batch": 72, "reward": null, "loss": 0.0008}
-{"batch": 73, "reward": null, "loss": 0.0002}
-{"batch": 74, "reward": null, "loss": 0.0012}
-{"batch": 75, "reward": null, "loss": 0.2767}
-{"batch": 76, "reward": null, "loss": 0.0011}
-{"batch": 77, "reward": null, "loss": 0.0757}
-{"batch": 78, "reward": null, "loss": 0.0259}
-{"batch": 79, "reward": null, "loss": -0.0429}
-{"batch": 80, "reward": null, "loss": -0.1146}
-{"batch": 81, "reward": null, "loss": 0.065}
-{"batch": 82, "reward": null, "loss": 0.0711}
-{"batch": 83, "reward": null, "loss": 0.1249}
-{"batch": 84, "reward": null, "loss": 0.0003}
-{"batch": 85, "reward": null, "loss": 0.0178}
-{"batch": 86, "reward": null, "loss": 0.0007}
-{"batch": 87, "reward": null, "loss": -0.0893}
-{"batch": 88, "reward": null, "loss": 0.069}
-{"batch": 89, "reward": null, "loss": -0.0719}
-{"batch": 90, "reward": null, "loss": 0.1625}
-{"batch": 91, "reward": null, "loss": 0.4483}
-{"batch": 92, "reward": null, "loss": 0.0172}
-{"batch": 93, "reward": null, "loss": -0.0406}
-{"batch": 94, "reward": null, "loss": -0.0199}
-{"batch": 95, "reward": null, "loss": 0.055}
-{"batch": 96, "reward": null, "loss": -0.0498}
-{"batch": 97, "reward": null, "loss": -0.0119}
-{"batch": 98, "reward": null, "loss": 0.0016}
-{"batch": 99, "reward": null, "loss": 0.0636}
-{"batch": 100, "reward": null, "loss": -0.1874}
-{"batch": 101, "reward": null, "loss": -0.0192}
-{"batch": 102, "reward": null, "loss": 0.0008}
-{"batch": 103, "reward": null, "loss": -0.0164}
-{"batch": 104, "reward": null, "loss": 0.0971}
-{"batch": 105, "reward": null, "loss": -0.1624}
-{"batch": 106, "reward": null, "loss": -0.0265}
-{"batch": 107, "reward": null, "loss": 0.0009}
-{"batch": 108, "reward": null, "loss": -0.0832}
-{"batch": 109, "reward": null, "loss": 0.0008}
-{"batch": 110, "reward": null, "loss": -0.102}
-{"batch": 111, "reward": null, "loss": -0.0063}
-{"batch": 112, "reward": null, "loss": 0.0372}
-{"batch": 113, "reward": null, "loss": 0.2034}
-{"batch": 114, "reward": null, "loss": 0.0009}
-{"batch": 115, "reward": null, "loss": 0.0016}
-{"batch": 116, "reward": null, "loss": -0.0357}
-{"batch": 117, "reward": null, "loss": -0.0206}
-{"batch": 118, "reward": null, "loss": -0.0184}
-{"batch": 119, "reward": null, "loss": 0.0014}
-{"batch": 120, "reward": null, "loss": -0.07}
-{"batch": 121, "reward": null, "loss": 0.0893}
-{"batch": 122, "reward": null, "loss": 0.1406}
-{"batch": 123, "reward": null, "loss": 0.0706}
-{"batch": 124, "reward": null, "loss": -0.0323}
-{"batch": 125, "reward": null, "loss": -0.0599}
-{"batch": 126, "reward": null, "loss": -0.0272}
-{"batch": 127, "reward": null, "loss": 0.028}
-{"batch": 128, "reward": null, "loss": 0.0517}
-{"batch": 129, "reward": null, "loss": -0.0453}
-{"batch": 130, "reward": null, "loss": 0.0409}
-{"batch": 131, "reward": null, "loss": 0.1233}
-{"batch": 132, "reward": null, "loss": 0.0385}
-{"batch": 133, "reward": null, "loss": -0.0431}
-{"batch": 134, "reward": null, "loss": 0.0568}
-{"batch": 135, "reward": null, "loss": -0.0328}
-{"batch": 136, "reward": null, "loss": 0.0237}
-{"batch": 137, "reward": null, "loss": -0.0437}
-{"batch": 138, "reward": null, "loss": 0.0639}
-{"batch": 139, "reward": null, "loss": -0.0434}
-{"batch": 140, "reward": null, "loss": 0.0614}
-{"batch": 141, "reward": null, "loss": -0.0629}
-{"batch": 142, "reward": null, "loss": -0.0654}
-{"batch": 143, "reward": null, "loss": -0.0044}
-{"batch": 144, "reward": null, "loss": -0.0299}
-{"batch": 145, "reward": null, "loss": 0.0016}
-{"batch": 146, "reward": null, "loss": -0.0391}
-{"batch": 147, "reward": null, "loss": -0.0656}
-{"batch": 148, "reward": null, "loss": 0.0452}
-{"batch": 149, "reward": null, "loss": 0.0262}
-{"batch": 150, "reward": null, "loss": 0.0412}
+{"batch": 33, "reward": null, "loss": 0.0}
+{"batch": 34, "reward": null, "loss": 0.0}
+{"batch": 35, "reward": null, "loss": 0.3886}
+{"batch": 36, "reward": null, "loss": 0.0}
+{"batch": 37, "reward": null, "loss": 0.0}
+{"batch": 38, "reward": null, "loss": 0.0}
+{"batch": 39, "reward": null, "loss": 0.0}
+{"batch": 40, "reward": null, "loss": 0.3769}
+{"batch": 41, "reward": null, "loss": 0.0}
+{"batch": 42, "reward": null, "loss": 0.0}
+{"batch": 43, "reward": null, "loss": 0.0}
+{"batch": 44, "reward": null, "loss": 0.0}
+{"batch": 45, "reward": null, "loss": 0.0}
+{"batch": 46, "reward": null, "loss": 0.0001}
+{"batch": 47, "reward": null, "loss": 0.0}
+{"batch": 48, "reward": null, "loss": -0.1641}
+{"batch": 49, "reward": null, "loss": 0.0}
+{"batch": 50, "reward": null, "loss": -0.0944}
+{"batch": 51, "reward": null, "loss": 0.0}
+{"batch": 52, "reward": null, "loss": 0.0}
+{"batch": 53, "reward": null, "loss": 0.0}
+{"batch": 54, "reward": null, "loss": 0.0}
+{"batch": 55, "reward": null, "loss": 0.0}
+{"batch": 56, "reward": null, "loss": 0.0}
+{"batch": 57, "reward": null, "loss": 0.0}
+{"batch": 58, "reward": null, "loss": 0.0}
+{"batch": 59, "reward": null, "loss": 0.0}
+{"batch": 60, "reward": null, "loss": 0.0}
+{"batch": 61, "reward": null, "loss": -0.4932}
+{"batch": 62, "reward": null, "loss": 0.0}
+{"batch": 63, "reward": null, "loss": 0.0335}
+{"batch": 64, "reward": null, "loss": 0.0}
+{"batch": 65, "reward": null, "loss": 0.0}
+{"batch": 66, "reward": null, "loss": -0.0811}
+{"batch": 67, "reward": null, "loss": 0.0}
+{"batch": 68, "reward": null, "loss": 0.0}
+{"batch": 69, "reward": null, "loss": 0.0}
+{"batch": 70, "reward": null, "loss": 0.3967}
+{"batch": 71, "reward": null, "loss": 0.0}
+{"batch": 72, "reward": null, "loss": 0.0002}
+{"batch": 73, "reward": null, "loss": 0.0}
+{"batch": 74, "reward": null, "loss": 0.0}
+{"batch": 75, "reward": null, "loss": 0.0}

training_args.bin

@@ -1,3 +1,3 @@
 version https://git-lfs.github.com/spec/v1
-oid sha256:764ad12f95df558b1167c061b7a4e858e3e8b2dfb0c189e9027a161141149641
+oid sha256:2b346bee98cdce11d13871c68c2cac5fc786cbd0bcbfcad7aaf525b08f9f5300
 size 7889

trajectory.jsonl

@@ -1,3 +1,3 @@
 version https://git-lfs.github.com/spec/v1
-oid sha256:08b1ec356fe40d72b7d8245a3720f8dc83dff2fc972629d5c941ec2c130ef62c
-size 534821
+oid sha256:bc3853aaa72c9955142cd99ccd7a616be3da3e960dc35f6791097451f37400de
+size 95401