Phase3 ckpt step 52

checkpoints/checkpoint-52/adapter_config.json

@@ -34,13 +34,13 @@
   "rank_pattern": {},
   "revision": null,
   "target_modules": [
+    "up_proj",
+    "v_proj",
     "gate_proj",
     "down_proj",
-    "k_proj",
-    "v_proj",
     "o_proj",
     "q_proj",
-    "up_proj"
+    "k_proj"
   ],
   "target_parameters": null,
   "task_type": "CAUSAL_LM",

checkpoints/checkpoint-52/adapter_model.safetensors

@@ -1,3 +1,3 @@
 version https://git-lfs.github.com/spec/v1
-oid sha256:6bb538618cf5b6c6c7fe08c677a6ae69a909b03cac8c45c503231f7046aea6cb
+oid sha256:aceec9bdf8378212454ec3f24b575da6d1b188e33b8090c043ae560028984f7d
 size 119801528

checkpoints/checkpoint-52/optimizer.pt

@@ -1,3 +1,3 @@
 version https://git-lfs.github.com/spec/v1
-oid sha256:3d8c3648df8959ccd799109c08e08eb2977d5321c1bbf0d3f2f63d557124e474
+oid sha256:49aefdc1676e3c200a2119748fe79abfd5b957cc28e90773e812e925c2d4d26a
 size 61397701

checkpoints/checkpoint-52/trainer_state.json

@@ -11,44 +11,44 @@
   "log_history": [
     {
       "epoch": 0.019417475728155338,
-      "grad_norm": 0.4687540531158447,
+      "grad_norm": 0.7171201109886169,
       "learning_rate": 0.0,
-      "loss": 1.063906192779541,
+      "loss": 0.9684253334999084,
       "step": 1
     },
     {
       "epoch": 0.1941747572815534,
-      "grad_norm": 0.6736085414886475,
+      "grad_norm": 0.5563909411430359,
       "learning_rate": 9.524135262330098e-05,
-      "loss": 1.0384196175469294,
+      "loss": 1.013859748840332,
       "step": 10
     },
     {
       "epoch": 0.3883495145631068,
-      "grad_norm": 0.5741080641746521,
+      "grad_norm": 0.6436744928359985,
       "learning_rate": 7.408768370508576e-05,
-      "loss": 1.033409881591797,
+      "loss": 1.0135980606079102,
       "step": 20
     },
     {
       "epoch": 0.5825242718446602,
-      "grad_norm": 0.5239073038101196,
+      "grad_norm": 0.6327415108680725,
       "learning_rate": 4.373333832178478e-05,
-      "loss": 1.0107914924621582,
+      "loss": 0.995547103881836,
       "step": 30
     },
     {
       "epoch": 0.7766990291262136,
-      "grad_norm": 0.5713074803352356,
+      "grad_norm": 0.6593179106712341,
       "learning_rate": 1.5772644703565565e-05,
-      "loss": 1.0056734085083008,
+      "loss": 0.9822251319885253,
       "step": 40
     },
     {
       "epoch": 0.970873786407767,
-      "grad_norm": 0.5943475365638733,
+      "grad_norm": 0.6111292243003845,
       "learning_rate": 8.856374635655695e-07,
-      "loss": 1.0117749214172362,
+      "loss": 0.9803412437438965,
       "step": 50
     }
   ],
@@ -69,7 +69,7 @@
       "attributes": {}
     }
   },
-  "total_flos": 6798745189023744.0,
+  "total_flos": 6772089250578432.0,
   "train_batch_size": 8,
   "trial_name": null,
   "trial_params": null

checkpoints/checkpoint-52/training_args.bin

@@ -1,3 +1,3 @@
 version https://git-lfs.github.com/spec/v1
-oid sha256:af77054b4761c728d1a9242b42f7bdd0489d431f0e4256f4c4a3c07166e09e21
+oid sha256:d0031c4b5488ff8b8f6fc0ee70e2b39e62c1725b61a0a23ef16c061911121f7a
 size 5777