Training in progress, step 60, checkpoint

last-checkpoint/adapter_config.json

@@ -20,8 +20,8 @@
   "rank_pattern": {},
   "revision": null,
   "target_modules": [
-    "query_key_valuelm_head",
-    "word_embeddings"
+    "word_embeddings",
+    "query_key_valuelm_head"
   ],
   "task_type": "CAUSAL_LM",
   "use_dora": false,

last-checkpoint/adapter_model.safetensors

@@ -1,3 +1,3 @@
 version https://git-lfs.github.com/spec/v1
-oid sha256:de52612f18d94eb85267cdbaa87cfac914b5a3415370220f048ff6627b5dc973
+oid sha256:7b5e1d7fec5642c5a7eebc261811b86c8379352efb270f0d53c3bdd0277083ab
 size 8077608

last-checkpoint/optimizer.pt

@@ -1,3 +1,3 @@
 version https://git-lfs.github.com/spec/v1
-oid sha256:c7da0cccb26e635a9bb8fe359b98f58b1fe7d55939c5f6f51514ca872caa19e9
+oid sha256:c295c871992ea2658a9d7ad487d59b40db23c7cfbb6615cf5f83edf6deb52d2c
 size 4052500

last-checkpoint/rng_state.pth

@@ -1,3 +1,3 @@
 version https://git-lfs.github.com/spec/v1
-oid sha256:386fcc8cc1089aade9450d86fb239ea3483f455fd2d78d8378645feecfec9d69
+oid sha256:384b1441690cc32db0a09ece90eb10816e434402fb6deac7559104dc40fcf585
 size 14244

last-checkpoint/scheduler.pt

@@ -1,3 +1,3 @@
 version https://git-lfs.github.com/spec/v1
-oid sha256:960f0c2ec75c7a0d06bf09f5118049f84af6550acbf637b5ec1d7409c50149b7
+oid sha256:b215c0947378bd341b267cea5a7112696ec035345c849e38fb6a3c58f28d2185
 size 1064

last-checkpoint/trainer_state.json

@@ -10,28 +10,28 @@
   "log_history": [
     {
       "epoch": 0.0,
-      "grad_norm": 87.1609115600586,
-      "learning_rate": 2.341780598043574e-05,
-      "loss": 4.6745,
+      "grad_norm": 79.12922668457031,
+      "learning_rate": 1e-06,
+      "loss": 4.6771,
       "step": 20
     },
     {
       "epoch": 0.0,
-      "grad_norm": 226.922119140625,
-      "learning_rate": 7.97387338950315e-06,
-      "loss": 4.5875,
+      "grad_norm": 239.7991943359375,
+      "learning_rate": 2e-06,
+      "loss": 4.5023,
       "step": 40
     },
     {
       "epoch": 0.0,
-      "grad_norm": 86.45769500732422,
-      "learning_rate": 0.0,
-      "loss": 4.8834,
+      "grad_norm": 77.5164794921875,
+      "learning_rate": 3e-06,
+      "loss": 4.8171,
       "step": 60
     }
   ],
   "logging_steps": 20,
-  "max_steps": 60,
+  "max_steps": 20000,
   "num_input_tokens_seen": 0,
   "num_train_epochs": 1,
   "save_steps": 20,

last-checkpoint/training_args.bin

@@ -1,3 +1,3 @@
 version https://git-lfs.github.com/spec/v1
-oid sha256:83256b55088d45d0ebeda1319a91e9b247c70b5f4d82cb04a018aa4a684554bc
+oid sha256:32e994f25267341e613f4d352977cc7a2847de358db5ece7a60fcf21be944170
 size 4984