Training in progress, step 20, checkpoint

last-checkpoint/adapter_config.json

@@ -20,8 +20,8 @@
   "rank_pattern": {},
   "revision": null,
   "target_modules": [
-    "word_embeddings",
-    "query_key_value"
+    "query_key_value",
+    "word_embeddings"
   ],
   "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 sha256:5b717fc47fc6ca0f43cd64801c3f49ad862250ba32a036c80639250fcb7e18cd
+oid sha256:e4c346635a502bcfa1a0f2d9987faa2f711d867fc2d1e08232dbd6532240dd22
 size 12803224

last-checkpoint/optimizer.pt

@@ -1,3 +1,3 @@
 version https://git-lfs.github.com/spec/v1
-oid sha256:5295139bb5ea53c3cb17ab006d58608d0220293d8748ba7d455a435ffff93806
+oid sha256:c098e01e074e9d4bfc800ec514479b770a71cb68eaefd4ebf3a41b18222ecd43
 size 6472698

last-checkpoint/scheduler.pt

@@ -1,3 +1,3 @@
 version https://git-lfs.github.com/spec/v1
-oid sha256:1fcf032e6256b2cd4d659b09985077ab1961c6c13c26975c094dd36ab0c1ff74
+oid sha256:647a2db5e5d077e3cb72e5db1d70d7d8e77b6ca8a1f7da7bfb358e39a0b5482f
 size 1064

last-checkpoint/tokenizer.json

@@ -1,3 +1,3 @@
 version https://git-lfs.github.com/spec/v1
-oid sha256:17a208233d2ee8d8c83b23bc214df737c44806a1919f444e89b31e586cd956ba
-size 14500471
+oid sha256:12cb93e09fed5cf770f42b645926821cd300de44a57db3ca5e284142571864c2
+size 14500638

last-checkpoint/trainer_state.json

@@ -1,7 +1,7 @@
 {
   "best_metric": null,
   "best_model_checkpoint": null,
-  "epoch": 0.002725538293813028,
+  "epoch": 1.5565050236199638e-05,
   "eval_steps": 500,
   "global_step": 20,
   "is_hyper_param_search": false,
@@ -10,16 +10,16 @@
   "log_history": [
     {
       "epoch": 0.0,
-      "grad_norm": 36.511024475097656,
-      "learning_rate": 1.0000000000000002e-06,
-      "loss": 4.6323,
+      "grad_norm": 51.71564483642578,
+      "learning_rate": 2.0000000000000003e-06,
+      "loss": 3.4075,
       "step": 20
     }
   ],
   "logging_steps": 20,
-  "max_steps": 20000,
+  "max_steps": 10000,
   "num_input_tokens_seen": 0,
-  "num_train_epochs": 3,
+  "num_train_epochs": 1,
   "save_steps": 20,
   "total_flos": 167893794816000.0,
   "train_batch_size": 1,

last-checkpoint/training_args.bin

@@ -1,3 +1,3 @@
 version https://git-lfs.github.com/spec/v1
-oid sha256:2e025a4e68c1d22e7dc695a50138fa18070a7258d30cfd192237c2cea9e5c360
+oid sha256:b64dd65077fdb49c3763fde38eccd7d99504d918155707100438c27c5ffb8875
 size 4984