Training in progress, step 100, checkpoint

checkpoint-100/adapter_config.json

@@ -23,13 +23,13 @@
   "rank_pattern": {},
   "revision": null,
   "target_modules": [
-    "o_proj",
-    "down_proj",
     "gate_proj",
     "k_proj",
+    "o_proj",
+    "up_proj",
     "q_proj",
-    "v_proj",
-    "up_proj"
+    "down_proj",
+    "v_proj"
   ],
   "task_type": "CAUSAL_LM",
   "use_dora": false,

checkpoint-100/adapter_model.safetensors

@@ -1,3 +1,3 @@
 version https://git-lfs.github.com/spec/v1
-oid sha256:9d669ca556525e4c398acc432d7b286cab3d23d295402cca13771e3c807f4635
+oid sha256:fe4afc1dd1279e164c88fd39db3593809491cb76f947aec02b1a63a72efbcfad
 size 145287696

checkpoint-100/optimizer.pt

@@ -1,3 +1,3 @@
 version https://git-lfs.github.com/spec/v1
-oid sha256:e918cc212e6fe1b9706095cfac57f55be36da441c619b8ade0cf4222990c1244
+oid sha256:32763b3d5c4d9a387b699d46b0d51e1280e26c9c98bdff5302b052631883b5ad
 size 290833618

checkpoint-100/rng_state.pth

@@ -1,3 +1,3 @@
 version https://git-lfs.github.com/spec/v1
-oid sha256:f78502ea07713778b3fe788600db560d993b4e0be1abcdaf33b7608606fefed4
+oid sha256:5f869d29017196ae5e7e99232cd3274dc3b10e36fabc2ae1653327876d18f97a
 size 14244

checkpoint-100/trainer_state.json

@@ -10,88 +10,88 @@
   "log_history": [
     {
       "epoch": 0.010869565217391304,
-      "grad_norm": 2.120478868484497,
+      "grad_norm": 2.4570870399475098,
       "learning_rate": 4.981884057971015e-05,
-      "loss": 2.4608,
+      "loss": 2.3612,
       "step": 10
     },
     {
       "epoch": 0.021739130434782608,
-      "grad_norm": 1.4271776676177979,
+      "grad_norm": 1.2042421102523804,
       "learning_rate": 4.963768115942029e-05,
-      "loss": 1.2741,
+      "loss": 1.1915,
       "step": 20
     },
     {
       "epoch": 0.03260869565217391,
-      "grad_norm": 0.9902746677398682,
+      "grad_norm": 1.2019174098968506,
       "learning_rate": 4.945652173913044e-05,
-      "loss": 0.9403,
+      "loss": 0.8992,
       "step": 30
     },
     {
       "epoch": 0.043478260869565216,
-      "grad_norm": 1.0730223655700684,
+      "grad_norm": 1.0546120405197144,
       "learning_rate": 4.9275362318840584e-05,
-      "loss": 0.7517,
+      "loss": 0.7276,
       "step": 40
     },
     {
       "epoch": 0.05434782608695652,
-      "grad_norm": 0.8592391610145569,
+      "grad_norm": 0.754091203212738,
       "learning_rate": 4.909420289855073e-05,
-      "loss": 0.6695,
+      "loss": 0.6419,
       "step": 50
     },
     {
       "epoch": 0.05434782608695652,
-      "eval_loss": 0.8473024964332581,
-      "eval_runtime": 10.3597,
-      "eval_samples_per_second": 46.334,
-      "eval_steps_per_second": 2.896,
+      "eval_loss": 0.7905128598213196,
+      "eval_runtime": 10.8418,
+      "eval_samples_per_second": 44.273,
+      "eval_steps_per_second": 2.767,
       "step": 50
     },
     {
       "epoch": 0.06521739130434782,
-      "grad_norm": 0.7817952632904053,
+      "grad_norm": 0.7751966714859009,
       "learning_rate": 4.891304347826087e-05,
-      "loss": 0.6325,
+      "loss": 0.5971,
       "step": 60
     },
     {
       "epoch": 0.07608695652173914,
-      "grad_norm": 0.8163846135139465,
+      "grad_norm": 0.6874057650566101,
       "learning_rate": 4.873188405797102e-05,
-      "loss": 0.5994,
+      "loss": 0.608,
       "step": 70
     },
     {
       "epoch": 0.08695652173913043,
-      "grad_norm": 0.7358843088150024,
+      "grad_norm": 0.7145748734474182,
       "learning_rate": 4.855072463768116e-05,
-      "loss": 0.6158,
+      "loss": 0.6111,
       "step": 80
     },
     {
       "epoch": 0.09782608695652174,
-      "grad_norm": 0.7856508493423462,
+      "grad_norm": 0.8841484189033508,
       "learning_rate": 4.836956521739131e-05,
-      "loss": 0.5983,
+      "loss": 0.5848,
       "step": 90
     },
     {
       "epoch": 0.10869565217391304,
-      "grad_norm": 0.8615956902503967,
+      "grad_norm": 0.7937784790992737,
       "learning_rate": 4.818840579710145e-05,
-      "loss": 0.5858,
+      "loss": 0.5699,
       "step": 100
     },
     {
       "epoch": 0.10869565217391304,
-      "eval_loss": 0.7398766875267029,
-      "eval_runtime": 10.4317,
-      "eval_samples_per_second": 46.014,
-      "eval_steps_per_second": 2.876,
+      "eval_loss": 0.6956210732460022,
+      "eval_runtime": 10.863,
+      "eval_samples_per_second": 44.187,
+      "eval_steps_per_second": 2.762,
       "step": 100
     }
   ],
@@ -112,7 +112,7 @@
       "attributes": {}
     }
   },
-  "total_flos": 8762320739893248.0,
+  "total_flos": 8760080023683072.0,
   "train_batch_size": 16,
   "trial_name": null,
   "trial_params": null

checkpoint-100/training_args.bin

@@ -1,3 +1,3 @@
 version https://git-lfs.github.com/spec/v1
-oid sha256:a3712c5fdd6e5241e8ea26ae62b28612fc514ec88e8a03bc6c1b3a46acfc611a
+oid sha256:81786d634f359bee447d9c2dad023b92502900c87e9208470ed326b1204c5b1a
 size 5816