Training in progress, step 20, checkpoint

last-checkpoint/adapter_config.json

@@ -20,13 +20,13 @@
   "rank_pattern": {},
   "revision": null,
   "target_modules": [
-    "up_proj",
-    "o_proj",
-    "gate_proj",
     "v_proj",
-    "k_proj",
+    "up_proj",
     "down_proj",
-    "q_proj"
+    "gate_proj",
+    "o_proj",
+    "q_proj",
+    "k_proj"
   ],
   "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 sha256:3c0ecc2e9a75d14c753ea2123cc052ce634a5ace8a27427e3887b3bd040ff88f
+oid sha256:6937250c3c124f08bcfec45cd4a325f3405cb3f1f96296f7ad31bd5500713d84
 size 100966336

last-checkpoint/optimizer.pt

@@ -1,3 +1,3 @@
 version https://git-lfs.github.com/spec/v1
-oid sha256:73dccc910fc0d80a1f1722383ca6876f72484cabb53773ec050fab503d4d93db
+oid sha256:bb1200ffd26bf33388a55adc3374fd634ce7689aa54e4542ed9e14be83ea8a9d
 size 202110330

last-checkpoint/trainer_state.json

@@ -18,9 +18,9 @@
     {
       "epoch": 0.0028268551236749115,
       "eval_loss": 7.129258632659912,
-      "eval_runtime": 13.6697,
-      "eval_samples_per_second": 43.6,
-      "eval_steps_per_second": 5.487,
+      "eval_runtime": 20.9973,
+      "eval_samples_per_second": 28.385,
+      "eval_steps_per_second": 3.572,
       "step": 1
     },
     {
@@ -32,14 +32,14 @@
     },
     {
       "epoch": 0.008480565371024734,
-      "grad_norm": 5.114774703979492,
+      "grad_norm": 5.082520484924316,
       "learning_rate": 2e-05,
       "loss": 7.4669,
       "step": 3
     },
     {
       "epoch": 0.011307420494699646,
-      "grad_norm": 4.991275310516357,
+      "grad_norm": 4.920047760009766,
       "learning_rate": 4e-05,
       "loss": 7.1521,
       "step": 4
@@ -48,144 +48,144 @@
       "epoch": 0.014134275618374558,
       "grad_norm": NaN,
       "learning_rate": 4e-05,
-      "loss": 7.7429,
+      "loss": 7.7426,
       "step": 5
     },
     {
       "epoch": 0.014134275618374558,
-      "eval_loss": 7.097643852233887,
-      "eval_runtime": 20.7137,
-      "eval_samples_per_second": 28.773,
-      "eval_steps_per_second": 3.621,
+      "eval_loss": 7.097348213195801,
+      "eval_runtime": 20.8903,
+      "eval_samples_per_second": 28.53,
+      "eval_steps_per_second": 3.59,
       "step": 5
     },
     {
       "epoch": 0.01696113074204947,
-      "grad_norm": 4.559197902679443,
+      "grad_norm": 4.500086784362793,
       "learning_rate": 6e-05,
-      "loss": 7.1861,
+      "loss": 7.1859,
       "step": 6
     },
     {
       "epoch": 0.019787985865724382,
-      "grad_norm": 3.6284019947052,
+      "grad_norm": 3.6024482250213623,
       "learning_rate": 8e-05,
-      "loss": 6.9673,
+      "loss": 6.9672,
       "step": 7
     },
     {
       "epoch": 0.022614840989399292,
-      "grad_norm": 4.3491082191467285,
+      "grad_norm": 4.331362724304199,
       "learning_rate": 0.0001,
-      "loss": 7.2919,
+      "loss": 7.2915,
       "step": 8
     },
     {
       "epoch": 0.025441696113074206,
-      "grad_norm": 3.08682918548584,
+      "grad_norm": 3.071237564086914,
       "learning_rate": 0.00012,
-      "loss": 7.258,
+      "loss": 7.2579,
       "step": 9
     },
     {
       "epoch": 0.028268551236749116,
-      "grad_norm": 3.268092393875122,
+      "grad_norm": 3.218928813934326,
       "learning_rate": 0.00014,
-      "loss": 7.0434,
+      "loss": 7.0417,
       "step": 10
     },
     {
       "epoch": 0.028268551236749116,
-      "eval_loss": 6.333721160888672,
-      "eval_runtime": 20.9199,
-      "eval_samples_per_second": 28.49,
-      "eval_steps_per_second": 3.585,
+      "eval_loss": 6.332116603851318,
+      "eval_runtime": 20.7973,
+      "eval_samples_per_second": 28.658,
+      "eval_steps_per_second": 3.606,
       "step": 10
     },
     {
       "epoch": 0.03109540636042403,
-      "grad_norm": 2.161461114883423,
+      "grad_norm": 2.1495420932769775,
       "learning_rate": 0.00016,
-      "loss": 6.1538,
+      "loss": 6.1519,
       "step": 11
     },
     {
       "epoch": 0.03392226148409894,
-      "grad_norm": 2.0395100116729736,
+      "grad_norm": 2.033008337020874,
       "learning_rate": 0.00018,
-      "loss": 6.6144,
+      "loss": 6.6133,
       "step": 12
     },
     {
       "epoch": 0.03674911660777385,
-      "grad_norm": 2.263904094696045,
+      "grad_norm": 2.2604849338531494,
       "learning_rate": 0.0002,
-      "loss": 6.1712,
+      "loss": 6.1704,
       "step": 13
     },
     {
       "epoch": 0.039575971731448764,
-      "grad_norm": 1.9835869073867798,
+      "grad_norm": 1.981244683265686,
       "learning_rate": 0.00019510565162951537,
-      "loss": 6.0672,
+      "loss": 6.0664,
       "step": 14
     },
     {
       "epoch": 0.04240282685512368,
-      "grad_norm": 1.9458274841308594,
+      "grad_norm": 1.9168858528137207,
       "learning_rate": 0.00018090169943749476,
-      "loss": 5.7719,
+      "loss": 5.7705,
       "step": 15
     },
     {
       "epoch": 0.04240282685512368,
-      "eval_loss": 5.498767375946045,
-      "eval_runtime": 20.8896,
-      "eval_samples_per_second": 28.531,
-      "eval_steps_per_second": 3.59,
+      "eval_loss": 5.497771263122559,
+      "eval_runtime": 20.7925,
+      "eval_samples_per_second": 28.664,
+      "eval_steps_per_second": 3.607,
       "step": 15
     },
     {
       "epoch": 0.045229681978798585,
-      "grad_norm": 1.6456574201583862,
+      "grad_norm": 1.6251788139343262,
       "learning_rate": 0.00015877852522924732,
-      "loss": 5.6195,
+      "loss": 5.6184,
       "step": 16
     },
     {
       "epoch": 0.0480565371024735,
-      "grad_norm": 1.413881540298462,
+      "grad_norm": 1.4069232940673828,
       "learning_rate": 0.00013090169943749476,
-      "loss": 5.3649,
+      "loss": 5.3642,
       "step": 17
     },
     {
       "epoch": 0.05088339222614841,
-      "grad_norm": 1.8699934482574463,
+      "grad_norm": 1.9275850057601929,
       "learning_rate": 0.0001,
-      "loss": 5.6544,
+      "loss": 5.6553,
       "step": 18
     },
     {
       "epoch": 0.05371024734982332,
-      "grad_norm": 2.320040464401245,
+      "grad_norm": 2.3857240676879883,
       "learning_rate": 6.909830056250527e-05,
-      "loss": 5.3966,
+      "loss": 5.3992,
       "step": 19
     },
     {
       "epoch": 0.05653710247349823,
-      "grad_norm": 1.5797550678253174,
+      "grad_norm": 1.5103741884231567,
       "learning_rate": 4.12214747707527e-05,
-      "loss": 5.695,
+      "loss": 5.6954,
       "step": 20
     },
     {
       "epoch": 0.05653710247349823,
-      "eval_loss": 5.102807998657227,
-      "eval_runtime": 20.9456,
-      "eval_samples_per_second": 28.455,
-      "eval_steps_per_second": 3.581,
+      "eval_loss": 5.104593276977539,
+      "eval_runtime": 11.9721,
+      "eval_samples_per_second": 49.783,
+      "eval_steps_per_second": 6.265,
       "step": 20
     }
   ],

last-checkpoint/training_args.bin

@@ -1,3 +1,3 @@
 version https://git-lfs.github.com/spec/v1
-oid sha256:676c5b957839737d77485c6332fc96de69385066f3eef0d74b47f677462f11dc
+oid sha256:0b55d7424e01d2da7c827809f0e75e814ca3307809331f5d1880db7934ea6a65
 size 6712