Training in progress, step 36, checkpoint

last-checkpoint/adapter_config.json

@@ -20,13 +20,13 @@
   "rank_pattern": {},
   "revision": null,
   "target_modules": [
-    "down_proj",
     "up_proj",
-    "k_proj",
+    "q_proj",
     "gate_proj",
+    "down_proj",
     "v_proj",
-    "o_proj",
-    "q_proj"
+    "k_proj",
+    "o_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:fc698f29b5bd4395eba500467e69f010931c71c5b2370377f772025a4cb34e76
+oid sha256:d232bf82608e68647edce48907e8bf0ca30351f4261536e91bd280996ffbccb5
 size 319876032

last-checkpoint/optimizer.pt

@@ -1,3 +1,3 @@
 version https://git-lfs.github.com/spec/v1
-oid sha256:b12ed38f2554f877e3dda7e6d31003b997e5ed3b760a92a5a57dc276279d868f
+oid sha256:dba6767df4e8b5b70d856a6a30f5b2e797e638f3c32d31b69b17564534ffc113
 size 640009682

last-checkpoint/trainer_state.json

@@ -1,5 +1,5 @@
 {
-  "best_metric": 0.2715974450111389,
+  "best_metric": 0.26949256658554077,
   "best_model_checkpoint": "miner_id_24/checkpoint-25",
   "epoch": 3.031578947368421,
   "eval_steps": 25,
@@ -10,7 +10,7 @@
   "log_history": [
     {
       "epoch": 0.08421052631578947,
-      "grad_norm": 5.253191947937012,
+      "grad_norm": 5.284335136413574,
       "learning_rate": 5e-05,
       "loss": 6.1391,
       "step": 1
@@ -18,262 +18,262 @@
     {
       "epoch": 0.08421052631578947,
       "eval_loss": 6.062305450439453,
-      "eval_runtime": 4.7344,
-      "eval_samples_per_second": 16.897,
-      "eval_steps_per_second": 2.112,
+      "eval_runtime": 4.7467,
+      "eval_samples_per_second": 16.854,
+      "eval_steps_per_second": 2.107,
       "step": 1
     },
     {
       "epoch": 0.16842105263157894,
-      "grad_norm": 10.75270938873291,
+      "grad_norm": 10.67638111114502,
       "learning_rate": 0.0001,
       "loss": 6.2114,
       "step": 2
     },
     {
       "epoch": 0.25263157894736843,
-      "grad_norm": 10.224058151245117,
+      "grad_norm": 10.675355911254883,
       "learning_rate": 9.978670881475172e-05,
-      "loss": 4.4507,
+      "loss": 4.4485,
       "step": 3
     },
     {
       "epoch": 0.3368421052631579,
-      "grad_norm": 4.666428089141846,
+      "grad_norm": 4.717409133911133,
       "learning_rate": 9.91486549841951e-05,
-      "loss": 5.8713,
+      "loss": 5.8646,
       "step": 4
     },
     {
       "epoch": 0.42105263157894735,
-      "grad_norm": 6.493696689605713,
+      "grad_norm": 6.8357834815979,
       "learning_rate": 9.809128215864097e-05,
-      "loss": 2.443,
+      "loss": 2.4496,
       "step": 5
     },
     {
       "epoch": 0.5052631578947369,
-      "grad_norm": 4.728562831878662,
+      "grad_norm": 4.734500885009766,
       "learning_rate": 9.662361147021779e-05,
-      "loss": 3.1197,
+      "loss": 3.132,
       "step": 6
     },
     {
       "epoch": 0.5894736842105263,
-      "grad_norm": 4.848873615264893,
+      "grad_norm": 4.709146976470947,
       "learning_rate": 9.475816456775313e-05,
-      "loss": 5.0906,
+      "loss": 5.0928,
       "step": 7
     },
     {
       "epoch": 0.6736842105263158,
-      "grad_norm": 3.4578592777252197,
+      "grad_norm": 3.474050998687744,
       "learning_rate": 9.251085678648072e-05,
-      "loss": 0.9384,
+      "loss": 0.9609,
       "step": 8
     },
     {
       "epoch": 0.7578947368421053,
-      "grad_norm": 2.790679693222046,
+      "grad_norm": 2.854706048965454,
       "learning_rate": 8.9900861364012e-05,
-      "loss": 2.7662,
+      "loss": 2.7636,
       "step": 9
     },
     {
       "epoch": 0.8421052631578947,
-      "grad_norm": 2.5729105472564697,
+      "grad_norm": 2.623455286026001,
       "learning_rate": 8.695044586103296e-05,
-      "loss": 2.3579,
+      "loss": 2.3552,
       "step": 10
     },
     {
       "epoch": 0.9263157894736842,
-      "grad_norm": 1.2617417573928833,
+      "grad_norm": 1.2726651430130005,
       "learning_rate": 8.368478218232787e-05,
-      "loss": 0.2876,
+      "loss": 0.2886,
       "step": 11
     },
     {
       "epoch": 1.0105263157894737,
-      "grad_norm": 2.1544504165649414,
+      "grad_norm": 2.1983087062835693,
       "learning_rate": 8.013173181896283e-05,
-      "loss": 1.4562,
+      "loss": 1.4454,
       "step": 12
     },
     {
       "epoch": 1.0947368421052632,
-      "grad_norm": 3.2978413105010986,
+      "grad_norm": 3.3619978427886963,
       "learning_rate": 7.63216081438678e-05,
-      "loss": 2.0684,
+      "loss": 2.0503,
       "step": 13
     },
     {
       "epoch": 1.1789473684210527,
-      "grad_norm": 1.267893671989441,
+      "grad_norm": 1.3169318437576294,
       "learning_rate": 7.228691778882693e-05,
-      "loss": 0.4268,
+      "loss": 0.4234,
       "step": 14
     },
     {
       "epoch": 1.263157894736842,
-      "grad_norm": 1.8825129270553589,
+      "grad_norm": 1.9388500452041626,
       "learning_rate": 6.806208330935766e-05,
-      "loss": 0.4462,
+      "loss": 0.435,
       "step": 15
     },
     {
       "epoch": 1.3473684210526315,
-      "grad_norm": 3.6305789947509766,
+      "grad_norm": 3.657085418701172,
       "learning_rate": 6.368314950360415e-05,
-      "loss": 0.9635,
+      "loss": 0.9435,
       "step": 16
     },
     {
       "epoch": 1.431578947368421,
-      "grad_norm": 0.539232611656189,
+      "grad_norm": 0.5060288906097412,
       "learning_rate": 5.918747589082853e-05,
-      "loss": 0.0226,
+      "loss": 0.0212,
       "step": 17
     },
     {
       "epoch": 1.5157894736842106,
-      "grad_norm": 3.158874750137329,
+      "grad_norm": 3.298569917678833,
       "learning_rate": 5.4613417973165106e-05,
-      "loss": 0.2222,
+      "loss": 0.2224,
       "step": 18
     },
     {
       "epoch": 1.6,
-      "grad_norm": 5.146640777587891,
+      "grad_norm": 5.861860752105713,
       "learning_rate": 5e-05,
-      "loss": 0.7041,
+      "loss": 0.7128,
       "step": 19
     },
     {
       "epoch": 1.6842105263157894,
-      "grad_norm": 0.28501152992248535,
+      "grad_norm": 0.3091553747653961,
       "learning_rate": 4.5386582026834906e-05,
-      "loss": 0.0074,
+      "loss": 0.0077,
       "step": 20
     },
     {
       "epoch": 1.768421052631579,
-      "grad_norm": 1.6088167428970337,
+      "grad_norm": 1.5811270475387573,
       "learning_rate": 4.0812524109171476e-05,
-      "loss": 0.2779,
+      "loss": 0.2771,
       "step": 21
     },
     {
       "epoch": 1.8526315789473684,
-      "grad_norm": 4.001155376434326,
+      "grad_norm": 3.9169883728027344,
       "learning_rate": 3.631685049639586e-05,
-      "loss": 0.389,
+      "loss": 0.3961,
       "step": 22
     },
     {
       "epoch": 1.936842105263158,
-      "grad_norm": 0.3760082423686981,
+      "grad_norm": 0.372968465089798,
       "learning_rate": 3.1937916690642356e-05,
-      "loss": 0.0252,
+      "loss": 0.0253,
       "step": 23
     },
     {
       "epoch": 2.0210526315789474,
-      "grad_norm": 0.9666143655776978,
+      "grad_norm": 1.0356941223144531,
       "learning_rate": 2.771308221117309e-05,
-      "loss": 0.2194,
+      "loss": 0.2199,
       "step": 24
     },
     {
       "epoch": 2.1052631578947367,
-      "grad_norm": 3.1850697994232178,
+      "grad_norm": 3.1314473152160645,
       "learning_rate": 2.3678391856132204e-05,
-      "loss": 0.4543,
+      "loss": 0.4549,
       "step": 25
     },
     {
       "epoch": 2.1052631578947367,
-      "eval_loss": 0.2715974450111389,
-      "eval_runtime": 4.2024,
-      "eval_samples_per_second": 19.037,
-      "eval_steps_per_second": 2.38,
+      "eval_loss": 0.26949256658554077,
+      "eval_runtime": 4.2099,
+      "eval_samples_per_second": 19.003,
+      "eval_steps_per_second": 2.375,
       "step": 25
     },
     {
       "epoch": 2.1894736842105265,
-      "grad_norm": 0.04599270597100258,
+      "grad_norm": 0.041909243911504745,
       "learning_rate": 1.9868268181037185e-05,
-      "loss": 0.0012,
+      "loss": 0.0011,
       "step": 26
     },
     {
       "epoch": 2.2736842105263158,
-      "grad_norm": 2.79744291305542,
+      "grad_norm": 2.73862886428833,
       "learning_rate": 1.631521781767214e-05,
-      "loss": 0.1034,
+      "loss": 0.1024,
       "step": 27
     },
     {
       "epoch": 2.3578947368421055,
-      "grad_norm": 2.700434684753418,
+      "grad_norm": 2.667339563369751,
       "learning_rate": 1.3049554138967051e-05,
-      "loss": 0.4264,
+      "loss": 0.4289,
       "step": 28
     },
     {
       "epoch": 2.442105263157895,
-      "grad_norm": 0.027330536395311356,
+      "grad_norm": 0.023501254618167877,
       "learning_rate": 1.0099138635988026e-05,
-      "loss": 0.0008,
+      "loss": 0.0007,
       "step": 29
     },
     {
       "epoch": 2.526315789473684,
-      "grad_norm": 4.523248195648193,
+      "grad_norm": 4.1375651359558105,
       "learning_rate": 7.489143213519301e-06,
-      "loss": 0.175,
+      "loss": 0.1674,
       "step": 30
     },
     {
       "epoch": 2.610526315789474,
-      "grad_norm": 2.505566358566284,
+      "grad_norm": 2.5133378505706787,
       "learning_rate": 5.241835432246889e-06,
-      "loss": 0.2446,
+      "loss": 0.2455,
       "step": 31
     },
     {
       "epoch": 2.694736842105263,
-      "grad_norm": 0.018715698271989822,
+      "grad_norm": 0.01599811762571335,
       "learning_rate": 3.376388529782215e-06,
       "loss": 0.0006,
       "step": 32
     },
     {
       "epoch": 2.7789473684210524,
-      "grad_norm": 4.404769420623779,
+      "grad_norm": 4.188090801239014,
       "learning_rate": 1.908717841359048e-06,
-      "loss": 0.3044,
+      "loss": 0.2993,
       "step": 33
     },
     {
       "epoch": 2.863157894736842,
-      "grad_norm": 1.9680789709091187,
+      "grad_norm": 1.9685863256454468,
       "learning_rate": 8.513450158049108e-07,
-      "loss": 0.1945,
+      "loss": 0.1988,
       "step": 34
     },
     {
       "epoch": 2.9473684210526314,
-      "grad_norm": 2.3198118209838867,
+      "grad_norm": 2.153817653656006,
       "learning_rate": 2.1329118524827662e-07,
-      "loss": 0.0721,
+      "loss": 0.0703,
       "step": 35
     },
     {
       "epoch": 3.031578947368421,
-      "grad_norm": 2.005392074584961,
+      "grad_norm": 1.772990345954895,
       "learning_rate": 0.0,
-      "loss": 0.1797,
+      "loss": 0.1774,
       "step": 36
     }
   ],

last-checkpoint/training_args.bin

@@ -1,3 +1,3 @@
 version https://git-lfs.github.com/spec/v1
-oid sha256:f4e8ae8d03d1a5cb5cfbeabe4bf2a0a9151d25911ec9c8c250d423b12fe826d4
+oid sha256:62fe8c0855098ea055d918c702f52c1e692bb20ceaf0bb7577a4d6f246e27023
 size 6776