Training in progress, step 20, checkpoint

last-checkpoint/adapter_config.json

@@ -21,12 +21,12 @@
   "revision": null,
   "target_modules": [
     "o_proj",
-    "down_proj",
     "k_proj",
     "v_proj",
     "q_proj",
-    "up_proj",
-    "gate_proj"
+    "gate_proj",
+    "down_proj",
+    "up_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:c129f22828d9a11a70b1b6a912223b1c18badbd13ded6a14546603a09d84074b
+oid sha256:3ffdcf9f926d8be928a0034929041b9c7f54b4c50c234ff9ef11d235f1552cb0
 size 50624

last-checkpoint/optimizer.pt

@@ -1,3 +1,3 @@
 version https://git-lfs.github.com/spec/v1
-oid sha256:08421e6f0b59e1bd7f90d8d5d4a01b9ed4e59046adf26cfad2d2fdd7067acd3f
+oid sha256:297b9ecd3a6c7081a6ffcb673aa55e7d02260ef54fa8d65a57beac6937393a81
 size 111142

last-checkpoint/trainer_state.json

@@ -10,7 +10,7 @@
   "log_history": [
     {
       "epoch": 0.0003251239535072746,
-      "grad_norm": 0.021640103310346603,
+      "grad_norm": 0.024652473628520966,
       "learning_rate": 1e-05,
       "loss": 11.7645,
       "step": 1
@@ -18,174 +18,174 @@
     {
       "epoch": 0.0003251239535072746,
       "eval_loss": 11.76447582244873,
-      "eval_runtime": 249.3031,
-      "eval_samples_per_second": 41.56,
-      "eval_steps_per_second": 20.782,
+      "eval_runtime": 276.626,
+      "eval_samples_per_second": 37.455,
+      "eval_steps_per_second": 18.729,
       "step": 1
     },
     {
       "epoch": 0.0006502479070145493,
-      "grad_norm": 0.024706723168492317,
+      "grad_norm": 0.027672940865159035,
       "learning_rate": 2e-05,
       "loss": 11.7658,
       "step": 2
     },
     {
       "epoch": 0.000975371860521824,
-      "grad_norm": 0.022929754108190536,
+      "grad_norm": 0.025643622502684593,
       "learning_rate": 3e-05,
       "loss": 11.7649,
       "step": 3
     },
     {
       "epoch": 0.0013004958140290985,
-      "grad_norm": 0.022171195596456528,
+      "grad_norm": 0.025678519159555435,
       "learning_rate": 4e-05,
       "loss": 11.7648,
       "step": 4
     },
     {
       "epoch": 0.0016256197675363732,
-      "grad_norm": 0.02208411693572998,
+      "grad_norm": 0.02572811394929886,
       "learning_rate": 5e-05,
       "loss": 11.7649,
       "step": 5
     },
     {
       "epoch": 0.0016256197675363732,
-      "eval_loss": 11.764402389526367,
-      "eval_runtime": 249.1579,
-      "eval_samples_per_second": 41.584,
-      "eval_steps_per_second": 20.794,
+      "eval_loss": 11.764395713806152,
+      "eval_runtime": 276.5749,
+      "eval_samples_per_second": 37.462,
+      "eval_steps_per_second": 18.733,
       "step": 5
     },
     {
       "epoch": 0.001950743721043648,
-      "grad_norm": 0.02443164400756359,
+      "grad_norm": 0.026358768343925476,
       "learning_rate": 6e-05,
       "loss": 11.7634,
       "step": 6
     },
     {
       "epoch": 0.0022758676745509225,
-      "grad_norm": 0.023470092564821243,
+      "grad_norm": 0.026500564068555832,
       "learning_rate": 7e-05,
       "loss": 11.7654,
       "step": 7
     },
     {
       "epoch": 0.002600991628058197,
-      "grad_norm": 0.023716744035482407,
+      "grad_norm": 0.02642131596803665,
       "learning_rate": 8e-05,
       "loss": 11.7644,
       "step": 8
     },
     {
       "epoch": 0.002926115581565472,
-      "grad_norm": 0.022970277816057205,
+      "grad_norm": 0.02585265040397644,
       "learning_rate": 9e-05,
       "loss": 11.7658,
       "step": 9
     },
     {
       "epoch": 0.0032512395350727465,
-      "grad_norm": 0.022244542837142944,
+      "grad_norm": 0.024854907765984535,
       "learning_rate": 0.0001,
       "loss": 11.7641,
       "step": 10
     },
     {
       "epoch": 0.0032512395350727465,
-      "eval_loss": 11.764139175415039,
-      "eval_runtime": 248.9949,
-      "eval_samples_per_second": 41.611,
-      "eval_steps_per_second": 20.808,
+      "eval_loss": 11.76411247253418,
+      "eval_runtime": 276.5048,
+      "eval_samples_per_second": 37.471,
+      "eval_steps_per_second": 18.737,
       "step": 10
     },
     {
       "epoch": 0.003576363488580021,
-      "grad_norm": 0.024796368554234505,
+      "grad_norm": 0.02757527120411396,
       "learning_rate": 9.755282581475769e-05,
-      "loss": 11.7653,
+      "loss": 11.7652,
       "step": 11
     },
     {
       "epoch": 0.003901487442087296,
-      "grad_norm": 0.024691808968782425,
+      "grad_norm": 0.028088003396987915,
       "learning_rate": 9.045084971874738e-05,
-      "loss": 11.7642,
+      "loss": 11.7641,
       "step": 12
     },
     {
       "epoch": 0.00422661139559457,
-      "grad_norm": 0.026635609567165375,
+      "grad_norm": 0.02876427210867405,
       "learning_rate": 7.938926261462366e-05,
-      "loss": 11.7643,
+      "loss": 11.7642,
       "step": 13
     },
     {
       "epoch": 0.004551735349101845,
-      "grad_norm": 0.024080852046608925,
+      "grad_norm": 0.027249282225966454,
       "learning_rate": 6.545084971874738e-05,
-      "loss": 11.7639,
+      "loss": 11.7638,
       "step": 14
     },
     {
       "epoch": 0.00487685930260912,
-      "grad_norm": 0.021976996213197708,
+      "grad_norm": 0.025040265172719955,
       "learning_rate": 5e-05,
       "loss": 11.763,
       "step": 15
     },
     {
       "epoch": 0.00487685930260912,
-      "eval_loss": 11.7637939453125,
-      "eval_runtime": 248.9817,
-      "eval_samples_per_second": 41.613,
-      "eval_steps_per_second": 20.809,
+      "eval_loss": 11.763741493225098,
+      "eval_runtime": 276.3837,
+      "eval_samples_per_second": 37.488,
+      "eval_steps_per_second": 18.746,
       "step": 15
     },
     {
       "epoch": 0.005201983256116394,
-      "grad_norm": 0.021976066753268242,
+      "grad_norm": 0.024846818298101425,
       "learning_rate": 3.4549150281252636e-05,
       "loss": 11.7639,
       "step": 16
     },
     {
       "epoch": 0.005527107209623669,
-      "grad_norm": 0.0250637736171484,
+      "grad_norm": 0.027951408177614212,
       "learning_rate": 2.061073738537635e-05,
-      "loss": 11.7633,
+      "loss": 11.7632,
       "step": 17
     },
     {
       "epoch": 0.005852231163130944,
-      "grad_norm": 0.024703800678253174,
+      "grad_norm": 0.02765224315226078,
       "learning_rate": 9.549150281252633e-06,
-      "loss": 11.7632,
+      "loss": 11.7631,
       "step": 18
     },
     {
       "epoch": 0.006177355116638218,
-      "grad_norm": 0.02371840551495552,
+      "grad_norm": 0.027219220995903015,
       "learning_rate": 2.4471741852423237e-06,
-      "loss": 11.7637,
+      "loss": 11.7636,
       "step": 19
     },
     {
       "epoch": 0.006502479070145493,
-      "grad_norm": 0.027609504759311676,
+      "grad_norm": 0.031280361115932465,
       "learning_rate": 0.0,
-      "loss": 11.764,
+      "loss": 11.7639,
       "step": 20
     },
     {
       "epoch": 0.006502479070145493,
-      "eval_loss": 11.763700485229492,
-      "eval_runtime": 248.9974,
-      "eval_samples_per_second": 41.611,
-      "eval_steps_per_second": 20.807,
+      "eval_loss": 11.763640403747559,
+      "eval_runtime": 276.2913,
+      "eval_samples_per_second": 37.5,
+      "eval_steps_per_second": 18.752,
       "step": 20
     }
   ],

last-checkpoint/training_args.bin

@@ -1,3 +1,3 @@
 version https://git-lfs.github.com/spec/v1
-oid sha256:9a413dd30c82bf1458b4c9aa8cd9fc6df506b56de5b15f10d2e2a254ff755ea0
+oid sha256:e17a6c83dec956fb9efedd2c17093957afde82b766a4186ec8d108b6cbea110d
 size 6776