Training in progress, step 3000, checkpoint

last-checkpoint/adapter_model.safetensors

@@ -1,3 +1,3 @@
 version https://git-lfs.github.com/spec/v1
-oid sha256:c2c2d7f612913d93a077fdb78b3b8d2e485d7a3d9b8edf61f4b3878a29927581
+oid sha256:c5a6fe2d4b3225fd2a69dde200b394c96d6cf4eca02bc329d2419c8c1466f332
 size 335605144

last-checkpoint/optimizer.pt

@@ -1,3 +1,3 @@
 version https://git-lfs.github.com/spec/v1
-oid sha256:c42dc892637ebac98711496849d8c72d68e7b4aac93a8d4107267a47139ac80a
+oid sha256:f22d4a12b719d135438b26a2f413193a9867d93caceb92688f5f079f11cfd624
 size 671467026

last-checkpoint/rng_state.pth

@@ -1,3 +1,3 @@
 version https://git-lfs.github.com/spec/v1
-oid sha256:150188ffb5cc1748a7977a1870ef1e2a294b0577c956793d6fdde54fa8cc8bf0
+oid sha256:b4bb882287d6976c062c089a7686086b15f0601112a612821d81ac9776ae1a44
 size 14244

last-checkpoint/scheduler.pt

@@ -1,3 +1,3 @@
 version https://git-lfs.github.com/spec/v1
-oid sha256:ac658b35d124fb9bc77fb9f0ca508648718ba76b0e5a26a1e7feca7ebaaaca02
+oid sha256:cb5106a3d40df1efd273b9eb04f3299504d470352eb1b64d92b1e010e88c8900
 size 1064

last-checkpoint/trainer_state.json

@@ -1,9 +1,9 @@
 {
-  "best_metric": 0.011984162032604218,
-  "best_model_checkpoint": "./mistral-magyar-portas-results/checkpoint-2000",
-  "epoch": 1.567398119122257,
+  "best_metric": 0.011125968769192696,
+  "best_model_checkpoint": "./mistral-magyar-portas-results/checkpoint-2500",
+  "epoch": 2.3510971786833856,
   "eval_steps": 500,
-  "global_step": 2000,
+  "global_step": 3000,
   "is_hyper_param_search": false,
   "is_local_process_zero": true,
   "is_world_process_zero": true,
@@ -319,6 +319,162 @@
       "eval_samples_per_second": 2.586,
       "eval_steps_per_second": 1.294,
       "step": 2000
+    },
+    {
+      "epoch": 1.6065830721003134,
+      "grad_norm": 0.09814453125,
+      "learning_rate": 9.275910610392104e-05,
+      "loss": 0.0091,
+      "step": 2050
+    },
+    {
+      "epoch": 1.64576802507837,
+      "grad_norm": 0.13671875,
+      "learning_rate": 8.856432702880984e-05,
+      "loss": 0.0096,
+      "step": 2100
+    },
+    {
+      "epoch": 1.6849529780564263,
+      "grad_norm": 0.07568359375,
+      "learning_rate": 8.43898474574128e-05,
+      "loss": 0.0092,
+      "step": 2150
+    },
+    {
+      "epoch": 1.7241379310344827,
+      "grad_norm": 0.046875,
+      "learning_rate": 8.02430775232462e-05,
+      "loss": 0.0084,
+      "step": 2200
+    },
+    {
+      "epoch": 1.7633228840125392,
+      "grad_norm": 0.062255859375,
+      "learning_rate": 7.61313781723508e-05,
+      "loss": 0.0087,
+      "step": 2250
+    },
+    {
+      "epoch": 1.8025078369905956,
+      "grad_norm": 0.06591796875,
+      "learning_rate": 7.206204809685029e-05,
+      "loss": 0.0086,
+      "step": 2300
+    },
+    {
+      "epoch": 1.841692789968652,
+      "grad_norm": 0.052978515625,
+      "learning_rate": 6.804231077901733e-05,
+      "loss": 0.0085,
+      "step": 2350
+    },
+    {
+      "epoch": 1.8808777429467085,
+      "grad_norm": 0.07421875,
+      "learning_rate": 6.407930166884409e-05,
+      "loss": 0.0094,
+      "step": 2400
+    },
+    {
+      "epoch": 1.9200626959247649,
+      "grad_norm": 0.1044921875,
+      "learning_rate": 6.018005551787984e-05,
+      "loss": 0.008,
+      "step": 2450
+    },
+    {
+      "epoch": 1.9592476489028212,
+      "grad_norm": 0.05615234375,
+      "learning_rate": 5.635149389181855e-05,
+      "loss": 0.0088,
+      "step": 2500
+    },
+    {
+      "epoch": 1.9592476489028212,
+      "eval_loss": 0.011125968769192696,
+      "eval_runtime": 438.9494,
+      "eval_samples_per_second": 2.586,
+      "eval_steps_per_second": 1.294,
+      "step": 2500
+    },
+    {
+      "epoch": 1.9984326018808778,
+      "grad_norm": 0.038330078125,
+      "learning_rate": 5.260041288400284e-05,
+      "loss": 0.0092,
+      "step": 2550
+    },
+    {
+      "epoch": 2.0376175548589344,
+      "grad_norm": 0.056396484375,
+      "learning_rate": 4.893347105165468e-05,
+      "loss": 0.0064,
+      "step": 2600
+    },
+    {
+      "epoch": 2.0768025078369905,
+      "grad_norm": 0.0517578125,
+      "learning_rate": 4.535717759624677e-05,
+      "loss": 0.0065,
+      "step": 2650
+    },
+    {
+      "epoch": 2.115987460815047,
+      "grad_norm": 0.049560546875,
+      "learning_rate": 4.187788080899591e-05,
+      "loss": 0.0064,
+      "step": 2700
+    },
+    {
+      "epoch": 2.1551724137931036,
+      "grad_norm": 0.049072265625,
+      "learning_rate": 3.8501756801988675e-05,
+      "loss": 0.0062,
+      "step": 2750
+    },
+    {
+      "epoch": 2.19435736677116,
+      "grad_norm": 0.0908203125,
+      "learning_rate": 3.5234798544942914e-05,
+      "loss": 0.0063,
+      "step": 2800
+    },
+    {
+      "epoch": 2.2335423197492164,
+      "grad_norm": 0.0595703125,
+      "learning_rate": 3.208280522706602e-05,
+      "loss": 0.0066,
+      "step": 2850
+    },
+    {
+      "epoch": 2.2727272727272725,
+      "grad_norm": 0.044677734375,
+      "learning_rate": 2.9051371962893358e-05,
+      "loss": 0.0063,
+      "step": 2900
+    },
+    {
+      "epoch": 2.311912225705329,
+      "grad_norm": 0.05224609375,
+      "learning_rate": 2.6145879860380773e-05,
+      "loss": 0.0065,
+      "step": 2950
+    },
+    {
+      "epoch": 2.3510971786833856,
+      "grad_norm": 0.0908203125,
+      "learning_rate": 2.337148646888061e-05,
+      "loss": 0.0065,
+      "step": 3000
+    },
+    {
+      "epoch": 2.3510971786833856,
+      "eval_loss": 0.01185206975787878,
+      "eval_runtime": 438.7429,
+      "eval_samples_per_second": 2.587,
+      "eval_steps_per_second": 1.295,
+      "step": 3000
     }
   ],
   "logging_steps": 50,
@@ -326,7 +482,7 @@
   "num_input_tokens_seen": 0,
   "num_train_epochs": 3,
   "save_steps": 500,
-  "total_flos": 2.2359343890432e+18,
+  "total_flos": 3.3539015835648e+18,
   "train_batch_size": 2,
   "trial_name": null,
   "trial_params": null