SFT checkpoint step 24

checkpoints/checkpoint-24/adapter_model.safetensors

@@ -1,3 +1,3 @@
 version https://git-lfs.github.com/spec/v1
-oid sha256:63e5deaea9432e094c15bb64c92fa81cbba3041810db39c721e22096a1702974
+oid sha256:0b606145073d15bc2135946e05c944e20860deae934882b9fdb438337056243b
 size 204500912

checkpoints/checkpoint-24/optimizer.pt

@@ -1,3 +1,3 @@
 version https://git-lfs.github.com/spec/v1
-oid sha256:5c539f10386e07d23265746ae848f134e6801e7ac62cbc5fecdea9f06b09838e
+oid sha256:250ec5c90b9926f38ce2814a51bf1b25af5843cfba9eb35ac72d18a21b221940
 size 620569547

checkpoints/checkpoint-24/scaler.pt

@@ -1,3 +1,3 @@
 version https://git-lfs.github.com/spec/v1
-oid sha256:6c8c484b6aeff8e48e491091e15944fd74e10a109ea0ef357d241072f6e7ea4a
+oid sha256:e67c4c81374c6f521df75510e2ff9132038e90aa5d6f4327585ee498a7c32acf
 size 1383

checkpoints/checkpoint-24/scheduler.pt

@@ -1,3 +1,3 @@
 version https://git-lfs.github.com/spec/v1
-oid sha256:cf03b2543d3b9777aee09aeaefb82e4e594fbd8d33431ba55fbfdf92efaf789a
+oid sha256:0688e184f56eee18176185f2f1a5a43c3abd2e7e0e3b23afd81829a9a6999304
 size 1465

checkpoints/checkpoint-24/time_state.json

@@ -1 +1 @@
-{"consumed_seconds": 263.88233065605164, "global_step": 24}
+{"consumed_seconds": 277.82182002067566, "global_step": 24}

checkpoints/checkpoint-24/trainer_state.json

@@ -17,15 +17,15 @@
       "num_input_tokens_seen": 1494,
       "num_tokens": 1494.0,
       "step": 1,
-      "train_runtime": 44.9864,
-      "train_tokens_per_second": 33.21
+      "train_runtime": 42.4619,
+      "train_tokens_per_second": 35.185
     },
     {
       "epoch": 0.06956521739130435,
       "eval_loss": 5.31978178024292,
       "eval_num_tokens": 4980.0,
-      "eval_runtime": 28.6663,
-      "eval_samples_per_second": 1.047,
+      "eval_runtime": 28.7144,
+      "eval_samples_per_second": 1.045,
       "eval_steps_per_second": 0.279,
       "num_input_tokens_seen": 4980,
       "step": 4
@@ -34,71 +34,71 @@
       "epoch": 0.1391304347826087,
       "eval_loss": 5.31978178024292,
       "eval_num_tokens": 9004.0,
-      "eval_runtime": 3.5559,
-      "eval_samples_per_second": 8.437,
-      "eval_steps_per_second": 2.25,
+      "eval_runtime": 3.4006,
+      "eval_samples_per_second": 8.822,
+      "eval_steps_per_second": 2.353,
       "num_input_tokens_seen": 9004,
       "step": 8
     },
     {
       "epoch": 0.17391304347826086,
-      "grad_norm": 28.138591766357422,
+      "grad_norm": 28.1390438079834,
       "learning_rate": 0.0,
       "loss": 2.5477417839898004,
       "num_input_tokens_seen": 11145,
       "num_tokens": 11145.0,
       "step": 10,
-      "train_runtime": 145.0656,
-      "train_tokens_per_second": 76.827
+      "train_runtime": 146.4448,
+      "train_tokens_per_second": 76.104
     },
     {
       "epoch": 0.20869565217391303,
-      "eval_loss": 1.795594573020935,
+      "eval_loss": 1.7850792407989502,
       "eval_num_tokens": 13293.0,
-      "eval_runtime": 3.5484,
-      "eval_samples_per_second": 8.455,
-      "eval_steps_per_second": 2.255,
+      "eval_runtime": 3.4816,
+      "eval_samples_per_second": 8.617,
+      "eval_steps_per_second": 2.298,
       "num_input_tokens_seen": 13293,
       "step": 12
     },
     {
       "epoch": 0.2782608695652174,
-      "eval_loss": 0.5860968232154846,
+      "eval_loss": 0.5965744256973267,
       "eval_num_tokens": 17742.0,
-      "eval_runtime": 3.4658,
-      "eval_samples_per_second": 8.656,
-      "eval_steps_per_second": 2.308,
+      "eval_runtime": 3.4391,
+      "eval_samples_per_second": 8.723,
+      "eval_steps_per_second": 2.326,
       "num_input_tokens_seen": 17742,
       "step": 16
     },
     {
       "epoch": 0.34782608695652173,
-      "grad_norm": 7.353201389312744,
-      "learning_rate": 0.0001442909649383465,
-      "loss": 0.8008382797241211,
+      "grad_norm": 6.518080711364746,
+      "learning_rate": 0.0001425726650926814,
+      "loss": 0.7919160842895507,
       "num_input_tokens_seen": 20264,
       "num_tokens": 20264.0,
       "step": 20,
-      "train_runtime": 227.664,
-      "train_tokens_per_second": 89.008
+      "train_runtime": 241.9804,
+      "train_tokens_per_second": 83.742
     },
     {
       "epoch": 0.34782608695652173,
-      "eval_loss": 0.39688172936439514,
+      "eval_loss": 0.6307523250579834,
       "eval_num_tokens": 20264.0,
-      "eval_runtime": 3.494,
-      "eval_samples_per_second": 8.586,
-      "eval_steps_per_second": 2.29,
+      "eval_runtime": 3.512,
+      "eval_samples_per_second": 8.542,
+      "eval_steps_per_second": 2.278,
       "num_input_tokens_seen": 20264,
       "step": 20
     },
     {
       "epoch": 0.41739130434782606,
-      "eval_loss": 0.5452550053596497,
+      "eval_loss": 0.3135659396648407,
       "eval_num_tokens": 24990.0,
-      "eval_runtime": 3.5856,
-      "eval_samples_per_second": 8.367,
-      "eval_steps_per_second": 2.231,
+      "eval_runtime": 3.4341,
+      "eval_samples_per_second": 8.736,
+      "eval_steps_per_second": 2.33,
       "num_input_tokens_seen": 24990,
       "step": 24
     }