Training in progress, step 1500, checkpoint

checkpoint-1500/config.json

@@ -16,7 +16,7 @@
   "intermediate_size": 3072,
   "layer_norm_eps": 1e-12,
   "max_blocks": 200,
-  "max_position_embeddings": 1801,
+  "max_position_embeddings": 1805,
   "max_relative_position_embeddings": 8,
   "model_type": "longelm",
   "node_size": 1,

checkpoint-1500/optimizer.pt

@@ -1,3 +1,3 @@
 version https://git-lfs.github.com/spec/v1
-oid sha256:e310a76c30e78eb5432c1c445292d9d464545794a9971682fd3709480b32e7a0
-size 1010407418
+oid sha256:a02841bfbf4f9b2cb3a583f435f75799c5eabc648a03a2322c59e3171a8df39b
+size 1010431994

checkpoint-1500/pytorch_model.bin

@@ -1,3 +1,3 @@
 version https://git-lfs.github.com/spec/v1
-oid sha256:bb612e933edc948fabaa46ffa9a8c1d5147d0038022b2e791db5bdcb06795f91
-size 505188394
+oid sha256:8d034f4700900cc033db4b94061e15a1d041ffac4e2e73ab9753e5b9fd1d5538
+size 505200682

checkpoint-1500/trainer_state.json

@@ -9,27 +9,27 @@
   "is_world_process_zero": true,
   "log_history": [
     {
-      "ep_loss": 4.7343,
+      "ep_loss": 4.4238,
       "epoch": 0.04,
       "learning_rate": 2.5e-05,
-      "loss": 11.6218,
-      "mlm_loss": 6.8875,
+      "loss": 11.1467,
+      "mlm_loss": 6.7229,
       "step": 500
     },
     {
-      "ep_loss": 0.4315,
+      "ep_loss": 0.4305,
       "epoch": 0.08,
       "learning_rate": 5e-05,
-      "loss": 3.0285,
-      "mlm_loss": 2.5969,
+      "loss": 2.9841,
+      "mlm_loss": 2.5537,
       "step": 1000
     },
     {
-      "ep_loss": 0.3811,
+      "ep_loss": 0.3724,
       "epoch": 0.12,
       "learning_rate": 7.5e-05,
-      "loss": 2.2571,
-      "mlm_loss": 1.8761,
+      "loss": 2.2371,
+      "mlm_loss": 1.8647,
       "step": 1500
     }
   ],

checkpoint-1500/training_args.bin

@@ -1,3 +1,3 @@
 version https://git-lfs.github.com/spec/v1
-oid sha256:cee56d4f38339eac5ecf098381d5222b3783455863547a17ac0df67c8caa7d15
+oid sha256:b26e03be70ebe5d40a81a81be94421c0578a9d8742b065bc029501df48632957
 size 4664