Training in progress, step 21000, checkpoint

last-checkpoint/model.safetensors

@@ -1,3 +1,3 @@
 version https://git-lfs.github.com/spec/v1
-oid sha256:5cf815e6e63a17cdb9abef6d6a034aa11cda747870482f601096bc7eaaced8b8
+oid sha256:76962f31b592fa8deacccb979f42e9210c0c2e3bf0764d57dee6473a4df753a6
 size 301235464

last-checkpoint/optimizer.pt

@@ -1,3 +1,3 @@
 version https://git-lfs.github.com/spec/v1
-oid sha256:ef8663185aa879cdc6c96017209b8ea400dc73cec0abf37f93a6946d8b844fc7
+oid sha256:425d92660fd73540ddd53e4612e3b0f78452740315b4f16696c04af26bbcfffa
 size 602335994

last-checkpoint/rng_state.pth

@@ -1,3 +1,3 @@
 version https://git-lfs.github.com/spec/v1
-oid sha256:0a7131d69ba7156fcfd8e4dfc557d7e121df2a1c1d83a47239788bb8dbd92b17
+oid sha256:6be325dc8d081573ecb7d82e811226f573dbeee4a5f3df2685ea5ac22d875758
 size 14244

last-checkpoint/scheduler.pt

@@ -1,3 +1,3 @@
 version https://git-lfs.github.com/spec/v1
-oid sha256:6662ae68d38995d5846f13e724946a2acb1395046b7d08977dde3dab733945c0
+oid sha256:1b368897bc9733fded7264e8f195c047c51921c4d323bbc9cd4cc617a19adfef
 size 1064

last-checkpoint/trainer_state.json

@@ -2,9 +2,9 @@
   "best_global_step": null,
   "best_metric": null,
   "best_model_checkpoint": null,
-  "epoch": 0.09540050323765457,
+  "epoch": 0.1001705283995373,
   "eval_steps": 500,
-  "global_step": 20000,
+  "global_step": 21000,
   "is_hyper_param_search": false,
   "is_local_process_zero": true,
   "is_world_process_zero": true,
@@ -3568,11 +3568,189 @@
       "eval_steps_per_second": 24.709,
       "num_input_tokens_seen": 5242880000,
       "step": 20000
+    },
+    {
+      "epoch": 0.09563900449574872,
+      "grad_norm": 0.17750214040279388,
+      "learning_rate": 0.001,
+      "loss": 2.7,
+      "num_input_tokens_seen": 5255987200,
+      "step": 20050
+    },
+    {
+      "epoch": 0.09587750575384285,
+      "grad_norm": 0.16490615904331207,
+      "learning_rate": 0.001,
+      "loss": 2.7188,
+      "num_input_tokens_seen": 5269094400,
+      "step": 20100
+    },
+    {
+      "epoch": 0.09611600701193698,
+      "grad_norm": 0.20347309112548828,
+      "learning_rate": 0.001,
+      "loss": 2.7034,
+      "num_input_tokens_seen": 5282201600,
+      "step": 20150
+    },
+    {
+      "epoch": 0.09635450827003113,
+      "grad_norm": 0.19717667996883392,
+      "learning_rate": 0.001,
+      "loss": 2.6864,
+      "num_input_tokens_seen": 5295308800,
+      "step": 20200
+    },
+    {
+      "epoch": 0.09659300952812526,
+      "grad_norm": 0.17054997384548187,
+      "learning_rate": 0.001,
+      "loss": 2.7068,
+      "num_input_tokens_seen": 5308416000,
+      "step": 20250
+    },
+    {
+      "epoch": 0.09683151078621939,
+      "grad_norm": 0.1771887093782425,
+      "learning_rate": 0.001,
+      "loss": 2.7037,
+      "num_input_tokens_seen": 5321523200,
+      "step": 20300
+    },
+    {
+      "epoch": 0.09707001204431354,
+      "grad_norm": 0.17556501924991608,
+      "learning_rate": 0.001,
+      "loss": 2.705,
+      "num_input_tokens_seen": 5334630400,
+      "step": 20350
+    },
+    {
+      "epoch": 0.09730851330240767,
+      "grad_norm": 0.1696256399154663,
+      "learning_rate": 0.001,
+      "loss": 2.7109,
+      "num_input_tokens_seen": 5347737600,
+      "step": 20400
+    },
+    {
+      "epoch": 0.09754701456050181,
+      "grad_norm": 0.18629619479179382,
+      "learning_rate": 0.001,
+      "loss": 2.7043,
+      "num_input_tokens_seen": 5360844800,
+      "step": 20450
+    },
+    {
+      "epoch": 0.09778551581859594,
+      "grad_norm": 0.18701018393039703,
+      "learning_rate": 0.001,
+      "loss": 2.7002,
+      "num_input_tokens_seen": 5373952000,
+      "step": 20500
+    },
+    {
+      "epoch": 0.09778551581859594,
+      "eval_loss": 2.587498903274536,
+      "eval_runtime": 51.1986,
+      "eval_samples_per_second": 97.659,
+      "eval_steps_per_second": 24.415,
+      "num_input_tokens_seen": 5373952000,
+      "step": 20500
+    },
+    {
+      "epoch": 0.09802401707669008,
+      "grad_norm": 0.1792842447757721,
+      "learning_rate": 0.001,
+      "loss": 2.7083,
+      "num_input_tokens_seen": 5387059200,
+      "step": 20550
+    },
+    {
+      "epoch": 0.09826251833478422,
+      "grad_norm": 0.18761058151721954,
+      "learning_rate": 0.001,
+      "loss": 2.6898,
+      "num_input_tokens_seen": 5400166400,
+      "step": 20600
+    },
+    {
+      "epoch": 0.09850101959287835,
+      "grad_norm": 0.1827591508626938,
+      "learning_rate": 0.001,
+      "loss": 2.6976,
+      "num_input_tokens_seen": 5413273600,
+      "step": 20650
+    },
+    {
+      "epoch": 0.09873952085097248,
+      "grad_norm": 0.16178373992443085,
+      "learning_rate": 0.001,
+      "loss": 2.7029,
+      "num_input_tokens_seen": 5426380800,
+      "step": 20700
+    },
+    {
+      "epoch": 0.09897802210906663,
+      "grad_norm": 0.1880313903093338,
+      "learning_rate": 0.001,
+      "loss": 2.6867,
+      "num_input_tokens_seen": 5439488000,
+      "step": 20750
+    },
+    {
+      "epoch": 0.09921652336716076,
+      "grad_norm": 0.17611584067344666,
+      "learning_rate": 0.001,
+      "loss": 2.6741,
+      "num_input_tokens_seen": 5452595200,
+      "step": 20800
+    },
+    {
+      "epoch": 0.09945502462525489,
+      "grad_norm": 0.17712561786174774,
+      "learning_rate": 0.001,
+      "loss": 2.7099,
+      "num_input_tokens_seen": 5465702400,
+      "step": 20850
+    },
+    {
+      "epoch": 0.09969352588334904,
+      "grad_norm": 0.18022434413433075,
+      "learning_rate": 0.001,
+      "loss": 2.6988,
+      "num_input_tokens_seen": 5478809600,
+      "step": 20900
+    },
+    {
+      "epoch": 0.09993202714144317,
+      "grad_norm": 0.17434161901474,
+      "learning_rate": 0.001,
+      "loss": 2.6869,
+      "num_input_tokens_seen": 5491916800,
+      "step": 20950
+    },
+    {
+      "epoch": 0.1001705283995373,
+      "grad_norm": 0.17802472412586212,
+      "learning_rate": 0.001,
+      "loss": 2.6935,
+      "num_input_tokens_seen": 5505024000,
+      "step": 21000
+    },
+    {
+      "epoch": 0.1001705283995373,
+      "eval_loss": 2.5838851928710938,
+      "eval_runtime": 50.1977,
+      "eval_samples_per_second": 99.606,
+      "eval_steps_per_second": 24.902,
+      "num_input_tokens_seen": 5505024000,
+      "step": 21000
     }
   ],
   "logging_steps": 50,
   "max_steps": 70000,
-  "num_input_tokens_seen": 5242880000,
+  "num_input_tokens_seen": 5505024000,
   "num_train_epochs": 1,
   "save_steps": 1000,
   "stateful_callbacks": {
@@ -3587,7 +3765,7 @@
       "attributes": {}
     }
   },
-  "total_flos": 1.4025215705088e+18,
+  "total_flos": 1.47264764903424e+18,
   "train_batch_size": 64,
   "trial_name": null,
   "trial_params": null