FastPlus Backup checkpoint-200 at global step 200

checkpoint-200/config.json

@@ -0,0 +1,19 @@
+{
+  "architectures": [
+    "FastPlusForCausalLM"
+  ],
+  "dtype": "float32",
+  "hidden_size": 768,
+  "initializer_range": 0.02,
+  "intermediate_size": 2304,
+  "kd_alpha": 0.4,
+  "kd_temperature": 2.5,
+  "max_position_embeddings": 512,
+  "model_type": "fastplus",
+  "num_attention_heads": 12,
+  "num_hidden_layers": 12,
+  "tie_word_embeddings": true,
+  "transformers_version": "5.12.0",
+  "use_cache": false,
+  "vocab_size": 31968
+}

checkpoint-200/generation_config.json

@@ -0,0 +1,6 @@
+{
+  "_from_model_config": true,
+  "output_attentions": false,
+  "output_hidden_states": false,
+  "transformers_version": "5.12.0"
+}

checkpoint-200/model.safetensors

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:d9347f5344c8fd71cc579ac9e5fc33393f2165ff35b250f86c50ad95dc6a68e1
+size 395782848

checkpoint-200/optimizer.pt

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:7922eb1caa7fa378405517f5fd57e7852dba943a46f9bbaa424d11446f619ee6
+size 444577931

checkpoint-200/rng_state.pth

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:61c19bab1174704a4a4441475683bf1270277af15d2e2c95e964789128e482c4
+size 14645

checkpoint-200/scaler.pt

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:4f6aeeac0ea461c1ec85dceeca5a6c03628efcb98037954f2a2f0e0c18ef5645
+size 1383

checkpoint-200/scheduler.pt

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:2c0e0dfffd6fba9a80ef6ed621acf2d555b6b74a647b4310093dfda237734471
+size 6009

checkpoint-200/trainer_state.json

@@ -0,0 +1,194 @@
+{
+  "best_global_step": null,
+  "best_metric": null,
+  "best_model_checkpoint": null,
+  "epoch": 0.006666666666666667,
+  "eval_steps": 500,
+  "global_step": 200,
+  "is_hyper_param_search": false,
+  "is_local_process_zero": true,
+  "is_world_process_zero": true,
+  "log_history": [
+    {
+      "epoch": 0.0003333333333333333,
+      "grad_norm": 252.23410034179688,
+      "learning_rate": 5.95e-06,
+      "loss": 166.30635986328124,
+      "original_ce_loss": 10.3935,
+      "step": 10
+    },
+    {
+      "epoch": 0.0006666666666666666,
+      "grad_norm": 251.30441284179688,
+      "learning_rate": 1.445e-05,
+      "loss": 166.0551513671875,
+      "original_ce_loss": 10.3778,
+      "step": 20
+    },
+    {
+      "epoch": 0.001,
+      "grad_norm": 252.8511199951172,
+      "learning_rate": 2.295e-05,
+      "loss": 165.40390625,
+      "original_ce_loss": 10.3371,
+      "step": 30
+    },
+    {
+      "epoch": 0.0013333333333333333,
+      "grad_norm": 279.1929016113281,
+      "learning_rate": 3.145e-05,
+      "loss": 164.07041015625,
+      "original_ce_loss": 10.2537,
+      "step": 40
+    },
+    {
+      "epoch": 0.0016666666666666668,
+      "grad_norm": 348.9263000488281,
+      "learning_rate": 3.9949999999999995e-05,
+      "loss": 160.942431640625,
+      "original_ce_loss": 10.0582,
+      "step": 50
+    },
+    {
+      "epoch": 0.002,
+      "grad_norm": 266.1781311035156,
+      "learning_rate": 4.76e-05,
+      "loss": 153.2720947265625,
+      "original_ce_loss": 9.5788,
+      "step": 60
+    },
+    {
+      "epoch": 0.0023333333333333335,
+      "grad_norm": 68.32039642333984,
+      "learning_rate": 5.61e-05,
+      "loss": 143.15457763671876,
+      "original_ce_loss": 8.9465,
+      "step": 70
+    },
+    {
+      "epoch": 0.0026666666666666666,
+      "grad_norm": NaN,
+      "learning_rate": 5.95e-05,
+      "loss": 138.36038818359376,
+      "original_ce_loss": 8.6468,
+      "step": 80
+    },
+    {
+      "epoch": 0.003,
+      "grad_norm": 45.997161865234375,
+      "learning_rate": 6.46e-05,
+      "loss": 136.0820068359375,
+      "original_ce_loss": 8.5044,
+      "step": 90
+    },
+    {
+      "epoch": 0.0033333333333333335,
+      "grad_norm": 41.57798385620117,
+      "learning_rate": 7.309999999999999e-05,
+      "loss": 130.63963623046874,
+      "original_ce_loss": 8.1643,
+      "step": 100
+    },
+    {
+      "epoch": 0.0036666666666666666,
+      "grad_norm": 36.38893508911133,
+      "learning_rate": 8.159999999999999e-05,
+      "loss": 125.012890625,
+      "original_ce_loss": 7.8126,
+      "step": 110
+    },
+    {
+      "epoch": 0.004,
+      "grad_norm": 34.60835647583008,
+      "learning_rate": 9.01e-05,
+      "loss": 118.98409423828124,
+      "original_ce_loss": 7.4358,
+      "step": 120
+    },
+    {
+      "epoch": 0.004333333333333333,
+      "grad_norm": 27.233781814575195,
+      "learning_rate": 9.86e-05,
+      "loss": 113.41510009765625,
+      "original_ce_loss": 7.0878,
+      "step": 130
+    },
+    {
+      "epoch": 0.004666666666666667,
+      "grad_norm": 14.006446838378906,
+      "learning_rate": 0.0001071,
+      "loss": 109.00380859375,
+      "original_ce_loss": 6.8121,
+      "step": 140
+    },
+    {
+      "epoch": 0.005,
+      "grad_norm": 15.040637969970703,
+      "learning_rate": 0.0001156,
+      "loss": 106.41650390625,
+      "original_ce_loss": 6.6504,
+      "step": 150
+    },
+    {
+      "epoch": 0.005333333333333333,
+      "grad_norm": 16.328678131103516,
+      "learning_rate": 0.00012409999999999998,
+      "loss": 104.68450927734375,
+      "original_ce_loss": 6.5421,
+      "step": 160
+    },
+    {
+      "epoch": 0.005666666666666667,
+      "grad_norm": 47.60938262939453,
+      "learning_rate": 0.0001326,
+      "loss": 104.513427734375,
+      "original_ce_loss": 6.5314,
+      "step": 170
+    },
+    {
+      "epoch": 0.006,
+      "grad_norm": 23.81610679626465,
+      "learning_rate": 0.0001411,
+      "loss": 104.72510986328125,
+      "original_ce_loss": 6.5447,
+      "step": 180
+    },
+    {
+      "epoch": 0.006333333333333333,
+      "grad_norm": 19.716007232666016,
+      "learning_rate": 0.00014959999999999998,
+      "loss": 104.5696044921875,
+      "original_ce_loss": 6.5349,
+      "step": 190
+    },
+    {
+      "epoch": 0.006666666666666667,
+      "grad_norm": 43.200706481933594,
+      "learning_rate": 0.0001581,
+      "loss": 104.36375732421875,
+      "original_ce_loss": 6.5221,
+      "step": 200
+    }
+  ],
+  "logging_steps": 10,
+  "max_steps": 30000,
+  "num_input_tokens_seen": 0,
+  "num_train_epochs": 9223372036854775807,
+  "save_steps": 200,
+  "stateful_callbacks": {
+    "TrainerControl": {
+      "args": {
+        "should_epoch_stop": false,
+        "should_evaluate": false,
+        "should_log": false,
+        "should_save": true,
+        "should_training_stop": false
+      },
+      "attributes": {}
+    }
+  },
+  "total_flos": 1.47372918177792e+16,
+  "train_batch_size": 16,
+  "trial_name": null,
+  "trial_params": null
+}

checkpoint-200/training_args.bin

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:175ec23ca20f10f250a237ce534b01197513d9d9393fd48c938e3ec11d0bdaec
+size 5137