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:b7b195386e89d68b0f9f39026d203a85980165d536bc749b9f93d59b472de9bb
+size 395782848

checkpoint-200/optimizer.pt

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:1ce84f7743782642a671257abb713dce68d2f73ce7f505736638af3a794bf710
+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 sha256:9918311b56fae2af220fee0d372f5247d5fc99fb53910e241b542c2dc7256a73
+size 1383

checkpoint-200/scheduler.pt

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:9bf99d97b19f95dd798ea5d287831556502f92e1b27a0619aae528b904c60f74
+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": 246.00827026367188,
+      "learning_rate": 4.76e-06,
+      "loss": 166.319970703125,
+      "original_ce_loss": 10.3945,
+      "step": 10
+    },
+    {
+      "epoch": 0.0006666666666666666,
+      "grad_norm": 244.34457397460938,
+      "learning_rate": 1.1560000000000001e-05,
+      "loss": 166.11895751953125,
+      "original_ce_loss": 10.3819,
+      "step": 20
+    },
+    {
+      "epoch": 0.001,
+      "grad_norm": 243.96890258789062,
+      "learning_rate": 1.836e-05,
+      "loss": 165.62098388671876,
+      "original_ce_loss": 10.3508,
+      "step": 30
+    },
+    {
+      "epoch": 0.0013333333333333333,
+      "grad_norm": 268.2583312988281,
+      "learning_rate": 2.516e-05,
+      "loss": 164.65194091796874,
+      "original_ce_loss": 10.2902,
+      "step": 40
+    },
+    {
+      "epoch": 0.0016666666666666668,
+      "grad_norm": 315.6184997558594,
+      "learning_rate": 3.1960000000000006e-05,
+      "loss": 162.7759521484375,
+      "original_ce_loss": 10.173,
+      "step": 50
+    },
+    {
+      "epoch": 0.002,
+      "grad_norm": 378.4802551269531,
+      "learning_rate": 3.876e-05,
+      "loss": 158.40428466796874,
+      "original_ce_loss": 9.8997,
+      "step": 60
+    },
+    {
+      "epoch": 0.0023333333333333335,
+      "grad_norm": 178.61724853515625,
+      "learning_rate": 4.4880000000000004e-05,
+      "loss": 148.0895263671875,
+      "original_ce_loss": 9.2551,
+      "step": 70
+    },
+    {
+      "epoch": 0.0026666666666666666,
+      "grad_norm": 58.28500747680664,
+      "learning_rate": 5.168e-05,
+      "loss": 139.22628173828124,
+      "original_ce_loss": 8.7011,
+      "step": 80
+    },
+    {
+      "epoch": 0.003,
+      "grad_norm": NaN,
+      "learning_rate": 5.780000000000001e-05,
+      "loss": 133.89996337890625,
+      "original_ce_loss": 8.3682,
+      "step": 90
+    },
+    {
+      "epoch": 0.0033333333333333335,
+      "grad_norm": NaN,
+      "learning_rate": 5.848e-05,
+      "loss": 131.97056884765624,
+      "original_ce_loss": 8.2476,
+      "step": 100
+    },
+    {
+      "epoch": 0.0036666666666666666,
+      "grad_norm": 45.780582427978516,
+      "learning_rate": 6.392000000000001e-05,
+      "loss": 129.279736328125,
+      "original_ce_loss": 8.0794,
+      "step": 110
+    },
+    {
+      "epoch": 0.004,
+      "grad_norm": 38.22166442871094,
+      "learning_rate": 7.072e-05,
+      "loss": 124.1231689453125,
+      "original_ce_loss": 7.7572,
+      "step": 120
+    },
+    {
+      "epoch": 0.004333333333333333,
+      "grad_norm": 34.60215377807617,
+      "learning_rate": 7.752e-05,
+      "loss": 119.131298828125,
+      "original_ce_loss": 7.4452,
+      "step": 130
+    },
+    {
+      "epoch": 0.004666666666666667,
+      "grad_norm": 27.215518951416016,
+      "learning_rate": 8.432e-05,
+      "loss": 114.3770263671875,
+      "original_ce_loss": 7.148,
+      "step": 140
+    },
+    {
+      "epoch": 0.005,
+      "grad_norm": 27.311050415039062,
+      "learning_rate": 9.112000000000001e-05,
+      "loss": 110.84078369140624,
+      "original_ce_loss": 6.927,
+      "step": 150
+    },
+    {
+      "epoch": 0.005333333333333333,
+      "grad_norm": 29.360546112060547,
+      "learning_rate": 9.792e-05,
+      "loss": 108.61451416015625,
+      "original_ce_loss": 6.7879,
+      "step": 160
+    },
+    {
+      "epoch": 0.005666666666666667,
+      "grad_norm": 30.510602951049805,
+      "learning_rate": 0.00010472,
+      "loss": 108.4002685546875,
+      "original_ce_loss": 6.7745,
+      "step": 170
+    },
+    {
+      "epoch": 0.006,
+      "grad_norm": 34.6984977722168,
+      "learning_rate": 0.00011152000000000002,
+      "loss": 108.417431640625,
+      "original_ce_loss": 6.7756,
+      "step": 180
+    },
+    {
+      "epoch": 0.006333333333333333,
+      "grad_norm": 26.99405288696289,
+      "learning_rate": 0.00011832,
+      "loss": 108.10020751953125,
+      "original_ce_loss": 6.7557,
+      "step": 190
+    },
+    {
+      "epoch": 0.006666666666666667,
+      "grad_norm": 53.1820068359375,
+      "learning_rate": 0.00012512,
+      "loss": 106.81435546875,
+      "original_ce_loss": 6.6754,
+      "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 sha256:25417f401430668fae7517ca595e0978a628165e7693ec7a7a5f83234373edb0
+size 5137