Training in progress, step 485000

manual/checkpoint-17000/config.json

@@ -0,0 +1,40 @@
+{
+  "_name_or_path": "trainer/configs/large_config.json",
+  "activation_function": "gelu_new",
+  "architectures": [
+    "SAFEDoubleHeadsModel"
+  ],
+  "attn_pdrop": 0.1,
+  "bos_token_id": 1,
+  "embd_pdrop": 0.1,
+  "eos_token_id": 2,
+  "id2label": {
+    "0": "LABEL_0"
+  },
+  "initializer_range": 0.02,
+  "label2id": {
+    "LABEL_0": 0
+  },
+  "layer_norm_epsilon": 1e-05,
+  "model_type": "gpt2",
+  "n_embd": 660,
+  "n_head": 10,
+  "n_inner": null,
+  "n_layer": 10,
+  "n_positions": 1024,
+  "pad_token_id": 3,
+  "reorder_and_upcast_attn": false,
+  "resid_pdrop": 0.1,
+  "scale_attn_by_inverse_layer_idx": false,
+  "scale_attn_weights": true,
+  "summary_activation": "tanh",
+  "summary_first_dropout": 0.1,
+  "summary_hidden_size": 128,
+  "summary_proj_to_labels": true,
+  "summary_type": "cls_index",
+  "summary_use_proj": true,
+  "torch_dtype": "float32",
+  "transformers_version": "4.42.3",
+  "use_cache": true,
+  "vocab_size": 1880
+}

manual/checkpoint-17000/generation_config.json

@@ -0,0 +1,7 @@
+{
+  "_from_model_config": true,
+  "bos_token_id": 1,
+  "eos_token_id": 2,
+  "pad_token_id": 3,
+  "transformers_version": "4.42.3"
+}

manual/checkpoint-17000/model.safetensors

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:b601a3ed10eef25f03ee3e8ab8ff8a6a9f7e8197f258d333f9726ea6f5ce4220
+size 217454844

manual/checkpoint-17000/optimizer.pt

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:16186e34e3a93366787c45d0851d4c2e4154e00aed7823170bd2fc8bbf676cf6
+size 434307002

manual/checkpoint-17000/rng_state.pth

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:b1346e9edf4eb92556d7e6a530ba146df909f84d7279c1ecee8a0595a021fa75
+size 14244

manual/checkpoint-17000/scheduler.pt

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:d7d9c20964d6b5816b3dbf35c5f6287479777a7ae3da85a79e7ca4f93bf07ccb
+size 1064

manual/checkpoint-17000/trainer_state.json

@@ -0,0 +1,271 @@
+{
+  "best_metric": null,
+  "best_model_checkpoint": null,
+  "epoch": 0.17,
+  "eval_steps": 500.0,
+  "global_step": 17000,
+  "is_hyper_param_search": false,
+  "is_local_process_zero": true,
+  "is_world_process_zero": true,
+  "log_history": [
+    {
+      "epoch": 0.005,
+      "grad_norm": 2.999999761581421,
+      "learning_rate": 5.000000000000001e-07,
+      "loss": 4.7795,
+      "step": 500
+    },
+    {
+      "epoch": 0.01,
+      "grad_norm": 2.9999992847442627,
+      "learning_rate": 1.0000000000000002e-06,
+      "loss": 2.4619,
+      "step": 1000
+    },
+    {
+      "epoch": 0.015,
+      "grad_norm": 2.999999761581421,
+      "learning_rate": 1.5e-06,
+      "loss": 2.0368,
+      "step": 1500
+    },
+    {
+      "epoch": 0.02,
+      "grad_norm": 2.9999992847442627,
+      "learning_rate": 2.0000000000000003e-06,
+      "loss": 1.7481,
+      "step": 2000
+    },
+    {
+      "epoch": 0.025,
+      "grad_norm": 2.999999761581421,
+      "learning_rate": 2.5e-06,
+      "loss": 1.4947,
+      "step": 2500
+    },
+    {
+      "epoch": 0.03,
+      "grad_norm": 2.9999992847442627,
+      "learning_rate": 3e-06,
+      "loss": 1.3007,
+      "step": 3000
+    },
+    {
+      "epoch": 0.035,
+      "grad_norm": 2.9999990463256836,
+      "learning_rate": 3.5e-06,
+      "loss": 1.1635,
+      "step": 3500
+    },
+    {
+      "epoch": 0.04,
+      "grad_norm": 2.999999761581421,
+      "learning_rate": 4.000000000000001e-06,
+      "loss": 1.0721,
+      "step": 4000
+    },
+    {
+      "epoch": 0.045,
+      "grad_norm": 2.9999990463256836,
+      "learning_rate": 4.5e-06,
+      "loss": 1.0072,
+      "step": 4500
+    },
+    {
+      "epoch": 0.05,
+      "grad_norm": 2.9999992847442627,
+      "learning_rate": 5e-06,
+      "loss": 0.9584,
+      "step": 5000
+    },
+    {
+      "epoch": 0.055,
+      "grad_norm": 2.9999992847442627,
+      "learning_rate": 5.500000000000001e-06,
+      "loss": 0.918,
+      "step": 5500
+    },
+    {
+      "epoch": 0.06,
+      "grad_norm": 2.9999992847442627,
+      "learning_rate": 6e-06,
+      "loss": 0.8862,
+      "step": 6000
+    },
+    {
+      "epoch": 0.065,
+      "grad_norm": 2.9999992847442627,
+      "learning_rate": 6.5000000000000004e-06,
+      "loss": 0.8607,
+      "step": 6500
+    },
+    {
+      "epoch": 0.07,
+      "grad_norm": 2.9999990463256836,
+      "learning_rate": 7e-06,
+      "loss": 0.8363,
+      "step": 7000
+    },
+    {
+      "epoch": 0.075,
+      "grad_norm": 2.9999992847442627,
+      "learning_rate": 7.500000000000001e-06,
+      "loss": 0.8174,
+      "step": 7500
+    },
+    {
+      "epoch": 0.08,
+      "grad_norm": 2.9999992847442627,
+      "learning_rate": 8.000000000000001e-06,
+      "loss": 0.8015,
+      "step": 8000
+    },
+    {
+      "epoch": 0.085,
+      "grad_norm": 2.999999523162842,
+      "learning_rate": 8.5e-06,
+      "loss": 0.7885,
+      "step": 8500
+    },
+    {
+      "epoch": 0.09,
+      "grad_norm": 2.762766122817993,
+      "learning_rate": 9e-06,
+      "loss": 0.7744,
+      "step": 9000
+    },
+    {
+      "epoch": 0.095,
+      "grad_norm": 2.5979771614074707,
+      "learning_rate": 9.5e-06,
+      "loss": 0.765,
+      "step": 9500
+    },
+    {
+      "epoch": 0.1,
+      "grad_norm": 2.4132823944091797,
+      "learning_rate": 1e-05,
+      "loss": 0.7546,
+      "step": 10000
+    },
+    {
+      "epoch": 0.105,
+      "grad_norm": 2.9999990463256836,
+      "learning_rate": 9.944444444444445e-06,
+      "loss": 0.7454,
+      "step": 10500
+    },
+    {
+      "epoch": 0.11,
+      "grad_norm": 1.9988462924957275,
+      "learning_rate": 9.88888888888889e-06,
+      "loss": 0.7354,
+      "step": 11000
+    },
+    {
+      "epoch": 0.115,
+      "grad_norm": 2.1960322856903076,
+      "learning_rate": 9.833333333333333e-06,
+      "loss": 0.7274,
+      "step": 11500
+    },
+    {
+      "epoch": 0.12,
+      "grad_norm": 2.3398022651672363,
+      "learning_rate": 9.777777777777779e-06,
+      "loss": 0.7211,
+      "step": 12000
+    },
+    {
+      "epoch": 0.125,
+      "grad_norm": 1.904077172279358,
+      "learning_rate": 9.722222222222223e-06,
+      "loss": 0.7152,
+      "step": 12500
+    },
+    {
+      "epoch": 0.13,
+      "grad_norm": 2.774881601333618,
+      "learning_rate": 9.666666666666667e-06,
+      "loss": 0.7099,
+      "step": 13000
+    },
+    {
+      "epoch": 0.135,
+      "grad_norm": 1.7885903120040894,
+      "learning_rate": 9.611111111111112e-06,
+      "loss": 0.7046,
+      "step": 13500
+    },
+    {
+      "epoch": 0.14,
+      "grad_norm": 1.9303686618804932,
+      "learning_rate": 9.555555555555556e-06,
+      "loss": 0.6997,
+      "step": 14000
+    },
+    {
+      "epoch": 0.145,
+      "grad_norm": 2.0811753273010254,
+      "learning_rate": 9.5e-06,
+      "loss": 0.6949,
+      "step": 14500
+    },
+    {
+      "epoch": 0.15,
+      "grad_norm": 2.0698301792144775,
+      "learning_rate": 9.444444444444445e-06,
+      "loss": 0.6918,
+      "step": 15000
+    },
+    {
+      "epoch": 0.155,
+      "grad_norm": 1.838094711303711,
+      "learning_rate": 9.38888888888889e-06,
+      "loss": 0.688,
+      "step": 15500
+    },
+    {
+      "epoch": 0.16,
+      "grad_norm": 1.7373666763305664,
+      "learning_rate": 9.333333333333334e-06,
+      "loss": 0.6835,
+      "step": 16000
+    },
+    {
+      "epoch": 0.165,
+      "grad_norm": 2.1607325077056885,
+      "learning_rate": 9.277777777777778e-06,
+      "loss": 0.6802,
+      "step": 16500
+    },
+    {
+      "epoch": 0.17,
+      "grad_norm": 1.8940073251724243,
+      "learning_rate": 9.222222222222224e-06,
+      "loss": 0.6761,
+      "step": 17000
+    }
+  ],
+  "logging_steps": 500,
+  "max_steps": 100000,
+  "num_input_tokens_seen": 0,
+  "num_train_epochs": 9223372036854775807,
+  "save_steps": 1000,
+  "stateful_callbacks": {
+    "TrainerControl": {
+      "args": {
+        "should_epoch_stop": false,
+        "should_evaluate": false,
+        "should_log": false,
+        "should_save": true,
+        "should_training_stop": false
+      },
+      "attributes": {}
+    }
+  },
+  "total_flos": 4.273433586508646e+16,
+  "train_batch_size": 64,
+  "trial_name": null,
+  "trial_params": null
+}

manual/checkpoint-17000/training_args.bin

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:e59b3a3018b4bac8c30c130dc256bae308fe3db1ac00a19ed74630792abf832e
+size 5304

model.safetensors

@@ -1,3 +1,3 @@
 version https://git-lfs.github.com/spec/v1
-oid sha256:b601a3ed10eef25f03ee3e8ab8ff8a6a9f7e8197f258d333f9726ea6f5ce4220
+oid sha256:8a98bf99e9eb418ca24f3398cc6ccf53c4083fb63f4c4820cab8d843feed8c30
 size 217454844

training_args.bin

@@ -1,3 +1,3 @@
 version https://git-lfs.github.com/spec/v1
-oid sha256:e59b3a3018b4bac8c30c130dc256bae308fe3db1ac00a19ed74630792abf832e
+oid sha256:c4b950106c8844bc1e46b02a9a7e75d3c88cba83db826edf2d777fae7a97508b
 size 5304