MHGanainy/mgpt-lora-multi-denmark-balanced-1024

README.md

@@ -15,6 +15,8 @@ should probably proofread and complete it, then remove this comment. -->
 # mgpt-lora-multi-denmark-balanced-1024
 
 This model is a fine-tuned version of [ai-forever/mGPT](https://huggingface.co/ai-forever/mGPT) on an unknown dataset.
+It achieves the following results on the evaluation set:
+- Loss: 1.8459
 
 ## Model description
 

all_results.json

@@ -0,0 +1,13 @@
+{
+    "epoch": 1.0,
+    "eval_loss": 1.8459237813949585,
+    "eval_runtime": 73.2243,
+    "eval_samples_per_second": 79.782,
+    "eval_steps_per_second": 9.983,
+    "perplexity": 6.333948272197309,
+    "total_flos": 3.478541817358582e+17,
+    "train_loss": 2.9001881553825744,
+    "train_runtime": 774.5812,
+    "train_samples_per_second": 30.024,
+    "train_steps_per_second": 3.753
+}

eval_results.json

@@ -0,0 +1,8 @@
+{
+    "epoch": 1.0,
+    "eval_loss": 1.8459237813949585,
+    "eval_runtime": 73.2243,
+    "eval_samples_per_second": 79.782,
+    "eval_steps_per_second": 9.983,
+    "perplexity": 6.333948272197309
+}

train_results.json

@@ -0,0 +1,8 @@
+{
+    "epoch": 1.0,
+    "total_flos": 3.478541817358582e+17,
+    "train_loss": 2.9001881553825744,
+    "train_runtime": 774.5812,
+    "train_samples_per_second": 30.024,
+    "train_steps_per_second": 3.753
+}

trainer_state.json

@@ -0,0 +1,245 @@
+{
+  "best_metric": null,
+  "best_model_checkpoint": null,
+  "epoch": 1.0,
+  "eval_steps": 500,
+  "global_step": 2907,
+  "is_hyper_param_search": false,
+  "is_local_process_zero": true,
+  "is_world_process_zero": true,
+  "log_history": [
+    {
+      "epoch": 0.03439972480220158,
+      "grad_norm": 10.833683013916016,
+      "learning_rate": 8.60215053763441e-07,
+      "loss": 6.3304,
+      "step": 100
+    },
+    {
+      "epoch": 0.06879944960440317,
+      "grad_norm": 5.736911773681641,
+      "learning_rate": 1.720430107526882e-06,
+      "loss": 5.8518,
+      "step": 200
+    },
+    {
+      "epoch": 0.10319917440660474,
+      "grad_norm": 2.5415754318237305,
+      "learning_rate": 2.580645161290323e-06,
+      "loss": 4.8713,
+      "step": 300
+    },
+    {
+      "epoch": 0.13759889920880633,
+      "grad_norm": 2.0687456130981445,
+      "learning_rate": 3.440860215053764e-06,
+      "loss": 4.2268,
+      "step": 400
+    },
+    {
+      "epoch": 0.1719986240110079,
+      "grad_norm": 2.140303134918213,
+      "learning_rate": 4.3010752688172045e-06,
+      "loss": 4.016,
+      "step": 500
+    },
+    {
+      "epoch": 0.20639834881320948,
+      "grad_norm": 1.2515285015106201,
+      "learning_rate": 5.161290322580646e-06,
+      "loss": 3.9056,
+      "step": 600
+    },
+    {
+      "epoch": 0.24079807361541108,
+      "grad_norm": 1.818322777748108,
+      "learning_rate": 6.021505376344087e-06,
+      "loss": 3.7952,
+      "step": 700
+    },
+    {
+      "epoch": 0.27519779841761266,
+      "grad_norm": 1.817495584487915,
+      "learning_rate": 6.881720430107528e-06,
+      "loss": 3.6559,
+      "step": 800
+    },
+    {
+      "epoch": 0.30959752321981426,
+      "grad_norm": 4.060177326202393,
+      "learning_rate": 7.741935483870968e-06,
+      "loss": 3.4939,
+      "step": 900
+    },
+    {
+      "epoch": 0.3439972480220158,
+      "grad_norm": 5.048155784606934,
+      "learning_rate": 8.602150537634409e-06,
+      "loss": 3.2084,
+      "step": 1000
+    },
+    {
+      "epoch": 0.3783969728242174,
+      "grad_norm": 4.375072479248047,
+      "learning_rate": 9.46236559139785e-06,
+      "loss": 2.7951,
+      "step": 1100
+    },
+    {
+      "epoch": 0.41279669762641896,
+      "grad_norm": 3.271152973175049,
+      "learning_rate": 1.0322580645161291e-05,
+      "loss": 2.4737,
+      "step": 1200
+    },
+    {
+      "epoch": 0.44719642242862057,
+      "grad_norm": 2.0996718406677246,
+      "learning_rate": 1.118279569892473e-05,
+      "loss": 2.4253,
+      "step": 1300
+    },
+    {
+      "epoch": 0.48159614723082217,
+      "grad_norm": 2.1390457153320312,
+      "learning_rate": 1.2043010752688173e-05,
+      "loss": 2.2669,
+      "step": 1400
+    },
+    {
+      "epoch": 0.5159958720330238,
+      "grad_norm": 1.7436959743499756,
+      "learning_rate": 1.2903225806451613e-05,
+      "loss": 2.2353,
+      "step": 1500
+    },
+    {
+      "epoch": 0.5503955968352253,
+      "grad_norm": 2.8704700469970703,
+      "learning_rate": 1.3763440860215056e-05,
+      "loss": 2.2071,
+      "step": 1600
+    },
+    {
+      "epoch": 0.5847953216374269,
+      "grad_norm": 1.6568603515625,
+      "learning_rate": 1.4623655913978497e-05,
+      "loss": 2.1548,
+      "step": 1700
+    },
+    {
+      "epoch": 0.6191950464396285,
+      "grad_norm": 3.1028966903686523,
+      "learning_rate": 1.5483870967741936e-05,
+      "loss": 2.1223,
+      "step": 1800
+    },
+    {
+      "epoch": 0.6535947712418301,
+      "grad_norm": 1.7996934652328491,
+      "learning_rate": 1.6344086021505377e-05,
+      "loss": 2.0797,
+      "step": 1900
+    },
+    {
+      "epoch": 0.6879944960440316,
+      "grad_norm": 1.737014651298523,
+      "learning_rate": 1.7204301075268818e-05,
+      "loss": 2.1156,
+      "step": 2000
+    },
+    {
+      "epoch": 0.7223942208462333,
+      "grad_norm": 1.8968210220336914,
+      "learning_rate": 1.806451612903226e-05,
+      "loss": 2.0334,
+      "step": 2100
+    },
+    {
+      "epoch": 0.7567939456484348,
+      "grad_norm": 1.7477818727493286,
+      "learning_rate": 1.89247311827957e-05,
+      "loss": 2.0278,
+      "step": 2200
+    },
+    {
+      "epoch": 0.7911936704506364,
+      "grad_norm": 2.6288206577301025,
+      "learning_rate": 1.978494623655914e-05,
+      "loss": 2.0297,
+      "step": 2300
+    },
+    {
+      "epoch": 0.8255933952528379,
+      "grad_norm": 1.7430673837661743,
+      "learning_rate": 1.9191636897958123e-05,
+      "loss": 2.0365,
+      "step": 2400
+    },
+    {
+      "epoch": 0.8599931200550396,
+      "grad_norm": 1.3275859355926514,
+      "learning_rate": 1.586037078125607e-05,
+      "loss": 1.9793,
+      "step": 2500
+    },
+    {
+      "epoch": 0.8943928448572411,
+      "grad_norm": 1.2883193492889404,
+      "learning_rate": 1.086259476800041e-05,
+      "loss": 1.9716,
+      "step": 2600
+    },
+    {
+      "epoch": 0.9287925696594427,
+      "grad_norm": 2.1552555561065674,
+      "learning_rate": 5.619523231433177e-06,
+      "loss": 1.9643,
+      "step": 2700
+    },
+    {
+      "epoch": 0.9631922944616443,
+      "grad_norm": 1.3895961046218872,
+      "learning_rate": 1.6221250761114803e-06,
+      "loss": 1.9634,
+      "step": 2800
+    },
+    {
+      "epoch": 0.9975920192638459,
+      "grad_norm": 1.249545931816101,
+      "learning_rate": 7.1378626715268295e-09,
+      "loss": 1.9456,
+      "step": 2900
+    },
+    {
+      "epoch": 1.0,
+      "step": 2907,
+      "total_flos": 3.478541817358582e+17,
+      "train_loss": 2.9001881553825744,
+      "train_runtime": 774.5812,
+      "train_samples_per_second": 30.024,
+      "train_steps_per_second": 3.753
+    }
+  ],
+  "logging_steps": 100,
+  "max_steps": 2907,
+  "num_input_tokens_seen": 0,
+  "num_train_epochs": 1,
+  "save_steps": 500,
+  "stateful_callbacks": {
+    "TrainerControl": {
+      "args": {
+        "should_epoch_stop": false,
+        "should_evaluate": false,
+        "should_log": false,
+        "should_save": false,
+        "should_training_stop": false
+      },
+      "attributes": {}
+    }
+  },
+  "total_flos": 3.478541817358582e+17,
+  "train_batch_size": 1,
+  "trial_name": null,
+  "trial_params": null
+}