初始化项目，由ModelHub XC社区提供模型

Model: abinayam/gpt-2-tamil Source: Original Platform
2026-07-17 07:06:11 +08:00
commit f301701ac2
55 changed files with 3107 additions and 0 deletions
--- a/scripts/wandb/run-20210715_085943-1ize2alk/files/wandb-metadata.json
+++ b/scripts/wandb/run-20210715_085943-1ize2alk/files/wandb-metadata.json
@@ -0,0 +1,49 @@
+{
+    "os": "Linux-5.4.0-1043-gcp-x86_64-with-glibc2.29",
+    "python": "3.8.10",
+    "heartbeatAt": "2021-07-15T08:59:45.122600",
+    "startedAt": "2021-07-15T08:59:43.232731",
+    "docker": null,
+    "cpu_count": 96,
+    "cuda": null,
+    "args": [
+        "--output_dir=../gpt-2-tamil",
+        "--model_type=gpt2",
+        "--config_name=../gpt-2-tamil",
+        "--tokenizer_name=../gpt-2-tamil",
+        "--dataset_name=oscar",
+        "--dataset_config_name=unshuffled_deduplicated_ta",
+        "--do_train",
+        "--do_eval",
+        "--block_size=512",
+        "--per_device_train_batch_size=128",
+        "--per_device_eval_batch_size=128",
+        "--learning_rate=3e-5",
+        "--warmup_steps=1000",
+        "--adam_beta1=0.9",
+        "--adam_beta2=0.98",
+        "--weight_decay=0.01",
+        "--overwrite_output_dir",
+        "--num_train_epochs=10",
+        "--report_to",
+        "wandb",
+        "--run_name",
+        "trial",
+        "--logging_steps=500",
+        "--save_steps=2500",
+        "--eval_steps=2500",
+        "--preprocessing_num_workers=90"
+    ],
+    "state": "running",
+    "program": "../src/run_clm_flax.py",
+    "codePath": "src/run_clm_flax.py",
+    "git": {
+        "remote": "https://github.com/AbinayaM02/GPT2-Tamil.git",
+        "commit": "69c9b7bf75b708a8f62cf5833d1b89acf5d1760b"
+    },
+    "email": "abinaya.m02@mphasis.com",
+    "root": "/home/tweety_abi/GPT2-Tamil",
+    "host": "t1v-n-ebe36c53-w-0",
+    "username": "tweety_abi",
+    "executable": "/home/tweety_abi/gpt2_env/bin/python"
+}