config.json

config.json

@@ -0,0 +1,34 @@
+{
+  "model_type": "unknown", // Specify the base model type if known (e.g., "bert", "gpt-2")
+  "architectures": ["unknown"], // Specify the architecture if known (e.g., "BertForMaskedLM", "GPT2LMHeadModel")
+  "finetuned_from": "vilsonrodrigues/falcon-7b-instruct-sharded",
+  "tasks": "conversational", // Specify the task if it's a standard task in Hugging Face
+
+  // Include training hyperparameters
+  "learning_rate": 0.0002,
+  "train_batch_size": 8,
+  "eval_batch_size": 8,
+  "seed": 42,
+  "gradient_accumulation_steps": 8,
+  "total_train_batch_size": 64,
+  "optimizer": "Adam with betas=(0.9, 0.999) and epsilon=1e-08",
+  "lr_scheduler_type": "cosine",
+  "lr_scheduler_warmup_ratio": 0.03,
+  "training_steps": 5,
+
+  // Include library versions used for training
+  "framework": "Pytorch",
+  "pytorch_version": "2.1.0+cu118",
+  "transformers_version": "4.34.1",
+  "datasets_version": "2.14.6",
+  "tokenizers_version": "0.14.1",
+
+  // Other relevant model configuration parameters
+  "vocab_size": 0, // Specify the vocabulary size
+  "hidden_size": 0, // Specify the hidden layer size
+  "num_hidden_layers": 0, // Specify the number of hidden layers
+  "num_attention_heads": 0, // Specify the number of attention heads
+  // ... other model-specific parameters
+
+  "unknown_parameters": "This model has been fine-tuned on an unknown dataset with specific parameters and may not conform to the standard tasks or architectures. Please refer to the model documentation for more details."
+}