Delete config.json

config.json

@@ -1,46 +0,0 @@
-{
-  "architectures": [
-    "LlamaForCausalLM"
-  ],
-  "attention_bias": false,
-  "attention_dropout": 0.0,
-  "bos_token_id": 128000,
-  "eos_token_id": [
-    128001,
-    128008,
-    128009
-  ],
-  "hidden_act": "silu",
-  "hidden_size": 4096,  // Verify based on the actual architecture
-  "initializer_range": 0.02,
-  "intermediate_size": 14336,  // This may vary depending on the actual model
-  "max_position_embeddings": 2048,  // Adjust if needed (default max sequence length)
-  "mlp_bias": false,
-  "model_type": "llama",
-  "num_attention_heads": 32,  // Adjust based on actual architecture, might differ
-  "num_hidden_layers": 32,  // Adjust based on actual architecture
-  "num_key_value_heads": 8,
-  "pretraining_tp": 1,
-  "rms_norm_eps": 1e-05,
-  "rope_scaling": {
-    "factor": 2.0,  // Modify based on your model's specs
-    "original_max_position_embeddings": 2048
-  },
-  "rope_theta": 10000.0,
-  "tie_word_embeddings": true,  // Keep this true to share word embeddings
-  "torch_dtype": "bfloat16",
-  "transformers_version": "4.44.2",  // Match this to your Transformers version
-  "use_cache": true,
-  "vocab_size": 128256,  // Ensure this matches the actual vocab size
-  "quantization": "PEFT",  // Include this since the model uses PEFT fine-tuning
-  "training_framework": "AutoTrain",
-  "training_metrics": {
-    "loss": 1.74,  // Based on training data from the README
-    "epochs": 5,
-    "learning_rate": "3e-05"
-  },
-  "datasets": {
-    "name": "conversational-dataset",
-    "type": "text"
-  }
-}