Handle unsharded Llama2 model types in conversion script (#27069)

Handle all unshared models types

src/transformers/models/llama/convert_llama_weights_to_hf.py

@@ -124,7 +124,7 @@ def write_model(model_path, input_base_path, model_size, tokenizer_path=None, sa
 
     print(f"Fetching all parameters from the checkpoint at {input_base_path}.")
     # Load weights
-    if model_size == "7B":
+    if num_shards == 1:
         # Not sharded
         # (The sharded implementation would also work, but this is simpler.)
         loaded = torch.load(os.path.join(input_base_path, "consolidated.00.pth"), map_location="cpu")
@@ -138,7 +138,7 @@ def write_model(model_path, input_base_path, model_size, tokenizer_path=None, sa
     index_dict = {"weight_map": {}}
     for layer_i in range(n_layers):
         filename = f"pytorch_model-{layer_i + 1}-of-{n_layers + 1}.bin"
-        if model_size == "7B":
+        if num_shards == 1:
             # Unsharded
             state_dict = {
                 f"model.layers.{layer_i}.self_attn.q_proj.weight": permute(
@@ -222,7 +222,7 @@ def write_model(model_path, input_base_path, model_size, tokenizer_path=None, sa
         torch.save(state_dict, os.path.join(tmp_model_path, filename))
 
     filename = f"pytorch_model-{n_layers + 1}-of-{n_layers + 1}.bin"
-    if model_size == "7B":
+    if num_shards == 1:
         # Unsharded
         state_dict = {
             "model.embed_tokens.weight": loaded["tok_embeddings.weight"],