Update app.py

app.py

@@ -22,53 +22,6 @@ HUGGINGFACE_TOKEN = os.getenv("HUGGINGFACE_TOKEN")
 
 global_data = {
     'models': {},
-    'tokens': {
-        'eos': 'eos_token',
-        'pad': 'pad_token',
-        'padding': 'padding_token',
-        'unk': 'unk_token',
-        'bos': 'bos_token',
-        'sep': 'sep_token',
-        'cls': 'cls_token',
-        'mask': 'mask_token',
-        'n_ctx': 'n_ctx_token',
-        'vocab_size': 'vocab_size_token',
-        'n_embd': 'n_embd_token',
-        'n_head': 'n_head_token',
-        'n_layer': 'n_layer_token',
-        'n_vocab': 'n_vocab_token',
-        'block_size': 'block_size_token',
-        'n_rot': 'n_rot_token',
-        'rope_dim': 'rope_dim_token',
-        'rope_scaling': 'rope_scaling_token',
-        'n_positions': 'n_positions_token',
-        'use_cache': 'use_cache_token',
-        'use_parallel_inference': 'use_parallel_inference_token',
-        'parallel_inference_count': 'parallel_inference_count_token',
-        'use_mlock': 'use_mlock_token',
-        'use_mmap': 'use_mmap_token',
-        'use_cpu': 'use_cpu_token',
-        'f16_kv': 'f16_kv_token',
-        'f16_quant': 'f16_quant_token',
-        'f16_output': 'f16_output_token',
-        'use_flash_attn': 'use_flash_attn_token',
-        'max_seq_len': 'max_seq_len_token',
-        'do_sample': 'do_sample_token',
-        'top_k': 'top_k_token',
-        'top_p': 'top_p_token',
-        'temperature': 'temperature_token',
-        'num_return_sequences': 'num_return_sequences_token',
-        'use_repetition_penalty': 'use_repetition_penalty_token',
-        'repetition_penalty': 'repetition_penalty_token',
-        'no_repeat_ngram_size': 'no_repeat_ngram_size_token',
-        'bad_words_ids': 'bad_words_ids_token',
-        'use_token_logging': 'use_token_logging_token',
-        'use_tensor_parallel': 'use_tensor_parallel_token',
-        'tensor_parallel_size': 'tensor_parallel_size_token',
-        'use_gpu_memory_growth': 'use_gpu_memory_growth_token',
-        'use_multi_gpu_inference': 'use_multi_gpu_inference_token',
-        'multi_gpu_inference_count': 'multi_gpu_inference_count_token'
-    }
 }
 
 model_configs = [
@@ -381,4 +334,4 @@ def anonymize_ip():
 Thread(target=anonymize_ip).start()
 
 if __name__ == "__main__":
-    iface.launch(share=True)
+    iface.launch(share=True)