Upload folder using huggingface_hub

.gitattributes

@@ -33,3 +33,4 @@ saved_model/**/* filter=lfs diff=lfs merge=lfs -text
 *.zip filter=lfs diff=lfs merge=lfs -text
 *.zst filter=lfs diff=lfs merge=lfs -text
 *tfevents* filter=lfs diff=lfs merge=lfs -text
+tokenizer.json filter=lfs diff=lfs merge=lfs -text

config.json

@@ -0,0 +1,42 @@
+{
+  "architectures": [
+    "UltravoxModel"
+  ],
+  "audio_model_id": "kotoba-tech/kotoba-whisper-v2.2",
+  "audio_model_lora_config": {
+    "lora_alpha": 8,
+    "r": 0,
+    "target_modules": [
+      "k_proj",
+      "q_proj",
+      "linear_k",
+      "linear_q"
+    ]
+  },
+  "auto_map": {
+    "AutoConfig": "ultravox_config.UltravoxConfig",
+    "AutoModel": "ultravox_model.UltravoxModel"
+  },
+  "hidden_size": 4096,
+  "ignore_index": -100,
+  "initializer_range": 0.02,
+  "model_type": "ultravox",
+  "norm_init": 0.4,
+  "pad_token_id": 1,
+  "projector_act": "swiglu",
+  "stack_factor": 8,
+  "text_model_id": "google/gemma-2-9b-it",
+  "text_model_lora_config": {
+    "lora_alpha": 8,
+    "r": 0,
+    "target_modules": [
+      "k_proj",
+      "q_proj",
+      "linear_k",
+      "linear_q"
+    ]
+  },
+  "torch_dtype": "bfloat16",
+  "transformers_version": "4.44.0",
+  "vocab_size": 256000
+}

generation_config.json

@@ -0,0 +1,8 @@
+{
+  "_from_model_config": true,
+  "bos_token_id": 2,
+  "cache_implementation": "hybrid",
+  "eos_token_id": 1,
+  "pad_token_id": 1,
+  "transformers_version": "4.44.0"
+}

model.safetensors

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:7287a044c945ad4f1903cef5887e534c70cddb3852b1a2bd80a230e61d00b22f
+size 98594264

optimizer.pt

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:2b3fde1382a141c8b9075d0038d9b4ecfa526a15bca7e0b376ced5c550dce6db
+size 197192018

rng_state.pth

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:50406e2683fae845a67d2522407cfd71c13ce88867a1aac0dc9d26a8b3a5f840
+size 14244

scheduler.pt

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:ae7d1df7fbe282a2a1f6e0263b266c20c101d0f96ff5d77217fca16dab8166f9
+size 1064

special_tokens_map.json

@@ -0,0 +1,28 @@
+{
+  "additional_special_tokens": [
+    "<start_of_turn>",
+    "<end_of_turn>"
+  ],
+  "bos_token": {
+    "content": "<bos>",
+    "lstrip": false,
+    "normalized": false,
+    "rstrip": false,
+    "single_word": false
+  },
+  "eos_token": {
+    "content": "<eos>",
+    "lstrip": false,
+    "normalized": false,
+    "rstrip": false,
+    "single_word": false
+  },
+  "pad_token": "<eos>",
+  "unk_token": {
+    "content": "<unk>",
+    "lstrip": false,
+    "normalized": false,
+    "rstrip": false,
+    "single_word": false
+  }
+}

tokenizer.json

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:3f289bc05132635a8bc7aca7aa21255efd5e18f3710f43e3cdb96bcd41be4922
+size 17525357

tokenizer.model

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:61a7b147390c64585d6c3543dd6fc636906c9af3865a5548f27f31aee1d4c8e2
+size 4241003

tokenizer_config.json

@@ -0,0 +1,2013 @@
+{
+  "add_bos_token": true,
+  "add_eos_token": false,
+  "added_tokens_decoder": {
+    "0": {
+      "content": "<pad>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "1": {
+      "content": "<eos>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "2": {
+      "content": "<bos>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "3": {
+      "content": "<unk>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "4": {
+      "content": "<mask>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "5": {
+      "content": "<2mass>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "6": {
+      "content": "[@BOS@]",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "7": {
+      "content": "<unused0>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "8": {
+      "content": "<unused1>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "9": {
+      "content": "<unused2>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "10": {
+      "content": "<unused3>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "11": {
+      "content": "<unused4>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "12": {
+      "content": "<unused5>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "13": {
+      "content": "<unused6>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "14": {
+      "content": "<unused7>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "15": {
+      "content": "<unused8>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "16": {
+      "content": "<unused9>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "17": {
+      "content": "<unused10>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "18": {
+      "content": "<unused11>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "19": {
+      "content": "<unused12>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "20": {
+      "content": "<unused13>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "21": {
+      "content": "<unused14>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "22": {
+      "content": "<unused15>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "23": {
+      "content": "<unused16>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "24": {
+      "content": "<unused17>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "25": {
+      "content": "<unused18>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "26": {
+      "content": "<unused19>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "27": {
+      "content": "<unused20>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "28": {
+      "content": "<unused21>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "29": {
+      "content": "<unused22>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "30": {
+      "content": "<unused23>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "31": {
+      "content": "<unused24>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "32": {
+      "content": "<unused25>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "33": {
+      "content": "<unused26>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "34": {
+      "content": "<unused27>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "35": {
+      "content": "<unused28>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "36": {
+      "content": "<unused29>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "37": {
+      "content": "<unused30>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "38": {
+      "content": "<unused31>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "39": {
+      "content": "<unused32>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "40": {
+      "content": "<unused33>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "41": {
+      "content": "<unused34>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "42": {
+      "content": "<unused35>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "43": {
+      "content": "<unused36>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "44": {
+      "content": "<unused37>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "45": {
+      "content": "<unused38>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "46": {
+      "content": "<unused39>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "47": {
+      "content": "<unused40>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "48": {
+      "content": "<unused41>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "49": {
+      "content": "<unused42>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "50": {
+      "content": "<unused43>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "51": {
+      "content": "<unused44>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "52": {
+      "content": "<unused45>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "53": {
+      "content": "<unused46>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "54": {
+      "content": "<unused47>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "55": {
+      "content": "<unused48>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "56": {
+      "content": "<unused49>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "57": {
+      "content": "<unused50>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "58": {
+      "content": "<unused51>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "59": {
+      "content": "<unused52>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "60": {
+      "content": "<unused53>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "61": {
+      "content": "<unused54>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "62": {
+      "content": "<unused55>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "63": {
+      "content": "<unused56>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "64": {
+      "content": "<unused57>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "65": {
+      "content": "<unused58>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "66": {
+      "content": "<unused59>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "67": {
+      "content": "<unused60>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "68": {
+      "content": "<unused61>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "69": {
+      "content": "<unused62>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "70": {
+      "content": "<unused63>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "71": {
+      "content": "<unused64>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "72": {
+      "content": "<unused65>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "73": {
+      "content": "<unused66>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "74": {
+      "content": "<unused67>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "75": {
+      "content": "<unused68>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "76": {
+      "content": "<unused69>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "77": {
+      "content": "<unused70>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "78": {
+      "content": "<unused71>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "79": {
+      "content": "<unused72>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "80": {
+      "content": "<unused73>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "81": {
+      "content": "<unused74>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "82": {
+      "content": "<unused75>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "83": {
+      "content": "<unused76>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "84": {
+      "content": "<unused77>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "85": {
+      "content": "<unused78>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "86": {
+      "content": "<unused79>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "87": {
+      "content": "<unused80>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "88": {
+      "content": "<unused81>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "89": {
+      "content": "<unused82>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "90": {
+      "content": "<unused83>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "91": {
+      "content": "<unused84>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "92": {
+      "content": "<unused85>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "93": {
+      "content": "<unused86>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "94": {
+      "content": "<unused87>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "95": {
+      "content": "<unused88>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "96": {
+      "content": "<unused89>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "97": {
+      "content": "<unused90>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "98": {
+      "content": "<unused91>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "99": {
+      "content": "<unused92>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "100": {
+      "content": "<unused93>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "101": {
+      "content": "<unused94>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "102": {
+      "content": "<unused95>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "103": {
+      "content": "<unused96>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "104": {
+      "content": "<unused97>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "105": {
+      "content": "<unused98>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "106": {
+      "content": "<start_of_turn>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "107": {
+      "content": "<end_of_turn>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "108": {
+      "content": "\n",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "109": {
+      "content": "\n\n",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "110": {
+      "content": "\n\n\n",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "111": {
+      "content": "\n\n\n\n",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "112": {
+      "content": "\n\n\n\n\n",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "113": {
+      "content": "\n\n\n\n\n\n",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "114": {
+      "content": "\n\n\n\n\n\n\n",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "115": {
+      "content": "\n\n\n\n\n\n\n\n",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "116": {
+      "content": "\n\n\n\n\n\n\n\n\n",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "117": {
+      "content": "\n\n\n\n\n\n\n\n\n\n",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "118": {
+      "content": "\n\n\n\n\n\n\n\n\n\n\n",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "119": {
+      "content": "\n\n\n\n\n\n\n\n\n\n\n\n",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "120": {
+      "content": "\n\n\n\n\n\n\n\n\n\n\n\n\n",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "121": {
+      "content": "\n\n\n\n\n\n\n\n\n\n\n\n\n\n",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "122": {
+      "content": "\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "123": {
+      "content": "\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "124": {
+      "content": "\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "125": {
+      "content": "\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "126": {
+      "content": "\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "127": {
+      "content": "\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "128": {
+      "content": "\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "129": {
+      "content": "\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "130": {
+      "content": "\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "131": {
+      "content": "\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "132": {
+      "content": "\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "133": {
+      "content": "\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "134": {
+      "content": "\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "135": {
+      "content": "\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "136": {
+      "content": "\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "137": {
+      "content": "\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "138": {
+      "content": "\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "139": {
+      "content": "▁▁",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "140": {
+      "content": "▁▁▁",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "141": {
+      "content": "▁▁▁▁",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "142": {
+      "content": "▁▁▁▁▁",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "143": {
+      "content": "▁▁▁▁▁▁",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "144": {
+      "content": "▁▁▁▁▁▁▁",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "145": {
+      "content": "▁▁▁▁▁▁▁▁",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "146": {
+      "content": "▁▁▁▁▁▁▁▁▁",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "147": {
+      "content": "▁▁▁▁▁▁▁▁▁▁",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "148": {
+      "content": "▁▁▁▁▁▁▁▁▁▁▁",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "149": {
+      "content": "▁▁▁▁▁▁▁▁▁▁▁▁",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "150": {
+      "content": "▁▁▁▁▁▁▁▁▁▁▁▁▁",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "151": {
+      "content": "▁▁▁▁▁▁▁▁▁▁▁▁▁▁",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "152": {
+      "content": "▁▁▁▁▁▁▁▁▁▁▁▁▁▁▁",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "153": {
+      "content": "▁▁▁▁▁▁▁▁▁▁▁▁▁▁▁▁",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "154": {
+      "content": "▁▁▁▁▁▁▁▁▁▁▁▁▁▁▁▁▁",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "155": {
+      "content": "▁▁▁▁▁▁▁▁▁▁▁▁▁▁▁▁▁▁",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "156": {
+      "content": "▁▁▁▁▁▁▁▁▁▁▁▁▁▁▁▁▁▁▁",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "157": {
+      "content": "▁▁▁▁▁▁▁▁▁▁▁▁▁▁▁▁▁▁▁▁",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "158": {
+      "content": "▁▁▁▁▁▁▁▁▁▁▁▁▁▁▁▁▁▁▁▁▁",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "159": {
+      "content": "▁▁▁▁▁▁▁▁▁▁▁▁▁▁▁▁▁▁▁▁▁▁",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "160": {
+      "content": "▁▁▁▁▁▁▁▁▁▁▁▁▁▁▁▁▁▁▁▁▁▁▁",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "161": {
+      "content": "▁▁▁▁▁▁▁▁▁▁▁▁▁▁▁▁▁▁▁▁▁▁▁▁",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "162": {
+      "content": "▁▁▁▁▁▁▁▁▁▁▁▁▁▁▁▁▁▁▁▁▁▁▁▁▁",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "163": {
+      "content": "▁▁▁▁▁▁▁▁▁▁▁▁▁▁▁▁▁▁▁▁▁▁▁▁▁▁",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "164": {
+      "content": "▁▁▁▁▁▁▁▁▁▁▁▁▁▁▁▁▁▁▁▁▁▁▁▁▁▁▁",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "165": {
+      "content": "▁▁▁▁▁▁▁▁▁▁▁▁▁▁▁▁▁▁▁▁▁▁▁▁▁▁▁▁",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "166": {
+      "content": "▁▁▁▁▁▁▁▁▁▁▁▁▁▁▁▁▁▁▁▁▁▁▁▁▁▁▁▁▁",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "167": {
+      "content": "▁▁▁▁▁▁▁▁▁▁▁▁▁▁▁▁▁▁▁▁▁▁▁▁▁▁▁▁▁▁",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "168": {
+      "content": "▁▁▁▁▁▁▁▁▁▁▁▁▁▁▁▁▁▁▁▁▁▁▁▁▁▁▁▁▁▁▁",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "169": {
+      "content": "<table>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "170": {
+      "content": "<caption>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "171": {
+      "content": "<thead>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "172": {
+      "content": "<tbody>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "173": {
+      "content": "<tfoot>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "174": {
+      "content": "<tr>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "175": {
+      "content": "<th>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "176": {
+      "content": "<td>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "177": {
+      "content": "</table>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "178": {
+      "content": "</caption>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "179": {
+      "content": "</thead>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "180": {
+      "content": "</tbody>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "181": {
+      "content": "</tfoot>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "182": {
+      "content": "</tr>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "183": {
+      "content": "</th>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "184": {
+      "content": "</td>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "185": {
+      "content": "<h1>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "186": {
+      "content": "<h2>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "187": {
+      "content": "<h3>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "188": {
+      "content": "<h4>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "189": {
+      "content": "<h5>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "190": {
+      "content": "<h6>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "191": {
+      "content": "<blockquote>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "192": {
+      "content": "</h1>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "193": {
+      "content": "</h2>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "194": {
+      "content": "</h3>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "195": {
+      "content": "</h4>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "196": {
+      "content": "</h5>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "197": {
+      "content": "</h6>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "198": {
+      "content": "</blockquote>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "199": {
+      "content": "<strong>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "200": {
+      "content": "<em>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "201": {
+      "content": "<b>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "202": {
+      "content": "<i>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "203": {
+      "content": "<u>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "204": {
+      "content": "<s>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "205": {
+      "content": "<sub>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "206": {
+      "content": "<sup>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "207": {
+      "content": "<code>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "208": {
+      "content": "</strong>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "209": {
+      "content": "</em>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "210": {
+      "content": "</b>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "211": {
+      "content": "</i>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "212": {
+      "content": "</u>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "213": {
+      "content": "</s>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "214": {
+      "content": "</sub>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "215": {
+      "content": "</sup>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "216": {
+      "content": "</code>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "255968": {
+      "content": "[toxicity=0]",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "255969": {
+      "content": "\t\t",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "255970": {
+      "content": "\t\t\t",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "255971": {
+      "content": "\t\t\t\t",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "255972": {
+      "content": "\t\t\t\t\t",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "255973": {
+      "content": "\t\t\t\t\t\t",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "255974": {
+      "content": "\t\t\t\t\t\t\t",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "255975": {
+      "content": "\t\t\t\t\t\t\t\t",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "255976": {
+      "content": "\t\t\t\t\t\t\t\t\t",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "255977": {
+      "content": "\t\t\t\t\t\t\t\t\t\t",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "255978": {
+      "content": "\t\t\t\t\t\t\t\t\t\t\t",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "255979": {
+      "content": "\t\t\t\t\t\t\t\t\t\t\t\t",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "255980": {
+      "content": "\t\t\t\t\t\t\t\t\t\t\t\t\t",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "255981": {
+      "content": "\t\t\t\t\t\t\t\t\t\t\t\t\t\t",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "255982": {
+      "content": "\t\t\t\t\t\t\t\t\t\t\t\t\t\t\t",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "255983": {
+      "content": "\t\t\t\t\t\t\t\t\t\t\t\t\t\t\t\t",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "255984": {
+      "content": "\t\t\t\t\t\t\t\t\t\t\t\t\t\t\t\t\t",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "255985": {
+      "content": "\t\t\t\t\t\t\t\t\t\t\t\t\t\t\t\t\t\t",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "255986": {
+      "content": "\t\t\t\t\t\t\t\t\t\t\t\t\t\t\t\t\t\t\t",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "255987": {
+      "content": "\t\t\t\t\t\t\t\t\t\t\t\t\t\t\t\t\t\t\t\t",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "255988": {
+      "content": "\t\t\t\t\t\t\t\t\t\t\t\t\t\t\t\t\t\t\t\t\t",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "255989": {
+      "content": "\t\t\t\t\t\t\t\t\t\t\t\t\t\t\t\t\t\t\t\t\t\t",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "255990": {
+      "content": "\t\t\t\t\t\t\t\t\t\t\t\t\t\t\t\t\t\t\t\t\t\t\t",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "255991": {
+      "content": "\t\t\t\t\t\t\t\t\t\t\t\t\t\t\t\t\t\t\t\t\t\t\t\t",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "255992": {
+      "content": "\t\t\t\t\t\t\t\t\t\t\t\t\t\t\t\t\t\t\t\t\t\t\t\t\t",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "255993": {
+      "content": "\t\t\t\t\t\t\t\t\t\t\t\t\t\t\t\t\t\t\t\t\t\t\t\t\t\t",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "255994": {
+      "content": "\t\t\t\t\t\t\t\t\t\t\t\t\t\t\t\t\t\t\t\t\t\t\t\t\t\t\t",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "255995": {
+      "content": "\t\t\t\t\t\t\t\t\t\t\t\t\t\t\t\t\t\t\t\t\t\t\t\t\t\t\t\t",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "255996": {
+      "content": "\t\t\t\t\t\t\t\t\t\t\t\t\t\t\t\t\t\t\t\t\t\t\t\t\t\t\t\t\t",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "255997": {
+      "content": "\t\t\t\t\t\t\t\t\t\t\t\t\t\t\t\t\t\t\t\t\t\t\t\t\t\t\t\t\t\t",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "255998": {
+      "content": "\t\t\t\t\t\t\t\t\t\t\t\t\t\t\t\t\t\t\t\t\t\t\t\t\t\t\t\t\t\t\t",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "255999": {
+      "content": "<unused99>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    }
+  },
+  "additional_special_tokens": [
+    "<start_of_turn>",
+    "<end_of_turn>"
+  ],
+  "bos_token": "<bos>",
+  "chat_template": "{{ bos_token }}{% if messages[0]['role'] == 'system' %}{{ raise_exception('System role not supported') }}{% endif %}{% for message in messages %}{% if (message['role'] == 'user') != (loop.index0 % 2 == 0) %}{{ raise_exception('Conversation roles must alternate user/assistant/user/assistant/...') }}{% endif %}{% if (message['role'] == 'assistant') %}{% set role = 'model' %}{% else %}{% set role = message['role'] %}{% endif %}{{ '<start_of_turn>' + role + '\n' + message['content'] | trim + '<end_of_turn>\n' }}{% endfor %}{% if add_generation_prompt %}{{'<start_of_turn>model\n'}}{% endif %}",
+  "clean_up_tokenization_spaces": false,
+  "eos_token": "<eos>",
+  "model_max_length": 1000000000000000019884624838656,
+  "pad_token": "<eos>",
+  "sp_model_kwargs": {},
+  "spaces_between_special_tokens": false,
+  "tokenizer_class": "GemmaTokenizer",
+  "unk_token": "<unk>",
+  "use_default_system_prompt": false
+}

trainer_state.json

@@ -0,0 +1,460 @@
+{
+  "best_metric": null,
+  "best_model_checkpoint": null,
+  "epoch": 3.5842293906810037,
+  "eval_steps": 1000,
+  "global_step": 6000,
+  "is_hyper_param_search": false,
+  "is_local_process_zero": true,
+  "is_world_process_zero": true,
+  "log_history": [
+    {
+      "epoch": 0.0005973715651135006,
+      "grad_norm": 8.9375,
+      "learning_rate": 2e-06,
+      "loss": 1.3168,
+      "step": 1
+    },
+    {
+      "epoch": 0.05973715651135006,
+      "grad_norm": 0.625,
+      "learning_rate": 0.0002,
+      "loss": 0.5498,
+      "step": 100
+    },
+    {
+      "epoch": 0.11947431302270012,
+      "grad_norm": 0.322265625,
+      "learning_rate": 0.0004,
+      "loss": 0.3884,
+      "step": 200
+    },
+    {
+      "epoch": 0.17921146953405018,
+      "grad_norm": 0.302734375,
+      "learning_rate": 0.0006,
+      "loss": 0.376,
+      "step": 300
+    },
+    {
+      "epoch": 0.23894862604540024,
+      "grad_norm": 0.12353515625,
+      "learning_rate": 0.0008,
+      "loss": 0.3635,
+      "step": 400
+    },
+    {
+      "epoch": 0.2986857825567503,
+      "grad_norm": 0.1826171875,
+      "learning_rate": 0.001,
+      "loss": 0.3559,
+      "step": 500
+    },
+    {
+      "epoch": 0.35842293906810035,
+      "grad_norm": 0.10888671875,
+      "learning_rate": 0.0012,
+      "loss": 0.3438,
+      "step": 600
+    },
+    {
+      "epoch": 0.41816009557945044,
+      "grad_norm": 0.11572265625,
+      "learning_rate": 0.0014,
+      "loss": 0.3412,
+      "step": 700
+    },
+    {
+      "epoch": 0.4778972520908005,
+      "grad_norm": 0.08984375,
+      "learning_rate": 0.0016,
+      "loss": 0.3392,
+      "step": 800
+    },
+    {
+      "epoch": 0.5376344086021505,
+      "grad_norm": 0.11376953125,
+      "learning_rate": 0.0018000000000000002,
+      "loss": 0.346,
+      "step": 900
+    },
+    {
+      "epoch": 0.5973715651135006,
+      "grad_norm": 0.08251953125,
+      "learning_rate": 0.002,
+      "loss": 0.3174,
+      "step": 1000
+    },
+    {
+      "epoch": 0.6571087216248507,
+      "grad_norm": 0.0703125,
+      "learning_rate": 0.0019998292504580526,
+      "loss": 0.3309,
+      "step": 1100
+    },
+    {
+      "epoch": 0.7168458781362007,
+      "grad_norm": 0.1533203125,
+      "learning_rate": 0.001999317060143023,
+      "loss": 0.323,
+      "step": 1200
+    },
+    {
+      "epoch": 0.7765830346475507,
+      "grad_norm": 0.09228515625,
+      "learning_rate": 0.001998463603967434,
+      "loss": 0.3157,
+      "step": 1300
+    },
+    {
+      "epoch": 0.8363201911589009,
+      "grad_norm": 0.072265625,
+      "learning_rate": 0.0019972691733857882,
+      "loss": 0.3155,
+      "step": 1400
+    },
+    {
+      "epoch": 0.8960573476702509,
+      "grad_norm": 0.12890625,
+      "learning_rate": 0.0019957341762950344,
+      "loss": 0.3148,
+      "step": 1500
+    },
+    {
+      "epoch": 0.955794504181601,
+      "grad_norm": 0.10205078125,
+      "learning_rate": 0.001993859136895274,
+      "loss": 0.3013,
+      "step": 1600
+    },
+    {
+      "epoch": 1.015531660692951,
+      "grad_norm": 0.11474609375,
+      "learning_rate": 0.0019916446955107426,
+      "loss": 0.3009,
+      "step": 1700
+    },
+    {
+      "epoch": 1.075268817204301,
+      "grad_norm": 0.12890625,
+      "learning_rate": 0.001989091608371146,
+      "loss": 0.3005,
+      "step": 1800
+    },
+    {
+      "epoch": 1.135005973715651,
+      "grad_norm": 0.087890625,
+      "learning_rate": 0.0019862007473534027,
+      "loss": 0.3028,
+      "step": 1900
+    },
+    {
+      "epoch": 1.1947431302270013,
+      "grad_norm": 0.10546875,
+      "learning_rate": 0.001982973099683902,
+      "loss": 0.3006,
+      "step": 2000
+    },
+    {
+      "epoch": 1.2544802867383513,
+      "grad_norm": 0.1044921875,
+      "learning_rate": 0.001979409767601366,
+      "loss": 0.2981,
+      "step": 2100
+    },
+    {
+      "epoch": 1.3142174432497014,
+      "grad_norm": 0.10888671875,
+      "learning_rate": 0.001975511967980437,
+      "loss": 0.2917,
+      "step": 2200
+    },
+    {
+      "epoch": 1.3739545997610514,
+      "grad_norm": 0.10107421875,
+      "learning_rate": 0.001971281031916114,
+      "loss": 0.2816,
+      "step": 2300
+    },
+    {
+      "epoch": 1.4336917562724014,
+      "grad_norm": 0.1337890625,
+      "learning_rate": 0.0019667184042691877,
+      "loss": 0.2754,
+      "step": 2400
+    },
+    {
+      "epoch": 1.4934289127837514,
+      "grad_norm": 0.08203125,
+      "learning_rate": 0.001961825643172819,
+      "loss": 0.2775,
+      "step": 2500
+    },
+    {
+      "epoch": 1.5531660692951015,
+      "grad_norm": 0.0634765625,
+      "learning_rate": 0.0019566044195004407,
+      "loss": 0.2756,
+      "step": 2600
+    },
+    {
+      "epoch": 1.6129032258064515,
+      "grad_norm": 0.1142578125,
+      "learning_rate": 0.0019510565162951536,
+      "loss": 0.27,
+      "step": 2700
+    },
+    {
+      "epoch": 1.6726403823178018,
+      "grad_norm": 0.09423828125,
+      "learning_rate": 0.0019451838281608197,
+      "loss": 0.2616,
+      "step": 2800
+    },
+    {
+      "epoch": 1.7323775388291518,
+      "grad_norm": 0.095703125,
+      "learning_rate": 0.0019389883606150567,
+      "loss": 0.2589,
+      "step": 2900
+    },
+    {
+      "epoch": 1.7921146953405018,
+      "grad_norm": 0.12353515625,
+      "learning_rate": 0.0019324722294043557,
+      "loss": 0.2567,
+      "step": 3000
+    },
+    {
+      "epoch": 1.8518518518518519,
+      "grad_norm": 0.138671875,
+      "learning_rate": 0.0019256376597815564,
+      "loss": 0.2462,
+      "step": 3100
+    },
+    {
+      "epoch": 1.911589008363202,
+      "grad_norm": 0.10205078125,
+      "learning_rate": 0.001918486985745923,
+      "loss": 0.246,
+      "step": 3200
+    },
+    {
+      "epoch": 1.971326164874552,
+      "grad_norm": 0.07861328125,
+      "learning_rate": 0.0019110226492460884,
+      "loss": 0.2378,
+      "step": 3300
+    },
+    {
+      "epoch": 2.031063321385902,
+      "grad_norm": 0.091796875,
+      "learning_rate": 0.0019032471993461289,
+      "loss": 0.2322,
+      "step": 3400
+    },
+    {
+      "epoch": 2.090800477897252,
+      "grad_norm": 0.09814453125,
+      "learning_rate": 0.0018951632913550625,
+      "loss": 0.2341,
+      "step": 3500
+    },
+    {
+      "epoch": 2.150537634408602,
+      "grad_norm": 0.0859375,
+      "learning_rate": 0.0018867736859200619,
+      "loss": 0.2314,
+      "step": 3600
+    },
+    {
+      "epoch": 2.2102747909199523,
+      "grad_norm": 0.09765625,
+      "learning_rate": 0.0018780812480836979,
+      "loss": 0.2297,
+      "step": 3700
+    },
+    {
+      "epoch": 2.270011947431302,
+      "grad_norm": 0.087890625,
+      "learning_rate": 0.0018690889463055284,
+      "loss": 0.224,
+      "step": 3800
+    },
+    {
+      "epoch": 2.3297491039426523,
+      "grad_norm": 0.0693359375,
+      "learning_rate": 0.0018597998514483724,
+      "loss": 0.2193,
+      "step": 3900
+    },
+    {
+      "epoch": 2.3894862604540026,
+      "grad_norm": 0.11669921875,
+      "learning_rate": 0.0018502171357296143,
+      "loss": 0.2158,
+      "step": 4000
+    },
+    {
+      "epoch": 2.4492234169653524,
+      "grad_norm": 0.07568359375,
+      "learning_rate": 0.0018403440716378927,
+      "loss": 0.2081,
+      "step": 4100
+    },
+    {
+      "epoch": 2.5089605734767026,
+      "grad_norm": 0.10498046875,
+      "learning_rate": 0.0018301840308155505,
+      "loss": 0.2092,
+      "step": 4200
+    },
+    {
+      "epoch": 2.5686977299880525,
+      "grad_norm": 0.10107421875,
+      "learning_rate": 0.0018197404829072212,
+      "loss": 0.2038,
+      "step": 4300
+    },
+    {
+      "epoch": 2.6284348864994027,
+      "grad_norm": 0.103515625,
+      "learning_rate": 0.0018090169943749475,
+      "loss": 0.199,
+      "step": 4400
+    },
+    {
+      "epoch": 2.6881720430107525,
+      "grad_norm": 0.087890625,
+      "learning_rate": 0.0017980172272802398,
+      "loss": 0.1955,
+      "step": 4500
+    },
+    {
+      "epoch": 2.7479091995221028,
+      "grad_norm": 0.0986328125,
+      "learning_rate": 0.0017867449380334832,
+      "loss": 0.1937,
+      "step": 4600
+    },
+    {
+      "epoch": 2.807646356033453,
+      "grad_norm": 0.1044921875,
+      "learning_rate": 0.0017752039761111298,
+      "loss": 0.1914,
+      "step": 4700
+    },
+    {
+      "epoch": 2.867383512544803,
+      "grad_norm": 0.0947265625,
+      "learning_rate": 0.001763398282741103,
+      "loss": 0.1875,
+      "step": 4800
+    },
+    {
+      "epoch": 2.927120669056153,
+      "grad_norm": 0.09765625,
+      "learning_rate": 0.0017513318895568735,
+      "loss": 0.1855,
+      "step": 4900
+    },
+    {
+      "epoch": 2.986857825567503,
+      "grad_norm": 0.08056640625,
+      "learning_rate": 0.001739008917220659,
+      "loss": 0.1795,
+      "step": 5000
+    },
+    {
+      "epoch": 3.046594982078853,
+      "grad_norm": 0.1259765625,
+      "learning_rate": 0.0017264335740162242,
+      "loss": 0.1742,
+      "step": 5100
+    },
+    {
+      "epoch": 3.106332138590203,
+      "grad_norm": 0.09228515625,
+      "learning_rate": 0.0017136101544117524,
+      "loss": 0.1759,
+      "step": 5200
+    },
+    {
+      "epoch": 3.166069295101553,
+      "grad_norm": 0.1259765625,
+      "learning_rate": 0.0017005430375932908,
+      "loss": 0.1756,
+      "step": 5300
+    },
+    {
+      "epoch": 3.225806451612903,
+      "grad_norm": 0.07470703125,
+      "learning_rate": 0.0016872366859692627,
+      "loss": 0.1766,
+      "step": 5400
+    },
+    {
+      "epoch": 3.2855436081242533,
+      "grad_norm": 0.087890625,
+      "learning_rate": 0.0016736956436465573,
+      "loss": 0.1666,
+      "step": 5500
+    },
+    {
+      "epoch": 3.3452807646356035,
+      "grad_norm": 0.0869140625,
+      "learning_rate": 0.0016599245348787228,
+      "loss": 0.169,
+      "step": 5600
+    },
+    {
+      "epoch": 3.4050179211469533,
+      "grad_norm": 0.08740234375,
+      "learning_rate": 0.0016459280624867873,
+      "loss": 0.1616,
+      "step": 5700
+    },
+    {
+      "epoch": 3.4647550776583036,
+      "grad_norm": 0.158203125,
+      "learning_rate": 0.001631711006253251,
+      "loss": 0.1595,
+      "step": 5800
+    },
+    {
+      "epoch": 3.5244922341696534,
+      "grad_norm": 0.078125,
+      "learning_rate": 0.001617278221289793,
+      "loss": 0.1626,
+      "step": 5900
+    },
+    {
+      "epoch": 3.5842293906810037,
+      "grad_norm": 0.0908203125,
+      "learning_rate": 0.0016026346363792565,
+      "loss": 0.1546,
+      "step": 6000
+    }
+  ],
+  "logging_steps": 100,
+  "max_steps": 18000,
+  "num_input_tokens_seen": 0,
+  "num_train_epochs": 11,
+  "save_steps": 3000,
+  "stateful_callbacks": {
+    "TrainerControl": {
+      "args": {
+        "should_epoch_stop": false,
+        "should_evaluate": true,
+        "should_log": false,
+        "should_save": true,
+        "should_training_stop": false
+      },
+      "attributes": {}
+    }
+  },
+  "total_flos": 1.1559951680731546e+17,
+  "train_batch_size": 6,
+  "trial_name": null,
+  "trial_params": null
+}

training_args.bin

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:91ac30b0e25fa5ad714bc4133367d871d355eb2cd2241f526ea5e733bbb709de
+size 5560

ultravox_config.py

@@ -0,0 +1,166 @@
+import dataclasses
+from enum import Enum
+from typing import Any, Dict, List, Optional
+
+import transformers
+
+
+@dataclasses.dataclass
+class LoraConfigSimplified:
+    """
+    Low Rank Approximation (LoRA) configuration.
+
+    Used for language and audio models separately.
+    """
+
+    # The rank of the approximation
+    r: int = 0
+    lora_alpha: float = 8
+    target_modules: Optional[List[str]] = dataclasses.field(
+        default_factory=lambda: ["k_proj", "q_proj", "linear_k", "linear_q"]
+    )
+
+
+class LossFunction(str, Enum):
+    CrossEntropy = "ce"
+    KL_Divergence = "kl"
+
+
+@dataclasses.dataclass
+class LossConfig:
+    loss_function: LossFunction = LossFunction.KL_Divergence
+    kl_temperature: float = 2.0
+
+    @property
+    def requires_alt_fields(self):
+        return self.loss_function == LossFunction.KL_Divergence
+
+
+class UltravoxConfig(transformers.PretrainedConfig):
+    r"""
+    This is the configuration class to store the configuration of a [`UltravoxForConditionalGeneration`]. It is used to instantiate an
+    Ultravox model according to the specified arguments, defining the model architecture.
+
+    Configuration objects inherit from [`PretrainedConfig`] and can be used to control the model outputs. Read the
+    documentation from [`PretrainedConfig`] for more information.
+
+    Args:
+        audio_config (`Wav2Vec2Config`,  *optional*):
+            Custom audio config or dict
+        text_config (`Union[AutoConfig, dict]`, *optional*):
+            The config object of the text backbone. Can be any of `LlamaConfig` or `MistralConfig`.
+        ignore_index (`int`, *optional*, defaults to -100):
+            The ignore index for the loss function.
+        audio_token_index (`int`, *optional*, defaults to 32000):
+            The audio token index to encode the audio prompt.
+        stack_factor (`int`, *optional*, defaults to 8):
+            Audio downsampling factor for the multimodal projector.
+        norm_init (`float`, *optional*, defaults to 0.4):
+            The initialization value for the layer normalization.
+        projector_act (`str`, *optional*, defaults to `"swiglu"`):
+            The activation function used by the multimodal projector.
+        text_model_lora_config (`LoraConfigSimplified`, *optional*):
+            The LoRA configuration for finetuning the text model.
+        audio_model_lora_config (`LoraConfigSimplified`, *optional*):
+            The LoRA configuration for finetuning the audio model.
+
+
+    Example:
+
+    ```python
+    >>> from transformers import UltravoxForConditionalGeneration, Wav2Vec2Config, UltravoxConfig, LlamaConfig
+
+    >>> # Initializing an audio encoder config
+    >>> audio_config = Wav2Vec2Config()
+
+    >>> # Initializing a Llama config
+    >>> text_config = LlamaConfig()
+
+    >>> # Initializing a default configuration
+    >>> configuration = UltravoxConfig(audio_config, text_config)
+
+    >>> # Initializing a completely untrained model from the configuration
+    >>> model = UltravoxForConditionalGeneration(configuration)
+
+    >>> # Accessing the model configuration
+    >>> configuration = model.config
+
+    >>> # Initialize a model from pretrained checkpoints and random projector weights
+    >>> config = UltravoxConfig(audio_model_id="facebook/wav2vec2-base-960h", text_model_id="meta-llama/Llama-2-7b-chat-hf")
+    ```"""
+
+    model_type = "ultravox"
+    is_composition = False
+
+    def __init__(
+        self,
+        audio_config: Optional[Dict[str, Any]] = None,
+        text_config: Optional[Dict[str, Any]] = None,
+        audio_model_id: Optional[str] = None,
+        text_model_id: Optional[str] = None,
+        ignore_index: int = -100,
+        hidden_size: int = 4096,
+        stack_factor: int = 8,
+        norm_init: float = 0.4,
+        projector_act: str = "swiglu",
+        text_model_lora_config: Optional[LoraConfigSimplified] = None,
+        audio_model_lora_config: Optional[LoraConfigSimplified] = None,
+        **kwargs,
+    ):
+        self.ignore_index = ignore_index
+
+        self.audio_model_id = audio_model_id
+        self.text_model_id = text_model_id
+
+        self.hidden_size = hidden_size
+        self.stack_factor = stack_factor
+        self.norm_init = norm_init
+        self.projector_act = projector_act
+
+        if text_model_id is not None:
+            self.text_config: transformers.LlamaConfig = (
+                transformers.AutoConfig.from_pretrained(text_model_id)
+            )
+        else:
+            text_config = text_config or {}
+            self.text_config = transformers.CONFIG_MAPPING[
+                text_config.get("model_type", "llama")
+            ](**text_config)
+
+        if audio_model_id is not None:
+            self.audio_config: transformers.PretrainedConfig = (
+                transformers.AutoConfig.from_pretrained(audio_model_id)
+            )
+        else:
+            audio_config = audio_config or {}
+            self.audio_config = transformers.CONFIG_MAPPING[
+                audio_config.get("model_type", "wav2vec2")
+            ](**audio_config)
+
+        self.text_model_lora_config = (
+            text_model_lora_config
+            if isinstance(text_model_lora_config, dict)
+            else dataclasses.asdict(text_model_lora_config or LoraConfigSimplified())
+        )
+        self.audio_model_lora_config = (
+            audio_model_lora_config
+            if isinstance(audio_model_lora_config, dict)
+            else dataclasses.asdict(audio_model_lora_config or LoraConfigSimplified())
+        )
+
+        self.vocab_size = self.text_config.vocab_size
+
+        self.initializer_range = self.text_config.initializer_range
+
+        super().__init__(**kwargs)
+
+    def to_diff_dict(self) -> Dict[str, Any]:
+        diff_dict = super().to_diff_dict()
+
+        # remove text_config and audio_config if text_model_id and audio_model_id are present
+        if self.text_model_id is not None:
+            diff_dict.pop("text_config", None)
+        if self.audio_model_id is not None:
+            diff_dict.pop("audio_config", None)
+
+        return diff_dict

ultravox_model.py

@@ -0,0 +1,669 @@
+import logging
+from typing import Any, Dict, Optional, Set, Tuple, Union
+
+import peft
+import torch
+import torch.nn as nn
+import torch.nn.functional as F
+import transformers
+import transformers.activations
+import transformers.modeling_outputs
+import transformers.models
+from transformers.models.whisper import modeling_whisper as whisper
+
+# We must use relative import in this directory to allow uploading to HF Hub
+# Even "from . import X" pattern doesn't work (undocumented and unclear why)
+from .ultravox_config import LossConfig
+from .ultravox_config import LossFunction
+from .ultravox_config import UltravoxConfig
+
+
+class UltravoxModel(transformers.LlamaPreTrainedModel):
+    """
+    The Ultravox model which consists of an audio encoder and a language model.
+
+    Audio input is processed by the audio encoder, then every `stack_factor` frames are stacked together and
+    projected to the language model's embedding space using a few linear layers.
+    The text is embedded by the language model as usual and then the audio and text embeddings are merged together.
+
+    A special token `<|audio|>` is used to indicate the start of the audio embeddings in the merged embeddings.
+
+    Parameters:
+        config: Model configuration class with all the parameters of the model.
+    """
+
+    config_class = UltravoxConfig
+    config: UltravoxConfig  # for type hinting
+    # Usually we load encoder and LLM weights from a pretrained model separately, so they are allowed to be missing
+    _keys_to_ignore_on_load_missing = ["audio_tower.*", "language_model.*"]
+
+    def __init__(self, config: UltravoxConfig):
+        super().__init__(config)
+        self._register_load_state_dict_pre_hook(self._pre_load_state_dict_hook)
+
+        self.keep_params: Set[str] = set()
+        self.vocab_size = config.vocab_size
+
+        self.audio_tower = self._create_audio_tower(config)
+        self.multi_modal_projector = self._create_multi_modal_projector(config)
+        self.language_model = self._create_language_model(config)
+
+        # Determine no_split_modules dynamically to use with FSDP auto_wrap policy.
+        # FSDP throws an error if some of the layer types are not found in the model.
+        # This would be something like ["LlamaDecoderLayer", "WhisperEncoderLayer"]
+        self._no_split_modules = (self.language_model._no_split_modules or []) + (
+            self.audio_tower._no_split_modules or []
+        )
+
+        self.loss_config = LossConfig()
+        self.post_init()
+
+    def get_input_embeddings(self):
+        return self.language_model.get_input_embeddings()
+
+    def set_input_embeddings(self, value):
+        self.language_model.set_input_embeddings(value)
+
+    def get_output_embeddings(self):
+        return self.language_model.get_output_embeddings()
+
+    def set_output_embeddings(self, new_embeddings):
+        self.language_model.set_output_embeddings(new_embeddings)
+
+    def set_decoder(self, decoder):
+        self.language_model.set_decoder(decoder)
+
+    def get_decoder(self):
+        return self.language_model.get_decoder()
+
+    def tie_weights(self):
+        return self.language_model.tie_weights()
+
+    def set_loss_config(self, loss_config: LossConfig):
+        self.loss_config = loss_config
+
+    def _setup_cache(
+        self, cache_cls, max_batch_size: int, max_cache_len: Optional[int] = None
+    ):
+        self.language_model._setup_cache(cache_cls, max_batch_size, max_cache_len)
+
+    def _reorder_cache(self, past_key_values, beam_idx):
+        return self.language_model._reorder_cache(past_key_values, beam_idx)
+
+    def resize_token_embeddings(
+        self,
+        new_num_tokens: Optional[int] = None,
+        pad_to_multiple_of: Optional[int] = None,
+    ) -> nn.Embedding:
+        model_embeds = self.language_model.resize_token_embeddings(
+            new_num_tokens, pad_to_multiple_of
+        )
+        # update vocab size
+        self.config.text_config.vocab_size = model_embeds.num_embeddings
+        self.config.vocab_size = model_embeds.num_embeddings
+        self.vocab_size = model_embeds.num_embeddings
+        return model_embeds
+
+    def _compute_kl_loss(
+        self,
+        lm_output: transformers.modeling_outputs.CausalLMOutputWithPast,
+        labels: Optional[torch.Tensor] = None,
+        past_key_values: Optional[Union[Tuple, transformers.cache_utils.Cache]] = None,
+        alt_input_ids: Optional[torch.Tensor] = None,
+        alt_attention_mask: Optional[torch.Tensor] = None,
+        alt_labels: Optional[torch.Tensor] = None,
+        **kwargs,
+    ):
+        # disable gradient computation for the teacher model
+        with torch.no_grad():
+            # compute the teacher (text-only) model's distribution
+            alt_inputs_embeds = self.get_input_embeddings().forward(alt_input_ids)
+            alt_lm_output = self.language_model.forward(
+                inputs_embeds=alt_inputs_embeds,
+                labels=alt_labels,
+                attention_mask=alt_attention_mask,
+                past_key_values=past_key_values,
+                **kwargs,
+            )
+        # compute the KL divergence loss between the two models
+        kl_loss = F.kl_div(
+            F.log_softmax(
+                lm_output.logits[labels != -100] / self.loss_config.kl_temperature,
+                dim=-1,
+            ),
+            F.softmax(
+                alt_lm_output.logits[alt_labels != -100]
+                / self.loss_config.kl_temperature,
+                dim=-1,
+            ),
+            reduction="batchmean",
+        )
+        return {"loss": kl_loss}
+
+    def forward(
+        self,
+        input_ids: torch.Tensor,
+        audio_values: Optional[torch.FloatTensor] = None,
+        inputs_embeds: Optional[torch.FloatTensor] = None,
+        labels: Optional[torch.Tensor] = None,
+        attention_mask: Optional[torch.Tensor] = None,
+        audio_token_start_idx: Optional[torch.Tensor] = None,
+        audio_len: Optional[torch.Tensor] = None,
+        audio_token_len: Optional[torch.Tensor] = None,
+        past_key_values: Optional[Union[Tuple, transformers.cache_utils.Cache]] = None,
+        # the alt_* fields are needed for KL divergence loss
+        alt_input_ids: Optional[torch.Tensor] = None,
+        alt_attention_mask: Optional[torch.Tensor] = None,
+        alt_labels: Optional[torch.Tensor] = None,
+        **kwargs,
+    ) -> Union[Tuple, transformers.modeling_outputs.CausalLMOutputWithPast]:
+        """
+        Forward pass for the Ultravox model.
+
+        `input_ids` are the tokenized text input. They are embedded by the language model as usual.
+        `audio_values` are processed by the audio encoder and then every `stack_factor` frames are stacked together and
+        projected to the language model's embedding space using a few linear layers.
+        The audio and text embeddings are merged together. A special token `<|audio|>` is used to indicate the start
+        of the audio embeddings in the merged embeddings.
+
+        Args:
+            input_ids: The tokenized text input.
+            audio_values: The processed audio values.
+            inputs_embeds: The embeddings for the input tokens.
+            labels: The tokenized text labels.
+            attention_mask: The attention mask for the input.
+            position_ids: The position ids for the input.
+            past_key_values: The past key value cache for the language model attention layers.
+            **kwargs: Additional keyword arguments. Passed directly to the language model.
+        """
+        if inputs_embeds is None:
+            # B x T  ->  B x T x D
+            inputs_embeds = self.get_input_embeddings().forward(input_ids)
+
+        if audio_values is not None:
+            assert (
+                audio_token_start_idx is not None and audio_token_len is not None
+            ), "audio_token_start_idx and audio_token_len must be provided if audio_values are provided."
+            assert (
+                len(audio_token_start_idx) == len(audio_token_len) == len(audio_values)
+            ), "audio_token_start_idx, audio_token_len, and audio_values must have the same batch size."
+
+            # B x A/3200 x D
+            audio_tower_output = self.audio_tower.forward(
+                audio_values.to(self.audio_tower.dtype), audio_len=audio_len
+            ).last_hidden_state
+            audio_tower_output = audio_tower_output.to(inputs_embeds.dtype)
+
+            audio_embeds = self.multi_modal_projector.forward(audio_tower_output)
+
+            # combine audio and text embeddings
+            for i, (audio, start, length) in enumerate(
+                zip(audio_embeds, audio_token_start_idx, audio_token_len)
+            ):
+                length = min(length, audio.shape[0])
+                inputs_embeds[i, start : start + length] = audio[:length]
+
+        lm_output = self.language_model.forward(
+            inputs_embeds=inputs_embeds,
+            labels=labels,
+            attention_mask=attention_mask,
+            past_key_values=past_key_values,
+            **kwargs,
+        )
+        if self.training:
+            if self.loss_config.loss_function == LossFunction.CrossEntropy:
+                return lm_output
+            elif self.loss_config.loss_function == LossFunction.KL_Divergence:
+                return self._compute_kl_loss(
+                    lm_output=lm_output,
+                    labels=labels,
+                    past_key_values=past_key_values,
+                    alt_input_ids=alt_input_ids,
+                    alt_attention_mask=alt_attention_mask,
+                    alt_labels=alt_labels,
+                    **kwargs,
+                )
+            else:
+                raise ValueError(
+                    f"Unsupported loss function: {self.loss_config.loss_function}"
+                )
+        else:
+            return lm_output
+
+    def prepare_inputs_for_generation(
+        self,
+        input_ids: torch.Tensor,
+        audio_values: Optional[torch.FloatTensor] = None,
+        audio_token_start_idx: Optional[torch.Tensor] = None,
+        audio_token_len: Optional[torch.Tensor] = None,
+        audio_len: Optional[torch.Tensor] = None,
+        past_key_values: Optional[Union[Tuple, transformers.cache_utils.Cache]] = None,
+        attention_mask: Optional[torch.Tensor] = None,
+        inputs_embeds: Optional[torch.Tensor] = None,
+        cache_position: Optional[torch.Tensor] = None,
+        **kwargs,
+    ) -> Dict[str, Any]:
+        model_input = self.language_model.prepare_inputs_for_generation(
+            input_ids=input_ids,
+            past_key_values=past_key_values,
+            attention_mask=attention_mask,
+            inputs_embeds=inputs_embeds,
+            cache_position=cache_position,
+            **kwargs,
+        )
+
+        # include audio information in model_input only when it is needed during prefilling
+        # audio_token_start_idx should always be relative to the current cache position
+        prefill_start_idx = 0 if cache_position is None else cache_position[0]
+        if (
+            audio_values is not None
+            and audio_token_start_idx is not None
+            and prefill_start_idx <= torch.max(audio_token_start_idx)
+        ):
+            model_input["audio_values"] = audio_values
+            model_input["audio_token_start_idx"] = (
+                audio_token_start_idx - prefill_start_idx
+            )
+            model_input["audio_token_len"] = audio_token_len
+            model_input["audio_len"] = audio_len
+
+        return model_input
+
+    @classmethod
+    def _create_multi_modal_projector(
+        cls, config: UltravoxConfig
+    ) -> "UltravoxProjector":
+        projector = UltravoxProjector(config)
+        projector.to(config.torch_dtype)
+        return projector
+
+    @classmethod
+    def _create_audio_tower(
+        cls, config: UltravoxConfig
+    ) -> Union[transformers.Wav2Vec2Model, "ModifiedWhisperEncoder"]:
+        if config.audio_model_id is not None:
+            if "whisper" in config.audio_model_id is not None:
+                audio_tower = ModifiedWhisperEncoder.from_pretrained(
+                    config.audio_model_id, torch_dtype=config.torch_dtype
+                )
+            else:
+                audio_tower = transformers.AutoModel.from_pretrained(
+                    config.audio_model_id, torch_dtype=config.torch_dtype
+                )
+        else:
+            if "whisper" in config.audio_config._name_or_path:
+                audio_tower = ModifiedWhisperEncoder(config.audio_config)
+            else:
+                with transformers.modeling_utils.no_init_weights():
+                    # we only ever use from_config if the weights are retrained, hence initializing is not
+                    # required. This makes the model quite creation faster since init on CPU is quite slow.
+                    audio_tower = transformers.AutoModel.from_config(
+                        config.audio_config
+                    )
+
+        if isinstance(
+            audio_tower,
+            (transformers.Wav2Vec2BertModel, transformers.WhisperModel),
+        ):
+            # For these models we only need the encoder part
+            # Wav2Vec2BertModel -> Wav2Vec2BertEncoder
+            # WhisperModel -> WhisperEncoder
+            audio_tower = audio_tower.encoder
+
+        audio_tower = apply_lora(audio_tower, config.audio_model_lora_config)
+        return audio_tower
+
+    @classmethod
+    def _create_language_model(
+        cls, config: UltravoxConfig
+    ) -> transformers.LlamaForCausalLM:
+        if config.text_model_id is not None:
+            language_model = transformers.AutoModelForCausalLM.from_pretrained(
+                config.text_model_id,
+                attn_implementation=config._attn_implementation,
+                torch_dtype=config.torch_dtype,
+                load_in_4bit=True,
+            )
+        else:
+            with transformers.modeling_utils.no_init_weights():
+                # we only ever use from_config if the weights are retrained, hence initializing is not
+                # required. This makes the model quite creation faster since init on CPU is quite slow.
+                language_model = transformers.AutoModelForCausalLM.from_config(
+                    config.text_config,
+                    attn_implementation=config._attn_implementation,
+                    torch_dtype=config.torch_dtype,
+                )
+
+        language_model = apply_lora(language_model, config.text_model_lora_config)
+        return language_model
+
+    def merge_and_unload(self):
+        if isinstance(self.language_model, peft.PeftModel):
+            self.language_model = self.language_model.merge_and_unload()
+            # no need to download base language model weights anymore, so we can remove the id
+            self.config.text_model_id = None
+            self.keep_params.update(
+                set(
+                    [
+                        f"language_model.{name}"
+                        for name, _ in self.language_model.named_parameters()
+                    ]
+                )
+            )
+
+        if isinstance(self.audio_tower, peft.PeftModel):
+            self.audio_tower = self.audio_tower.merge_and_unload()
+            # no need to download base audio model weights anymore, so we can remove the id
+            self.config.audio_model_id = None
+            self.keep_params.update(
+                set(
+                    [
+                        f"audio_tower.{name}"
+                        for name, _ in self.audio_tower.named_parameters()
+                    ]
+                )
+            )
+
+        for param in ["text_model_lora_config", "audio_model_lora_config"]:
+            if hasattr(self.config, param):
+                delattr(self.config, param)
+
+    def push_to_hub(self, *args, **kwargs):
+        self.merge_and_unload()
+        self.to(self.language_model.dtype)
+        return super().push_to_hub(*args, **kwargs)
+
+    def save_pretrained(
+        self, *args, state_dict: Optional[Dict[str, Any]] = None, **kwargs
+    ):
+        if state_dict is None:
+            state_dict = super().state_dict()
+
+        named_params = dict(self.named_parameters())
+
+        state_dict = {
+            k: v
+            for k, v in state_dict.items()
+            if k in self.keep_params
+            or (k in named_params and named_params[k].requires_grad)
+        }
+
+        super().save_pretrained(*args, state_dict=state_dict, **kwargs)
+
+    def _pre_load_state_dict_hook(self, state_dict: Dict[str, Any], *args, **kwargs):
+        self.keep_params.update(set(state_dict.keys()))
+
+    def print_trainable_parameters(self):
+        """
+        Prints the number of trainable parameters in the model (reuses Peft model's method)
+        """
+        count_params = peft.peft_model.PeftModel.get_nb_trainable_parameters
+
+        trainable_params, all_param = count_params(self)
+
+        logging.info(
+            f"trainable params: {trainable_params:,d} || all params: {all_param:,d}"
+            f" || trainable%: {100 * trainable_params / all_param:.1f}%"
+        )
+
+        lm_trainable_params, lm_all_params = count_params(self.language_model)
+        audio_trainable_params, audio_all_params = count_params(self.audio_tower)
+
+        projector_trainable_params = (
+            trainable_params - lm_trainable_params - audio_trainable_params
+        )
+        projector_all_params = all_param - lm_all_params - audio_all_params
+
+        logging.info(
+            f"Trainable%:   "
+            f" LLM: {100 * lm_trainable_params / lm_all_params:.1f}%"
+            f" || Audio Encoder: {100 * audio_trainable_params / audio_all_params:.1f}%"
+            f" || Projector: {100 * projector_trainable_params / projector_all_params:.1f}%"
+        )
+
+
+def is_cache_empty(
+    past_key_values: Optional[Union[Tuple, transformers.cache_utils.Cache]]
+) -> bool:
+    """
+    Check if the cache is empty.
+    """
+    if past_key_values is None:
+        return True
+    if isinstance(past_key_values, tuple):
+        return all(len(c) == 0 for c in past_key_values)
+    return past_key_values.get_seq_length() == 0
+
+
+def apply_lora(model: torch.nn.Module, lora_config: dict) -> torch.nn.Module:
+    """
+    Applies LoRA finetuning to the model. If the `r` parameter is set to 0, the model is frozen instead.
+    """
+    lora_config = peft.LoraConfig(**lora_config or {})
+
+    if lora_config.r == 0:
+        # freeze the model entirely
+        for param in model.parameters():
+            param.requires_grad = False
+    else:
+        model = peft.get_peft_model(model, lora_config)
+
+    return model
+
+
+class StackAudioFrames(nn.Module):
+    """
+    Stack the audio embedding frames to reduce the sequence length by a factor of `stack_factor`.
+
+    The number of output frames will be `ceil(T / stack_factor) + 1` where `T` is the number of input frames.
+    NOTE: the extra +1 is intentional: in case the number of audio tokens are over-estimated by the processor,
+    we want to make sure `processor.audio_token_replacement` (i.e. EOS) doesn't get leaked into the middle of embeddings.
+    In most cases this extra padding will get removed in the model's forward function so it has no effect.
+    """
+
+    def __init__(self, stack_factor: int = 8):
+        super().__init__()
+        self.stack_factor = stack_factor
+
+    def forward(self, audio_embeds: torch.Tensor) -> torch.Tensor:
+        B, T, C = audio_embeds.shape
+        T_pad = (T + self.stack_factor - 1) // self.stack_factor * self.stack_factor
+        audio_embeds = F.pad(audio_embeds, (0, 0, 0, T_pad - T + self.stack_factor))
+        B, T, C = audio_embeds.shape
+        audio_embeds = audio_embeds.view(
+            B, T // self.stack_factor, C * self.stack_factor
+        )
+        return audio_embeds
+
+
+class RMSNorm(transformers.models.llama.modeling_llama.LlamaRMSNorm):
+    def __init__(self, hidden_size: int, init: float = 1, eps: float = 1e-6):
+        super().__init__(hidden_size=hidden_size, eps=eps)
+        self.weight.data.fill_(init)
+
+
+class SwiGLU(nn.Module):
+    def forward(self, x):
+        x, gate = x.chunk(2, dim=-1)
+        return F.silu(gate) * x
+
+
+class UltravoxProjector(nn.Sequential):
+    def __init__(self, config: UltravoxConfig):
+        super().__init__()
+        self.hidden_dim = config.hidden_size
+        self._pad_and_stack = StackAudioFrames(config.stack_factor)
+        dim = config.audio_config.hidden_size * config.stack_factor
+        self.ln_pre = RMSNorm(dim, init=config.norm_init)
+        self.linear_1 = nn.Linear(dim, self.hidden_dim, bias=False)
+        dim = self.hidden_dim
+        self.act = transformers.activations.get_activation(config.projector_act)
+        dim = dim // 2 if config.projector_act == "swiglu" else dim
+        self.linear_2 = nn.Linear(dim, config.text_config.hidden_size, bias=False)
+        self.ln_post = RMSNorm(config.text_config.hidden_size, init=config.norm_init)
+
+    def forward(self, audio_features: torch.Tensor) -> torch.Tensor:
+        audio_features = self._pad_and_stack(audio_features)
+        audio_features = self.ln_pre(audio_features)
+        hidden_states = self.linear_1(audio_features)
+        hidden_states = self.act(hidden_states)
+        hidden_states = self.linear_2(hidden_states)
+        hidden_states = self.ln_post(hidden_states)
+        return hidden_states
+
+
+class ModifiedWhisperEncoder(
+    whisper.WhisperEncoder, transformers.modeling_utils.ModuleUtilsMixin
+):
+    """
+    Encoder portion of OpenAI's Whisper model.
+
+    This implementation is a slightly modified version of HF Transformers' Whisper Encoder, with only a few fixes:
+    1. base_model_prefix updated to allow for doing `.from_pretrained` directly on the encoder
+    2. allow less than 30 second of audio padding to be passed in:
+        - relaxed ValueError check for `input_features` length to be less than or equal to `expected_seq_length` instead of strictly equal
+        - embed_pos is now sliced to match the length of `inputs_embeds`
+
+    Original: https://github.com/huggingface/transformers/blob/main/src/transformers/models/whisper/modeling_whisper.py
+    """
+
+    base_model_prefix = "model.encoder"
+    _no_split_modules = ["WhisperEncoderLayer"]
+
+    def forward(
+        self,
+        input_features,
+        audio_len=None,
+        head_mask=None,
+        output_attentions=None,
+        output_hidden_states=None,
+        return_dict=None,
+    ):
+        expected_seq_length = (
+            self.config.max_source_positions
+            * self.conv1.stride[0]
+            * self.conv2.stride[0]
+        )
+        if input_features.shape[-1] > expected_seq_length:
+            raise ValueError(
+                f"Whisper expects the mel input features to be of length {expected_seq_length} or less, but found {input_features.shape[-1]}. Make sure to pad the input mel features to {expected_seq_length}."
+            )
+
+        output_attentions = (
+            output_attentions
+            if output_attentions is not None
+            else self.config.output_attentions
+        )
+        output_hidden_states = (
+            output_hidden_states
+            if output_hidden_states is not None
+            else self.config.output_hidden_states
+        )
+        return_dict = (
+            return_dict if return_dict is not None else self.config.use_return_dict
+        )
+        inputs_embeds = nn.functional.gelu(self.conv1(input_features))
+        inputs_embeds = nn.functional.gelu(self.conv2(inputs_embeds))
+
+        inputs_embeds = inputs_embeds.permute(0, 2, 1)
+        embed_pos = self.embed_positions.weight[: inputs_embeds.size(-2)]
+
+        hidden_states = inputs_embeds + embed_pos
+        hidden_states = nn.functional.dropout(
+            hidden_states, p=self.dropout, training=self.training
+        )
+
+        encoder_states = () if output_hidden_states else None
+        all_attentions = () if output_attentions else None
+
+        # Create attention mask based on audio lengths to mask out padding tokens
+        # For each sample in batch:
+        # - Convert raw audio length to feature length after convolutions
+        # - Create boolean mask that is True for valid positions and False for padding
+        # - Convert to extended attention mask format expected by transformer layers
+        #   (1.0 for positions to attend to, large negative for positions to ignore)
+        # This masking ensures consistent behavior between training and inference
+        # by preventing the model from attending to padding tokens in both cases
+        attention_mask = None
+        if audio_len != None:
+            audio_feature_len = self._get_feat_extract_output_lengths(audio_len)
+            batch_size = hidden_states.shape[0]
+            max_seq_len = hidden_states.shape[1]
+            attention_mask = (
+                torch.arange(max_seq_len, device=hidden_states.device)[None, :]
+                .expand(batch_size, -1)
+                .lt(audio_feature_len.view(batch_size, 1))
+            )
+            attention_mask = self.get_extended_attention_mask(
+                attention_mask,
+                None,
+                device=hidden_states.device,
+                dtype=hidden_states.dtype,
+            )
+
+        # check if head_mask has a correct number of layers specified if desired
+        if head_mask is not None:
+            assert head_mask.size()[0] == (
+                len(self.layers)
+            ), f"The head_mask should be specified for {len(self.layers)} layers, but it is for {head_mask.size()[0]}."
+
+        for idx, encoder_layer in enumerate(self.layers):
+            if output_hidden_states:
+                encoder_states = encoder_states + (hidden_states,)
+            # add LayerDrop (see https://arxiv.org/abs/1909.11556 for description)
+            to_drop = False
+            if self.training:
+                dropout_probability = torch.rand([])
+                if dropout_probability < self.layerdrop:  # skip the layer
+                    to_drop = True
+
+            if to_drop:
+                layer_outputs = (None, None)
+            else:
+                if self.gradient_checkpointing and self.training:
+                    layer_outputs = self._gradient_checkpointing_func(
+                        encoder_layer.__call__,
+                        hidden_states,
+                        attention_mask,
+                        (head_mask[idx] if head_mask is not None else None),
+                        output_attentions,
+                    )
+                else:
+                    layer_outputs = encoder_layer(
+                        hidden_states,
+                        attention_mask,
+                        layer_head_mask=(
+                            head_mask[idx] if head_mask is not None else None
+                        ),
+                        output_attentions=output_attentions,
+                    )
+
+                hidden_states = layer_outputs[0]
+
+            if output_attentions:
+                all_attentions = all_attentions + (layer_outputs[1],)
+
+        hidden_states = self.layer_norm(hidden_states)
+        if output_hidden_states:
+            encoder_states = encoder_states + (hidden_states,)
+
+        if not return_dict:
+            return tuple(
+                v
+                for v in [hidden_states, encoder_states, all_attentions]
+                if v is not None
+            )
+        return transformers.modeling_outputs.BaseModelOutput(
+            last_hidden_state=hidden_states,
+            hidden_states=encoder_states,
+            attentions=all_attentions,
+        )
+
+
+UltravoxConfig.register_for_auto_class()
+UltravoxModel.register_for_auto_class()
+
+transformers.AutoConfig.register("ultravox", UltravoxConfig)
+transformers.AutoModel.register(UltravoxConfig, UltravoxModel)
+
+transformers.activations.ACT2FN["swiglu"] = SwiGLU