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

README.md

@@ -0,0 +1,128 @@
+---
+library_name: transformers
+pipeline_tag: text-generation
+inference: true
+widget:
+  - text: Hello!
+    example_title: Hello world
+    group: Python
+---
+
+This tiny model is for debugging. It is randomly initialized with the config adapted from [meta-llama/Llama-4-Maverick-17B-128E-Instruct](https://huggingface.co/meta-llama/Llama-4-Maverick-17B-128E-Instruct).
+
+### Example usage:
+
+```python
+import torch
+
+from transformers import AutoProcessor, Llama4ForConditionalGeneration
+
+model_id = "yujiepan/llama-4-tiny-random"
+processor = AutoProcessor.from_pretrained(model_id)
+model = Llama4ForConditionalGeneration.from_pretrained(
+    model_id,
+    attn_implementation="sdpa",  # flex attention / flash_attention_2 do not work, debugging...
+    device_map="auto",
+    torch_dtype=torch.bfloat16,
+)
+
+url1 = "https://huggingface.co/datasets/huggingface/documentation-images/resolve/0052a70beed5bf71b92610a43a52df6d286cd5f3/diffusers/rabbit.jpg"
+url2 = "https://huggingface.co/datasets/huggingface/documentation-images/resolve/main/datasets/cat_style_layout.png"
+messages = [
+    {
+        "role": "user",
+        "content": [
+            {"type": "image", "url": url1},
+            {"type": "image", "url": url2},
+            {"type": "text", "text": "Can you describe how these two images are similar, and how they differ?"},
+        ]
+    },
+]
+
+inputs = processor.apply_chat_template(
+    messages,
+    add_generation_prompt=True,
+    tokenize=True,
+    return_dict=True,
+    return_tensors="pt",
+).to(model.device)
+
+outputs = model.generate(
+    **inputs,
+    max_new_tokens=32,
+)
+
+response = processor.batch_decode(outputs[:, inputs["input_ids"].shape[-1]:])[0]
+print(response)
+print(outputs[0])
+```
+
+### Codes to create this repo:
+
+```python
+import json
+
+import torch
+
+from huggingface_hub import hf_hub_download
+from transformers import (
+    AutoConfig,
+    AutoModelForCausalLM,
+    AutoProcessor,
+    AutoTokenizer,
+    GenerationConfig,
+    Llama4ForConditionalGeneration,
+    pipeline,
+    set_seed,
+)
+
+source_model_id = "meta-llama/Llama-4-Maverick-17B-128E-Instruct"
+save_folder = "/tmp/yujiepan/llama-4-tiny-random"
+
+processor = AutoProcessor.from_pretrained(source_model_id)
+processor.save_pretrained(save_folder)
+
+with open(hf_hub_download(source_model_id, filename='config.json', repo_type='model'), 'r') as f:
+    config_json = json.load(f)
+config_json["text_config"]["num_hidden_layers"] = 4  # ensure to trigger no-rope & moe
+config_json["text_config"]["hidden_size"] = 32
+config_json["text_config"]["head_dim"] = 32  # vllm requires dim >= 32
+config_json["text_config"]["num_attention_heads"] = 1
+config_json["text_config"]["num_key_value_heads"] = 1
+config_json['text_config']["use_qk_norm"] = True
+config_json["text_config"]["intermediate_size"] = 64
+config_json["text_config"]["intermediate_size_mlp"] = 128
+config_json["text_config"]["num_local_experts"] = 8
+config_json["text_config"]["tie_word_embeddings"] = True
+
+config_json["vision_config"]["num_hidden_layers"] = 2
+config_json["vision_config"]["hidden_size"] = 32
+config_json["vision_config"]["intermediate_size"] = 128
+assert config_json["vision_config"]["intermediate_size"] == int(
+    config_json["vision_config"]["hidden_size"] // config_json["vision_config"]["pixel_shuffle_ratio"] ** 2
+)
+config_json["vision_config"]["num_attention_heads"] = 1
+config_json["vision_config"]["projector_input_dim"] = 32
+config_json["vision_config"]["projector_output_dim"] = 32
+config_json["vision_config"]["vision_output_dim"] = 32
+with open(f"{save_folder}/config.json", "w") as f:
+    json.dump(config_json, f, indent=2)
+
+config = AutoConfig.from_pretrained(
+    save_folder,
+)
+print(config)
+torch.set_default_dtype(torch.bfloat16)
+model = Llama4ForConditionalGeneration(config)
+torch.set_default_dtype(torch.float32)
+model.generation_config = GenerationConfig.from_pretrained(
+    source_model_id, trust_remote_code=True,
+)
+set_seed(42)
+with torch.no_grad():
+    for name, p in sorted(model.named_parameters()):
+        torch.nn.init.normal_(p, 0, 0.5)
+        print(name, p.shape)
+        pass
+model.save_pretrained(save_folder)
+```

chat_template.json

@@ -0,0 +1,3 @@
+{
+  "chat_template": "{{- bos_token }}\n{%- if custom_tools is defined %}\n    {%- set tools = custom_tools %}\n{%- endif %}\n{%- if not tools_in_user_message is defined %}\n    {%- set tools_in_user_message = true %}\n{%- endif %}\n{%- if not date_string is defined %}\n    {%- if strftime_now is defined %}\n        {%- set date_string = strftime_now(\"%d %b %Y\") %}\n    {%- else %}\n        {%- set date_string = \"26 Jul 2024\" %}\n    {%- endif %}\n{%- endif %}\n{%- if not tools is defined %}\n    {%- set tools = none %}\n{%- endif %}\n\n{#- This block extracts the system message, so we can slot it into the right place. #}\n{%- if messages[0]['role'] == 'system' %}    \n    {%- if messages[0]['content'] is string %}\n        {%- set system_message = messages[0]['content']|trim %}\n    {%- else %}\n        {#- FIXME: The processor requires an array, always. #}\n        {%- set system_message = messages[0]['content'][0]['text']|trim %}\n    {%- endif %}\n    {%- set messages = messages[1:] %}\n    {%- set user_supplied_system_message = true %}\n{%- else %}\n    {%- set system_message = \"\" %}\n    {%- set user_supplied_system_message = false %}\n{%- endif %}\n\n{#- System message if the user supplied one #}\n{%- if user_supplied_system_message %}\n    {{- \"<|header_start|>system<|header_end|>\\n\\n\" }}\n    {%- if tools is not none %}\n        {{- \"Environment: ipython\\n\" }}\n    {%- endif %}\n    {%- if tools is not none and not tools_in_user_message %}\n        {{- \"You have access to the following functions. To call a function, please respond with JSON for a function call.\" }}\n        {{- 'Respond in the format {\"name\": function name, \"parameters\": dictionary of argument name and its value}.' }}\n        {{- \"Do not use variables.\\n\\n\" }}\n        {%- for t in tools %}\n            {{- t | tojson(indent=4) }}\n            {{- \"\\n\\n\" }}\n        {%- endfor %}\n    {%- endif %}\n    {{- system_message }}\n    {{- \"<|eot|>\" }}\n{%- endif %}\n\n{#- Custom tools are passed in a user message with some extra guidance #}\n{%- if tools_in_user_message and not tools is none %}\n    {#- Extract the first user message so we can plug it in here #}\n    {%- if messages | length != 0 %}\n        {%- set first_user_message = messages[0]['content']|trim %}\n        {%- set messages = messages[1:] %}\n    {%- else %}\n        {{- raise_exception(\"Cannot put tools in the first user message when there's no first user message!\") }}\n{%- endif %}\n    {{- '<|header_start|>user<|header_end|>\\n\\n' -}}\n    {{- \"Given the following functions, please respond with a JSON for a function call \" }}\n    {{- \"with its proper arguments that best answers the given prompt.\\n\\n\" }}\n    {{- 'Respond in the format {\"name\": function name, \"parameters\": dictionary of argument name and its value}.' }}\n    {{- \"Do not use variables.\\n\\n\" }}\n    {%- for t in tools %}\n        {{- t | tojson(indent=4) }}\n        {{- \"\\n\\n\" }}\n    {%- endfor %}\n    {{- first_user_message + \"<|eot|>\"}}\n{%- endif %}\n\n{%- for message in messages %}\n    {%- if not (message.role == 'ipython' or message.role == 'tool' or 'tool_calls' in message) %}\n    {{- '<|header_start|>' + message['role'] + '<|header_end|>\\n\\n' }}\n        {%- if message['content'] is string %}\n            {{- message['content'] }}\n        {%- else %}\n            {%- for content in message['content'] %}\n                {%- if content['type'] == 'image' %}\n                    {{- '<|image|>' }}\n                {%- elif content['type'] == 'text' %}\n                    {{- content['text'] }}\n                {%- endif %}\n            {%- endfor %}\n        {%- endif %}\n        {{- \"<|eot|>\" }}\n    {%- elif 'tool_calls' in message and message.tool_calls|length > 0 %}\n       {{- '<|header_start|>assistant<|header_end|>\\n\\n' -}}\n       {{- '<|python_start|>' }}\n        {%- if message['content'] is string %}\n            {{- message['content'] }}\n        {%- else %}\n            {%- for content in message['content'] %}\n                {%- if content['type'] == 'image' %}\n                    {{- '<|image|>' }}\n                {%- elif content['type'] == 'text' %}\n                    {{- content['text'] }}\n                {%- endif %}\n            {%- endfor %}\n        {%- endif %}\n       {{- '<|python_end|>' }}\n        {%- for tool_call in message.tool_calls %}\n           {{- '{\"name\": \"' + tool_call.function.name + '\", ' }}\n           {{- '\"parameters\": ' }}\n           {{- tool_call.function.arguments | tojson }}\n           {{- \"}\" }}\n        {%- endfor %}\n       {{- \"<|eot|>\" }}\n    {%- elif message.role == \"tool\" or message.role == \"ipython\" %}\n        {{- \"<|header_start|>ipython<|header_end|>\\n\\n\" }}\n        {%- if message.content is mapping or message.content is iterable %}\n            {{- message.content | tojson }}\n        {%- else %}\n            {{- message.content }}\n        {%- endif %}\n        {{- \"<|eot|>\" }}\n    {%- endif %}\n{%- endfor %}\n{%- if add_generation_prompt %}\n    {{- '<|header_start|>assistant<|header_end|>\\n\\n' }}\n{%- endif %}\n"
+}

config.json

@@ -0,0 +1,88 @@
+{
+  "architectures": [
+    "Llama4ForConditionalGeneration"
+  ],
+  "boi_token_index": 200080,
+  "eoi_token_index": 200081,
+  "image_token_index": 200092,
+  "model_type": "llama4",
+  "text_config": {
+    "_attn_implementation_autoset": true,
+    "attention_bias": false,
+    "attention_chunk_size": 8192,
+    "attention_dropout": 0.0,
+    "attn_scale": 0.1,
+    "attn_temperature_tuning": 4,
+    "bos_token_id": 200000,
+    "eos_token_id": [
+      200001,
+      200007,
+      200008
+    ],
+    "floor_scale": 8192,
+    "for_llm_compressor": false,
+    "head_dim": 32,
+    "hidden_act": "silu",
+    "hidden_size": 32,
+    "initializer_range": 0.02,
+    "interleave_moe_layer_step": 2,
+    "intermediate_size": 64,
+    "intermediate_size_mlp": 128,
+    "max_position_embeddings": 1048576,
+    "model_type": "llama4_text",
+    "moe_layers": [
+      1,
+      3
+    ],
+    "no_rope_layers": [
+      1,
+      1,
+      1,
+      0
+    ],
+    "num_attention_heads": 1,
+    "num_experts_per_tok": 1,
+    "num_hidden_layers": 4,
+    "num_key_value_heads": 1,
+    "num_local_experts": 8,
+    "output_router_logits": false,
+    "pad_token_id": 200018,
+    "rms_norm_eps": 1e-05,
+    "rope_scaling": null,
+    "rope_theta": 500000.0,
+    "router_aux_loss_coef": 0.001,
+    "router_jitter_noise": 0.0,
+    "tie_word_embeddings": true,
+    "torch_dtype": "bfloat16",
+    "use_cache": true,
+    "use_qk_norm": true,
+    "vocab_size": 202048
+  },
+  "tie_word_embeddings": false,
+  "torch_dtype": "bfloat16",
+  "transformers_version": "4.51.0",
+  "vision_config": {
+    "_attn_implementation_autoset": true,
+    "attention_dropout": 0.0,
+    "hidden_act": "gelu",
+    "hidden_size": 32,
+    "image_size": 336,
+    "initializer_range": 0.02,
+    "intermediate_size": 128,
+    "model_type": "llama4_vision_model",
+    "multi_modal_projector_bias": false,
+    "norm_eps": 1e-05,
+    "num_attention_heads": 1,
+    "num_channels": 3,
+    "num_hidden_layers": 2,
+    "patch_size": 14,
+    "pixel_shuffle_ratio": 0.5,
+    "projector_dropout": 0.0,
+    "projector_input_dim": 32,
+    "projector_output_dim": 32,
+    "rope_theta": 10000,
+    "vision_feature_layer": -1,
+    "vision_feature_select_strategy": "default",
+    "vision_output_dim": 32
+  }
+}

generation_config.json

@@ -0,0 +1,14 @@
+{
+  "bos_token_id": 200000,
+  "do_sample": true,
+  "eos_token_id": [
+    200001,
+    200007,
+    200008
+  ],
+  "pad_token_id": 200018,
+  "temperature": 0.6,
+  "top_p": 0.9,
+  "transformers_version": "4.51.0",
+  "trust_remote_code": true
+}

model.safetensors

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:dba211539db36f030df1135f8052cafcd5598cf4bfdfca30bcdab171df3249da
+size 13384256

preprocessor_config.json

@@ -0,0 +1,33 @@
+{
+  "crop_size": null,
+  "data_format": "channels_first",
+  "default_to_square": true,
+  "device": null,
+  "do_center_crop": null,
+  "do_convert_rgb": true,
+  "do_normalize": true,
+  "do_rescale": true,
+  "do_resize": true,
+  "image_mean": [
+    0.5,
+    0.5,
+    0.5
+  ],
+  "image_processor_type": "Llama4ImageProcessorFast",
+  "image_std": [
+    0.5,
+    0.5,
+    0.5
+  ],
+  "input_data_format": null,
+  "max_patches": 16,
+  "processor_class": "Llama4Processor",
+  "resample": 2,
+  "rescale_factor": 0.00392156862745098,
+  "resize_to_max_canvas": false,
+  "return_tensors": null,
+  "size": {
+    "height": 336,
+    "width": 336
+  }
+}

processor_config.json

@@ -0,0 +1,6 @@
+{
+  "fake_image_token": "<|image|>",
+  "image_token": "<|image|>",
+  "patch_size": 14,
+  "processor_class": "Llama4Processor"
+}

special_tokens_map.json

@@ -0,0 +1,23 @@
+{
+  "bos_token": {
+    "content": "<|begin_of_text|>",
+    "lstrip": false,
+    "normalized": false,
+    "rstrip": false,
+    "single_word": false
+  },
+  "eos_token": {
+    "content": "<|eot|>",
+    "lstrip": false,
+    "normalized": false,
+    "rstrip": false,
+    "single_word": false
+  },
+  "pad_token": {
+    "content": "<|finetune_right_pad_id|>",
+    "lstrip": false,
+    "normalized": false,
+    "rstrip": false,
+    "single_word": false
+  }
+}

tokenizer.json

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:172c9eb4beafc72601690da3ccfcede5c2e6806a8d5ec1fca33e22acea8023a4
+size 27948578