Duplicate from FastFlowLM/Llama-3.1-8B-NPU2

Co-authored-by: FastFlowLM <FastFlowLM@users.noreply.huggingface.co>

.gitattributes

@@ -0,0 +1,7 @@
+model.q4nx filter=lfs diff=lfs merge=lfs -text
+tokenizer.json filter=lfs diff=lfs merge=lfs -text
+layer.xclbin filter=lfs diff=lfs merge=lfs -text
+lm_head.xclbin filter=lfs diff=lfs merge=lfs -text
+attn.xclbin filter=lfs diff=lfs merge=lfs -text
+dequant.xclbin filter=lfs diff=lfs merge=lfs -text
+mm.xclbin filter=lfs diff=lfs merge=lfs -text

README.md

@@ -0,0 +1,67 @@
+---
+license: llama3
+language:
+- en
+tags:
+- llama
+- llama-3.1
+- text-generation
+- AMD
+- Ryzen
+- NPU
+pipeline_tag: text-generation
+base_model:
+- meta-llama/Llama-3.1-8B-Instruct
+---
+
+# 🦙 LLaMA 3.1 (8B) – Optimized for FastFlowLM on AMD Ryzen™ AI NPU (XDNA2 Only)
+
+## Model Summary
+This is a derivative of Meta AI’s LLaMA 3.1 base model. The model retains the core architecture and weights from Meta’s release and may include fine-tuning, quantization, or adaptation for specific applications.
+
+> ⚠️ **This model is subject to Meta’s LLaMA 3 license. You must accept Meta’s terms to use or download it.**
+
+## 📝 License & Usage Terms
+
+### Meta LLaMA 3 License
+- Base model is governed by Meta AI's license:  
+  👉 https://ai.meta.com/llama/license/
+
+- You must agree to their license terms to access and use the weights, which include:
+  - No commercial use without permission
+  - Redistribution only allowed under specific conditions
+  - Attribution required
+
+### Redistribution Notice
+- This repository does **not** include original Meta weights.
+- You must obtain base weights directly from Meta:  
+  👉 https://huggingface.co/meta-llama
+
+### If Fine-tuned
+If this model has been fine-tuned, the downstream weights are provided under the following conditions:
+
+- **Base Model License**: Meta’s LLaMA 3 License
+- **Derivative Weights License**: [e.g., CC-BY-NC-4.0, MIT, custom, etc.]
+- **Training Dataset License(s)**:  
+  - [Dataset A] – [license]  
+  - [Dataset B] – [license]
+
+Make sure you have rights to use and distribute any data used in fine-tuning.
+
+## Intended Use
+- **Use Cases**: Research, experimentation, academic NLP, code generation (if applicable)
+- **Not Intended For**: Use in production systems without further evaluation, sensitive applications, or commercial deployments without Meta’s explicit permission
+
+## Limitations & Risks
+- May generate incorrect or harmful content
+- Does not have knowledge past its training cutoff
+- Biases in training data may persist
+
+## Citation
+```bibtex
+@misc{touvron2024llama3,
+  title={LLaMA 3: Open Foundation and Instruction Models},
+  author={Touvron, Hugo and others},
+  year={2024},
+  url={https://ai.meta.com/llama/}
+}

attn.xclbin

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:cb8e7bef68c19797d9f57c163ba8943fa3bd474af6c9e82c0c4a7554852afd4a
+size 698331

config.json

@@ -0,0 +1,45 @@
+{
+  "architectures": [
+    "LlamaForCausalLM"
+  ],
+  "attention_bias": false,
+  "attention_dropout": 0.0,
+  "bos_token_id": 128000,
+  "eos_token_id": [
+    128001,
+    128008,
+    128009
+  ],
+  "head_dim": 128,
+  "hidden_act": "silu",
+  "hidden_size": 4096,
+  "initializer_range": 0.02,
+  "intermediate_size": 14336,
+  "max_position_embeddings": 131072,
+  "mlp_bias": false,
+  "model_type": "llama",
+  "num_attention_heads": 32,
+  "num_hidden_layers": 32,
+  "num_key_value_heads": 8,
+  "pretraining_tp": 1,
+  "rms_norm_eps": 1e-05,
+  "rope_scaling": {
+    "factor": 8.0,
+    "high_freq_factor": 4.0,
+    "low_freq_factor": 1.0,
+    "original_max_position_embeddings": 8192,
+    "rope_type": "llama3"
+  },
+  "rope_theta": 500000.0,
+  "tie_word_embeddings": false,
+  "torch_dtype": "float32",
+  "transformers_version": "4.52.0.dev0",
+  "use_cache": true,
+  "vocab_size": 128256,
+  "addr_qk": 9216,
+  "addr_kv": 34816,
+  "addr_l_begin_mha": 54272,
+  "addr_l_end_mha": 25600,
+  "addr_kk": 45056,
+  "flm_version": "0.9.21"
+}

dequant.xclbin

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:171e48055c262c07e7232be872e1a7a4ae92a1158ab8a1d455bb4d61c2a7fd3c
+size 115691

layer.xclbin

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:da989ed295cbd49c07b185ba99ab867da07eff6cc44bc8f9f452b0b0ff0389bd
+size 331659

lm_head.xclbin

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:6d2ebc16daf6a75e4bdefa499fccccc2bafd7c122bf4a1c3e73d66e8538f592d
+size 152843

mm.xclbin

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:b1e88cd899e3fbef3b402ecaf2ab644f21cca2ee85dafe6e2521d89222c5a644
+size 347675

model.q4nx

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:0c03021f1cd21e43376234817b34ad32ccf07dca94785ede7640833d15169d8a
+size 5741650592

tokenizer.json

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:79e3e522635f3171300913bb421464a87de6222182a0570b9b2ccba2a964b2b4
+size 9085657

tokenizer_config.json

@@ -0,0 +1,12 @@
+{ 
+  "bos_token": "<|begin_of_text|>",
+  "eos_token": "<|eot_id|>",
+  "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    {%- set system_message = messages[0]['content']|trim %}\n    {%- set messages = messages[1:] %}\n{%- else %}\n    {%- if user_system_prompt is defined %}{%- set system_message = user_system_prompt %} {%- else %} {%-  set system_message = \"\" %}\n{%- endif %}\n{%- endif %}\n\n{#- System message #}\n{{- \"<|start_header_id|>system<|end_header_id|>\\n\\n\" }}\n{%- if tools is not none %}\n    {{- \"Environment: ipython\\n\" }}\n{%- endif %}\n{{- \"Cutting Knowledge Date: December 2023\\n\" }}\n{{- \"Today Date: \" + date_string + \"\\n\\n\" }}\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_id|>\" }}\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    {{- '<|start_header_id|>user<|end_header_id|>\\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_id|>\"}}\n{%- endif %}\n\n{%- for message in messages %}\n    {%- if not (message.role == 'ipython' or message.role == 'tool' or 'tool_calls' in message) %}\n        {{- '<|start_header_id|>' + message['role'] + '<|end_header_id|>\\n\\n'+ message['content'] | trim + '<|eot_id|>' }}\n    {%- elif 'tool_calls' in message %}\n        {%- if not message.tool_calls|length == 1 %}\n            {{- raise_exception(\"This model only supports single tool-calls at once!\") }}\n        {%- endif %}\n        {%- set tool_call = message.tool_calls[0].function %}\n        {{- '<|start_header_id|>assistant<|end_header_id|>\\n\\n' -}}\n        {{- '{\"name\": \"' + tool_call.name + '\", ' }}\n        {{- '\"parameters\": ' }}\n        {{- tool_call.arguments | tojson }}\n        {{- \"}\" }}\n        {{- \"<|eot_id|>\" }}\n    {%- elif message.role == \"tool\" or message.role == \"ipython\" %}\n        {{- \"<|start_header_id|>ipython<|end_header_id|>\\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_id|>\" }}\n    {%- endif %}\n{%- endfor %}\n{%- if add_generation_prompt %}\n    {{- '<|start_header_id|>assistant<|end_header_id|>\\n\\n' }}\n{%- endif %}\n",
+  "model_max_length": 131072,
+  "bos_token_id": 128000,
+  "eos_token_id": [
+    128001,
+    128008,
+    128009
+  ]
+}