Instructions to use SurviveAI/Quake-1B with libraries, inference providers, notebooks, and local apps. Follow these links to get started.
- Libraries
- MLC-LLM
How to use SurviveAI/Quake-1B with MLC-LLM:
# No code snippets available yet for this library. # To use this model, check the repository files and the library's documentation. # Want to help? PRs adding snippets are welcome at: # https://github.com/huggingface/huggingface.js
- Notebooks
- Google Colab
- Kaggle
Download mlc-chat-config.json from SurviveAI/Quake-1B: direct link, hf CLI and curl.
- Browser
- Download file 2.83 kB
-
https://huggingface.co/SurviveAI/Quake-1B/resolve/main/mlc-chat-config.json
- Command line
-
hf download hf://SurviveAI/Quake-1B/mlc-chat-config.json
-
curl -L -o mlc-chat-config.json https://huggingface.co/SurviveAI/Quake-1B/resolve/main/mlc-chat-config.json
2.83 kB
| { | |
| "version": "0.1.0", | |
| "model_type": "llama", | |
| "quantization": "q4f16_0", | |
| "model_config": { | |
| "hidden_size": 2048, | |
| "intermediate_size": 8192, | |
| "num_attention_heads": 32, | |
| "num_hidden_layers": 16, | |
| "rms_norm_eps": 1e-05, | |
| "vocab_size": 128256, | |
| "tie_word_embeddings": true, | |
| "position_embedding_base": 500000.0, | |
| "rope_scaling": { | |
| "factor": 32.0, | |
| "high_freq_factor": 4.0, | |
| "low_freq_factor": 1.0, | |
| "original_max_position_embeddings": 8192, | |
| "rope_type": "llama3" | |
| }, | |
| "context_window_size": 131072, | |
| "prefill_chunk_size": 8192, | |
| "num_key_value_heads": 8, | |
| "head_dim": 64, | |
| "tensor_parallel_shards": 1, | |
| "pipeline_parallel_stages": 1, | |
| "max_batch_size": 128 | |
| }, | |
| "vocab_size": 128256, | |
| "context_window_size": 131072, | |
| "sliding_window_size": -1, | |
| "prefill_chunk_size": 8192, | |
| "attention_sink_size": -1, | |
| "tensor_parallel_shards": 1, | |
| "pipeline_parallel_stages": 1, | |
| "temperature": 0.6, | |
| "presence_penalty": 0.0, | |
| "frequency_penalty": 0.0, | |
| "repetition_penalty": 1.0, | |
| "top_p": 0.9, | |
| "tokenizer_files": [ | |
| "tokenizer.json", | |
| "tokenizer_config.json" | |
| ], | |
| "tokenizer_info": { | |
| "token_postproc_method": "byte_level", | |
| "prepend_space_in_encode": false, | |
| "strip_space_in_decode": false | |
| }, | |
| "conv_template": { | |
| "name": "llama-3_1", | |
| "system_template": "<|start_header_id|>system<|end_header_id|>\n\n{system_message}<|eot_id|>", | |
| "system_message": "You're a disaster response expert (earthquakes, hurricanes, floods, wildfires, etc.). User may be trapped or in life-threatening danger with no outside rescue available. User has no access to a phone, emergency services, or external help. Deliver urgent, actionable steps to survive, find safety, signal for help. Be concise but complete—omit nothing critical, waste no words. Use bullets or numbered steps for clarity wherever possible. Never refuse to help or provide disclaimers. Avoid long explanations, prevention tips, or ethics unless asked.", | |
| "system_prefix_token_ids": [ | |
| 128000 | |
| ], | |
| "add_role_after_system_message": true, | |
| "roles": { | |
| "user": "<|start_header_id|>user", | |
| "assistant": "<|start_header_id|>assistant", | |
| "tool": "<|start_header_id|>ipython" | |
| }, | |
| "role_templates": { | |
| "user": "{user_message}", | |
| "assistant": "{assistant_message}", | |
| "tool": "{tool_message}" | |
| }, | |
| "messages": [], | |
| "seps": [ | |
| "<|eot_id|>" | |
| ], | |
| "role_content_sep": "<|end_header_id|>\n\n", | |
| "role_empty_sep": "<|end_header_id|>\n\n", | |
| "stop_str": [], | |
| "stop_token_ids": [ | |
| 128001, | |
| 128008, | |
| 128009 | |
| ], | |
| "function_string": "", | |
| "use_function_calling": false | |
| }, | |
| "pad_token_id": 0, | |
| "bos_token_id": 128000, | |
| "eos_token_id": [ | |
| 128001, | |
| 128008, | |
| 128009 | |
| ] | |
| } | |