Upload folder using huggingface_hub

.gitattributes

@@ -33,3 +33,5 @@ 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
+llm.mnn filter=lfs diff=lfs merge=lfs -text
+llm.mnn.weight filter=lfs diff=lfs merge=lfs -text

README.md

@@ -1,3 +1,20 @@
----
-license: mit
----
+---
+license: apache-2.0
+language:
+- en
+base_model: shisa-ai/shisa-v2.1-unphi4-14b
+base_model_relation: quantized
+pipeline_tag: text-generation
+library_name: mnn
+tags:
+- code
+- mnn
+---
+
+Warning: This model was not verified with the MNN Chat app; it wouldn't load on my 12 GB RAM phone.
+
+This model [DeProgrammer/shisa-v2.1-unphi4-14b](https://huggingface.co/DeProgrammer/shisa-v2.1-unphi4-14b-MNN) was
+converted to MNN format from [shisa-ai/shisa-v2.1-unphi4-14b](https://huggingface.co/shisa-ai/shisa-v2.1-unphi4-14b)
+using [llmexport.py](https://github.com/alibaba/MNN/issues/4153#issuecomment-3866182869) in [MNN version **3.4.0**](https://github.com/alibaba/MNN/commit/a874b302f094599e2838a9186e5ce2cf6a81a7a7) with default settings (4-bit quantization).
+
+Inference can be run via MNN, e.g., MNN Chat on Android.

config.json

@@ -0,0 +1,10 @@
+{
+    "llm_model": "llm.mnn",
+    "llm_weight": "llm.mnn.weight",
+    "backend_type": "cpu",
+    "thread_num": 4,
+    "precision": "low",
+    "memory": "low",
+    "sampler_type": "penalty",
+    "penalty": 1.1
+}

embeddings_bf16.bin

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:98d761d26e7667c8bd09ba71a7c496c54c5309778de0e77b112ab28a0263603e
+size 1027604480

export_args.json

@@ -0,0 +1,42 @@
+{
+    "path": "shisa-ai/shisa-v2.1-unphi4-14b",
+    "type": null,
+    "tokenizer_path": "shisa-ai/shisa-v2.1-unphi4-14b",
+    "eagle_path": null,
+    "lora_path": null,
+    "gptq_path": null,
+    "dst_path": "./model",
+    "verbose": false,
+    "test": null,
+    "export": "mnn",
+    "onnx_slim": false,
+    "quant_bit": 4,
+    "quant_block": 64,
+    "visual_quant_bit": null,
+    "visual_quant_block": null,
+    "lm_quant_bit": 4,
+    "lm_quant_block": 64,
+    "mnnconvert": "../../../build/MNNConvert",
+    "ppl": false,
+    "awq": false,
+    "hqq": false,
+    "omni": false,
+    "transformer_fuse": false,
+    "group_conv_native": false,
+    "smooth": false,
+    "sym": false,
+    "visual_sym": false,
+    "seperate_embed": false,
+    "lora_split": false,
+    "calib_data": null,
+    "act_bit": 16,
+    "embed_bit": 16,
+    "act_sym": false,
+    "quant_config": null,
+    "generate_for_npu": false,
+    "skip_weight": false,
+    "omni_epochs": 20,
+    "omni_lr": 0.005,
+    "omni_wd": 0.0001,
+    "tie_word_embeddings": false
+}

llm.mnn

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:75c7945829b747825e3fc92f33dae0f588bfb75b7b66754eca5533a9ddf8338f
+size 696896

llm.mnn.weight

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:6d4537c84c8771d2788ac84cf5fc005dcda45d519ebf7491f3783ca3034ea2f6
+size 8844130346

llm_config.json

@@ -0,0 +1,11 @@
+{
+    "model_type": "llama",
+    "hidden_size": 5120,
+    "attention_mask": "float",
+    "attention_type": "full",
+    "jinja": {
+        "chat_template": "{% for message in messages %}{% if (message['role'] == 'system') %}{{'<|im_start|>system<|im_sep|>' + message['content'] + '<|im_end|>'}}{% elif (message['role'] == 'user') %}{{'<|im_start|>user<|im_sep|>' + message['content'] + '<|im_end|>'}}{% elif (message['role'] == 'assistant') %}{{'<|im_start|>assistant<|im_sep|>' + message['content'] + '<|im_end|>'}}{% endif %}{% endfor %}{% if add_generation_prompt %}{{ '<|im_start|>assistant<|im_sep|>' }}{% endif %}",
+        "bos": "<|endoftext|>",
+        "eos": "<|im_end|>"
+    }
+}