Upload 7 files

README.md

@@ -1,3 +1,184 @@
----
-license: llama2
----
+---
+license: llama2
+---
+
+## Installation from source
+
+```bash
+git clone https://github.com/foundation-model-stack/fms-extras
+cd fms-extras
+pip install -e .
+```
+
+
+## Description
+
+This model is intended to be used as an accelerator for [llama 13B (chat)](https://huggingface.co/meta-llama/Llama-2-13b-chat-hf) and takes inspiration 
+from the Medusa speculative decoding architecture. This accelerator modifies the MLP into a multi-stage MLP, where each stage predicts 
+a single token in the draft based on both a state vector and sampled token
+from the prior stage (the base model can be considered stage 0).
+The state vector from the base model provides contextual information to the accelerator, 
+while conditioning on prior sampled tokens allows it to produce higher-quality draft n-grams.
+
+Note: The underlying MLP speculator is a generic architecture that can be trained with any generative model to accelerate inference. 
+Training is light-weight and can be completed in only a few days depending on base model size and speed.
+
+## Repository Links
+
+1. [Paged Attention KV-Cache / Speculator](https://github.com/foundation-model-stack/fms-extras)
+2. [Production Server with speculative decoding](https://github.com/IBM/text-generation-inference.git)
+3. [Speculator training](https://github.com/foundation-model-stack/fms-fsdp/pull/35)
+
+## Samples
+
+_Note: For all samples, your environment must have access to cuda_
+
+### Use in IBM Production TGIS
+
+*To try this out running in a production-like environment, please use the pre-built docker image:*
+
+#### Setup
+
+```bash
+HF_HUB_CACHE=/hf_hub_cache
+chmod a+w $HF_HUB_CACHE
+HF_HUB_TOKEN="your huggingface hub token"
+TGIS_IMAGE=quay.io/wxpe/text-gen-server:main.ddc56ee
+
+docker pull $TGIS_IMAGE
+
+# optionally download llama-2-13b-chat if the weights do not already exist
+docker run --rm \
+    -v $HF_HUB_CACHE:/models \
+    -e HF_HUB_CACHE=/models \
+    -e TRANSFORMERS_CACHE=/models \
+    $TGIS_IMAGE \
+    text-generation-server download-weights \
+    meta-llama/Llama-2-13b-chat-hf \
+    --token $HF_HUB_TOKEN
+
+# optionally download the speculator model if the weights do not already exist
+docker run --rm \
+    -v $HF_HUB_CACHE:/models \
+    -e HF_HUB_CACHE=/models \
+    -e TRANSFORMERS_CACHE=/models \
+    $TGIS_IMAGE \
+    text-generation-server download-weights \
+    ibm-fms/llama-13b-accelerator \
+    --token $HF_HUB_TOKEN
+
+# note: if the weights were downloaded separately (not with the above commands), please place them in the HF_HUB_CACHE directory and refer to them with /models/<model_name>
+docker run -d --rm --gpus all \
+    --name my-tgis-server \
+    -p 8033:8033 \
+    -v $HF_HUB_CACHE:/models \
+    -e HF_HUB_CACHE=/models \
+    -e TRANSFORMERS_CACHE=/models \
+    -e MODEL_NAME=meta-llama/Llama-2-13b-chat-hf \
+    -e SPECULATOR_NAME=ibm-fms/llama-13b-accelerator \
+    -e FLASH_ATTENTION=true \
+    -e PAGED_ATTENTION=true \
+    -e DTYPE=float16 \
+    $TGIS_IMAGE
+
+# check logs and wait for "gRPC server started on port 8033" and "HTTP server started on port 3000"
+docker logs my-tgis-server -f
+
+# get the client sample (Note: The first prompt will take longer as there is a warmup time)
+conda create -n tgis-client-env python=3.11
+conda activate tgis-client-env
+git clone --branch main --single-branch https://github.com/IBM/text-generation-inference.git
+cd text-generation-inference/integration_tests
+make gen-client
+pip install . --no-cache-dir
+```
+
+#### Run Sample
+
+```bash
+python sample_client.py
+```
+
+_Note: first prompt may be slower as there is a slight warmup time_
+
+### Use in Huggingface TGI
+
+#### start the server
+
+```bash
+model=ibm-fms/llama-13b-accelerator
+volume=$PWD/data # share a volume with the Docker container to avoid downloading weights every run
+docker run --gpus all --shm-size 1g -p 8080:80 -v $volume:/data ghcr.io/huggingface/text-generation-inference:latest --model-id $model
+```
+
+_note: for tensor parallel, add --num-shard_
+
+#### make a request
+
+```bash
+curl 127.0.0.1:8080/generate_stream \
+    -X POST \
+    -d '{"inputs":"What is Deep Learning?","parameters":{"max_new_tokens":20}}' \
+    -H 'Content-Type: application/json'
+```
+
+### Minimal Sample
+
+*To try this out with the fms-native compiled model, please execute the following:*
+
+#### Install
+
+```bash
+git clone https://github.com/foundation-model-stack/fms-extras
+(cd fms-extras && pip install -e .)
+pip install transformers==4.35.0 sentencepiece numpy
+```
+
+#### Run Sample
+
+##### batch_size=1 (compile + cudagraphs)
+
+```bash
+MODEL_PATH=/path/to/llama/hf/13B-F
+python fms-extras/scripts/paged_speculative_inference.py \
+    --variant=13b \
+    --model_path=$MODEL_PATH \
+    --model_source=hf \
+    --tokenizer=$MODEL_PATH \
+    --speculator_path=ibm-fms/llama-13b-accelerator \
+    --speculator_source=hf \
+    --speculator_variant=840m \
+    --compile \
+    --compile_mode=reduce-overhead
+```
+
+##### batch_size=1 (compile)
+
+```bash
+MODEL_PATH=/path/to/llama/hf/13B-F
+python fms-extras/scripts/paged_speculative_inference.py \
+    --variant=13b \
+    --model_path=$MODEL_PATH \
+    --model_source=hf \
+    --tokenizer=$MODEL_PATH \
+    --speculator_path=ibm-fms/llama-13b-accelerator \
+    --speculator_source=hf \
+    --speculator_variant=840m \
+    --compile
+```
+
+##### batch_size=4 (compile)
+
+```bash
+MODEL_PATH=/path/to/llama/hf/13B-F
+python fms-extras/scripts/paged_speculative_inference.py \
+    --variant=13b \
+    --model_path=$MODEL_PATH \
+    --model_source=hf \
+    --tokenizer=$MODEL_PATH \
+    --speculator_path=ibm-fms/llama-13b-accelerator \
+    --speculator_source=hf \
+    --speculator_variant=840m \
+    --batch_input \
+    --compile
+```

config.json

@@ -0,0 +1,19 @@
+{
+  "base_model_name_or_path": "meta-llama/Llama-2-13b-chat-hf",
+  "architectures": [
+    "MLPSpeculatorPreTrainedModel"
+  ],
+  "emb_dim": 5120,
+  "inner_dim": 4096,
+  "model_type": "mlp_speculator",
+  "n_candidates": 5,
+  "n_predict": 3,
+  "top_k_tokens_per_head": [
+    5,
+    3,
+    2
+  ],
+  "torch_dtype": "float16",
+  "transformers_version": "4.35.0",
+  "vocab_size": 32000
+}

model.safetensors

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:de321baca8edf17872498e6ce2721870745f41baf33473a49717a52fa4a79f28
+size 1681966568

special_tokens_map.json

@@ -0,0 +1,23 @@
+{
+  "bos_token": {
+    "content": "<s>",
+    "lstrip": false,
+    "normalized": true,
+    "rstrip": false,
+    "single_word": false
+  },
+  "eos_token": {
+    "content": "</s>",
+    "lstrip": false,
+    "normalized": true,
+    "rstrip": false,
+    "single_word": false
+  },
+  "unk_token": {
+    "content": "<unk>",
+    "lstrip": false,
+    "normalized": true,
+    "rstrip": false,
+    "single_word": false
+  }
+}

tokenizer.model

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:9e556afd44213b6bd1be2b850ebbbd98f5481437a8021afaf58ee7fb1818d347
+size 499723

tokenizer_config.json

@@ -0,0 +1,37 @@
+{
+  "added_tokens_decoder": {
+    "0": {
+      "content": "<unk>",
+      "lstrip": false,
+      "normalized": true,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "1": {
+      "content": "<s>",
+      "lstrip": false,
+      "normalized": true,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "2": {
+      "content": "</s>",
+      "lstrip": false,
+      "normalized": true,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    }
+  },
+  "bos_token": "<s>",
+  "clean_up_tokenization_spaces": false,
+  "eos_token": "</s>",
+  "model_max_length": 1000000000000000019884624838656,
+  "pad_token": null,
+  "sp_model_kwargs": {},
+  "tokenizer_class": "LlamaTokenizer",
+  "unk_token": "<unk>",
+  "use_default_system_prompt": false
+}