Skip to content

fix doc #1

New issue

Have a question about this project? Sign up for a free GitHub account to open an issue and contact its maintainers and the community.

Sign up for GitHub

By clicking “Sign up for GitHub”, you agree to our terms of service and privacy statement. We’ll occasionally send you account related emails.

Already on GitHub? Sign in to your account

Merged
Zhou-sx merged 2 commits into from
Mar 1, 2025
Merged

fix doc #1

Changes from all commits
Commits
Show all changes
2 commits
Select commit Hold shift + click to select a range
5fc7af5
fix
Achazwl Mar 1, 2025
6c738ea
fix model path
Achazwl Mar 1, 2025
File filter

Filter by extension

Filter by extension
Conversations
Failed to load comments.
Loading
Jump to
Jump to file
Failed to load files.
Loading
Diff view
Diff view
19 changes: 6 additions & 13 deletions docs/backend/speculative_decoding.ipynb
View file
Open in desktop
Original file line number Diff line number Diff line change
Expand Up @@ -150,18 +150,11 @@
"source": [
"### EAGLE Decoding via Frequency-Ranked Speculative Sampling\n",
"\n",
"By employing a truncated high-frequency token vocabulary in the draft model, Eagle speculative decoding reduces lm_head computational overhead while accelerating the pipeline without quality degradation.For more details checkout [this paper](https://arxiv.org/pdf/arXiv:2502.14856)\n",
"By employing a truncated high-frequency token vocabulary in the draft model, Eagle speculative decoding reduces lm_head computational overhead while accelerating the pipeline without quality degradation. For more details checkout [this paper](https://arxiv.org/pdf/arXiv:2502.14856)\n",
"\n",
"Set `--speculative-token-map` to use this optimization. You can get the high-frequency token in FR-Spec from https://huggingface.co/thunlp/LLaMA3-Instruct-8B-FR-Spec. Or you can obtain high-frequency token by yourself.\n",
"+ Execute inference on your dataset using sglang's standard inference mode and persist the outputs.\n",
"+ Extract the top-k high-frequency tokens from the saved file. There is a reference implementation (https://gist.github.com/Zhou-sx/71a9196d2f324c93f79016579fdf57da). \n"
"Set `--speculative-token-map` to use this optimization. You can get the high-frequency token in FR-Spec from https://huggingface.co/thunlp/LLaMA3-Instruct-8B-FR-Spec. Or you can obtain high-frequency token by yourself (https://github.com/thunlp/FR-Spec/tree/main?tab=readme-ov-file#prepare-fr-spec-vocabulary-subset).\n"
]
},
{
"cell_type": "markdown",
"metadata": {},
"source": []
},
{
"cell_type": "code",
"execution_count": null,
Expand All @@ -179,9 +172,9 @@
"\n",
"server_process, port = launch_server_cmd(\n",
" \"\"\"\n",
"python3 -m sglang.launch_server --model meta-llama/Llama-3.1-8B-Instruct --speculative-algo EAGLE \\\n",
" --speculative-draft lmzheng/sglang-EAGLE-LLaMA3-Instruct-8B --speculative-num-steps 3 \\\n",
" --speculative-eagle-topk 4 --speculative-num-draft-tokens 16 --speculative-token-map {hot_token_ids.pt} \n",
"python3 -m sglang.launch_server --model meta-llama/Meta-Llama-3-8B-Instruct --speculative-algo EAGLE \\\n",
" --speculative-draft lmzheng/sglang-EAGLE-LLaMA3-Instruct-8B --speculative-num-steps 5 \\\n",
" --speculative-eagle-topk 8 --speculative-num-draft-tokens 64 --speculative-token-map {hot_token_ids.pt} \n",
"\"\"\"\n",
")\n",
"\n",
Expand All @@ -199,7 +192,7 @@
"client = openai.Client(base_url=f\"http://127.0.0.1:{port}/v1\", api_key=\"None\")\n",
"\n",
"response = client.chat.completions.create(\n",
" model=\"meta-llama/Meta-Llama-3.1-8B-Instruct\",\n",
" model=\"meta-llama/Meta-Llama-3-8B-Instruct\",\n",
" messages=[\n",
" {\"role\": \"user\", \"content\": \"List 3 countries and their capitals.\"},\n",
" ],\n",
Expand Down