commit files to HF hub

README.md

@@ -0,0 +1,228 @@
+# lmarena-ai/p2l-360m-grk-01112025
+
+Large language model (LLM) evaluations typically rely on aggregated metrics like accuracy or human preference, averaging across users and prompts. This averaging obscures user- and prompt-specific variations in model performance.
+To address this, we propose Prompt-to-Leaderboard (P2L), a method that produces leaderboards specific to a prompt.
+The core idea is to train an LLM taking natural language prompts as input to output a vector of coefficients which are then used to predict the human preference vote.
+The resulting prompt-dependent leaderboards allow for unsupervised task-specific evaluation, optimal routing of queries to models, personalization, and automated evaluation of model strengths and weaknesses. 
+Data from Chatbot Arena suggest that P2L better captures the nuanced landscape of language model performance than the averaged leaderboard.
+
+**Paper**: [Prompt-to-Leaderboard](https://arxiv.org/abs/2502.14855)
+
+**Code**: [lmarena/p2l](https://github.com/lmarena/p2l)
+
+This particular P2L model has a *Grounded Rao-Kupper* regression head, which we define below:
+
+Let
+$$
+Y\in \{\mathsf{A}, \mathsf{B}, \mathsf{tie}, \mathsf{bad}\}
+$$
+and for the sake of notational convenience, let 
+$$
+\theta^*(z) = \big(\beta^*(z), \eta^*(z)\big); \ \beta^*(z) \in \mathbb{R}^M, \eta^*(z) \in \mathbb{R}_{\geq 1}\}
+$$
+
+For notational convenience, we define: 
+$$
+\varphi^*(z)_i := \exp(\beta^*(z)_i)
+$$
+
+Then grounded Rao-Kupper model is defined as:
+$$
+    g_{\theta^*(z)}(y ; x) =
+    \begin{cases}
+        \frac{\varphi^*(z)_A}{\varphi^*(z)_A + \eta^*(z)\varphi^*(z)_B + 1} &  y = \mathsf{A} \\
+        \frac{\varphi^*(z)_B}{\varphi^*(z)_B + \eta^*(z)\varphi^*(z)_A + 1} &  y = \mathsf{B}\\
+        \frac{1}{1 + \varphi^*(z)_A + \varphi^*(z)_B} & y = \mathsf{bad}\\
+        1 - \frac{\varphi^*(z)_A}{\varphi^*(z)_A + \eta^*(z)\varphi^*(z)_B + 1}  - \frac{\varphi^*(z)_B}{\varphi^*(z)_B + \eta^*(z)\varphi^*(z)_A + 1} - \frac{1}{1 + \varphi^*(z)_A + \varphi^*(z)_B} & y = \mathsf{tie}.
+    \end{cases}
+$$
+
+See section 2.2 in our paper for more details on various regression heads.
+
+## Serving
+To serve a P2L model, please see our documentation on GitHub: [Serving P2L](https://github.com/lmarena/p2l?tab=readme-ov-file#serving-p2l).
+
+Note: the P2L model outputs with this structure:
+
+
+```python
+class P2LOutputs(ModelOutput):
+    coefs: torch.FloatTensor = None # "betas" as described above
+    eta: Optional[torch.FloatTensor] = None # tie coefficent (also eta above)
+    last_hidden_state: torch.FloatTensor = None # last hidden state from the transformer
+```
+
+To understand which coefficient index corresponds with which model, see the [`model_list.json`](./model_list.json) found in the repo of each P2L model. As a general rule, the models will always be in sorted order.
+
+The easiest way to get this list from inside code is with the following:
+
+```python
+import json
+from huggingface_hub import hf_hub_download
+
+fname = hf_hub_download(
+        repo_id="lmarena-ai/p2l-360m-grk-01112025", filename="model_list.json", repo_type="model"
+    )
+
+with open(fname) as fin:
+    model_list = json.load(fin)
+```
+
+
+
+### Loading from Pretrained
+
+To define and load the model:
+
+```python
+
+import torch
+from transformers import (
+    Qwen2Model,
+    Qwen2PreTrainedModel,
+    LlamaModel,
+    LlamaPreTrainedModel,
+    PreTrainedModel,
+    AutoTokenizer,
+)
+from transformers import AutoTokenizer
+from transformers.utils import ModelOutput
+from dataclasses import dataclass
+import torch.nn as nn
+import torch.nn.functional as F
+from typing import Dict, Tuple, Callable, Optional
+from huggingface_hub import hf_hub_download
+import json
+
+
+@dataclass
+class HeadOutputs(ModelOutput):
+    coefs: torch.FloatTensor = None
+    eta: Optional[torch.FloatTensor] = None
+    gamma: Optional[torch.FloatTensor] = None
+
+
+@dataclass
+class P2LOutputs(ModelOutput):
+    coefs: torch.FloatTensor = None
+    eta: Optional[torch.FloatTensor] = None
+    gamma: Optional[torch.FloatTensor] = None
+    loss: Optional[torch.FloatTensor] = None
+    last_hidden_state: torch.FloatTensor = None
+
+class RKHead(nn.Module):
+    def __init__(
+        self,
+        input_dim,
+        output_dim,
+        **kwargs,
+    ) -> None:
+        super().__init__()
+        self.head = nn.Linear(
+            in_features=input_dim, out_features=output_dim, bias=True
+        )
+        self.eta_head = nn.Linear(
+            in_features=input_dim, out_features=1, bias=True
+        )
+
+    def forward(self, last_hidden_dim: torch.Tensor):
+        coefs = self.head(last_hidden_dim)
+        eta = self.eta_head(last_hidden_dim)
+
+        return HeadOutputs(coefs=coefs, eta=eta)
+
+class P2LModel(LlamaPreTrainedModel):
+    def __init__(
+        self,
+        config,
+        CLS_id,
+        num_models,
+        head_kwargs={},
+        **kwargs,
+    ):
+        super().__init__(config)
+
+        self.num_models = num_models
+        self.cls_token_id = CLS_id
+
+        self.model = LlamaModel(config)
+
+        self.head = RKHead(
+            input_dim=config.hidden_size,
+            output_dim=self.num_models,
+            **head_kwargs,
+        )
+
+        self.post_init()
+
+    def freeze_transformer(self):
+        for param in self.model.parameters():
+            param.requires_grad = False
+
+    def get_input_embeddings(self):
+        return self.model.embed_tokens
+
+    def set_input_embeddings(self, value):
+        self.model.embed_tokens = value
+
+    def forward(self, input_ids, attention_mask, labels=None, weights=None):
+        batch_size = input_ids.shape[0]
+
+        hidden_outputs = self.model(
+            input_ids=input_ids,
+            attention_mask=attention_mask,
+            output_hidden_states=False,
+        ).last_hidden_state  # (bs, num_token, embed_dim)
+
+        cls_mask = input_ids == self.cls_token_id
+
+        # double check this is getting the current CLS token
+        cls_hidden_dim = hidden_outputs[cls_mask]
+
+        assert (
+            cls_hidden_dim.shape[0] == batch_size
+        ), f"input ids {input_ids.shape}, cls_mask {cls_mask.shape}, cls_logit {cls_hidden_dim.shape}"
+
+        head_output = self.head(cls_hidden_dim)
+
+    
+        outputs = P2LOutputs(
+            coefs=head_output.coefs,
+            last_hidden_state=cls_hidden_dim,
+            eta=head_output.eta,
+            gamma=head_output.gamma,
+        )
+
+        return outputs
+
+
+fname = hf_hub_download(
+        repo_id="lmarena-ai/p2l-360m-grk-01112025", filename="model_list.json", repo_type="model"
+    )
+
+with open(fname) as fin:
+    model_list = json.load(fin)
+
+tokenizer = AutoTokenizer.from_pretrained("lmarena-ai/p2l-360m-grk-01112025")
+model = P2LModel.from_pretrained(
+    "lmarena-ai/p2l-360m-grk-01112025",
+    CLS_id=tokenizer.cls_token_id,
+    num_models=len(model_list),
+    torch_dtype=torch.bfloat16,
+)
+
+```
+
+## Citation
+
+```
+@misc{frick2025prompttoleaderboard,
+      title={Prompt-to-Leaderboard}, 
+      author={Evan Frick and Connor Chen and Joseph Tennyson and Tianle Li and Wei-Lin Chiang and Anastasios N. Angelopoulos and Ion Stoica},
+      year={2025},
+      eprint={2502.14855},
+      archivePrefix={arXiv},
+      primaryClass={cs.LG},
+      url={https://arxiv.org/abs/2502.14855}, 
+}
+```

added_tokens.json

@@ -0,0 +1,3 @@
+{
+  "<|cls|>": 49152
+}

config.json

@@ -0,0 +1,39 @@
+{
+  "_name_or_path": "HuggingFaceTB/SmolLM2-360M-Instruct",
+  "architectures": [
+    "P2LModel"
+  ],
+  "attention_bias": false,
+  "attention_dropout": 0.0,
+  "bos_token_id": 1,
+  "eos_token_id": 2,
+  "head_dim": 64,
+  "hidden_act": "silu",
+  "hidden_size": 960,
+  "initializer_range": 0.02,
+  "intermediate_size": 2560,
+  "is_llama_config": true,
+  "max_position_embeddings": 8192,
+  "mlp_bias": false,
+  "model_type": "llama",
+  "num_attention_heads": 15,
+  "num_hidden_layers": 32,
+  "num_key_value_heads": 5,
+  "pad_token_id": 2,
+  "pretraining_tp": 1,
+  "rms_norm_eps": 1e-05,
+  "rope_interleaved": false,
+  "rope_scaling": null,
+  "rope_theta": 100000,
+  "tie_word_embeddings": true,
+  "torch_dtype": "bfloat16",
+  "transformers.js_config": {
+    "kv_cache_dtype": {
+      "fp16": "float16",
+      "q4f16": "float16"
+    }
+  },
+  "transformers_version": "4.47.1",
+  "use_cache": true,
+  "vocab_size": 49153
+}

model.safetensors

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:6472f1007af72c4a61ba2063f2a9517bf2ecb1f9a12e9b7f0dece3c286e7405f
+size 723928926

model_list.json

@@ -0,0 +1,132 @@
+[
+ "amazon-nova-lite-v1.0",
+ "amazon-nova-micro-v1.0",
+ "amazon-nova-pro-v1.0",
+ "athene-70b-0725",
+ "athene-v2-chat",
+ "c4ai-aya-expanse-32b",
+ "c4ai-aya-expanse-8b",
+ "chatgpt-4o-latest-20240808",
+ "chatgpt-4o-latest-20240903",
+ "chatgpt-4o-latest-20241120",
+ "claude-3-5-haiku-20241022",
+ "claude-3-5-sonnet-20240620",
+ "claude-3-5-sonnet-20241022",
+ "claude-3-haiku-20240307",
+ "claude-3-opus-20240229",
+ "claude-3-sonnet-20240229",
+ "codestral-2405",
+ "command-r",
+ "command-r-08-2024",
+ "command-r-plus",
+ "command-r-plus-08-2024",
+ "dbrx-instruct-preview",
+ "deepseek-coder-v2",
+ "deepseek-coder-v2-0724",
+ "deepseek-v2-api-0628",
+ "deepseek-v2.5",
+ "deepseek-v2.5-1210",
+ "deepseek-v3",
+ "gemini-1.5-flash-001",
+ "gemini-1.5-flash-002",
+ "gemini-1.5-flash-8b-001",
+ "gemini-1.5-flash-8b-exp-0827",
+ "gemini-1.5-flash-exp-0827",
+ "gemini-1.5-pro-001",
+ "gemini-1.5-pro-002",
+ "gemini-1.5-pro-api-0409-preview",
+ "gemini-1.5-pro-exp-0801",
+ "gemini-1.5-pro-exp-0827",
+ "gemini-2.0-flash-exp",
+ "gemini-2.0-flash-thinking-exp-1219",
+ "gemini-advanced-0514",
+ "gemini-exp-1114",
+ "gemini-exp-1121",
+ "gemini-exp-1206",
+ "gemma-1.1-2b-it",
+ "gemma-1.1-7b-it",
+ "gemma-2-27b-it",
+ "gemma-2-2b-it",
+ "gemma-2-9b-it",
+ "gemma-2-9b-it-simpo",
+ "glm-4-0116",
+ "glm-4-0520",
+ "glm-4-plus",
+ "gpt-3.5-turbo-0125",
+ "gpt-4-0125-preview",
+ "gpt-4-0314",
+ "gpt-4-0613",
+ "gpt-4-1106-preview",
+ "gpt-4-turbo-2024-04-09",
+ "gpt-4o-2024-05-13",
+ "gpt-4o-2024-08-06",
+ "gpt-4o-mini-2024-07-18",
+ "granite-3.0-2b-instruct",
+ "granite-3.0-8b-instruct",
+ "grok-2-2024-08-13",
+ "grok-2-mini-2024-08-13",
+ "hunyuan-standard-256k",
+ "internlm2_5-20b-chat",
+ "jamba-1.5-large",
+ "jamba-1.5-mini",
+ "llama-2-13b-chat",
+ "llama-2-70b-chat",
+ "llama-3-70b-instruct",
+ "llama-3-8b-instruct",
+ "llama-3.1-405b-instruct-bf16",
+ "llama-3.1-405b-instruct-fp8",
+ "llama-3.1-70b-instruct",
+ "llama-3.1-8b-instruct",
+ "llama-3.1-nemotron-51b-instruct",
+ "llama-3.1-nemotron-70b-instruct",
+ "llama-3.1-tulu-3-70b",
+ "llama-3.1-tulu-3-8b",
+ "llama-3.2-1b-instruct",
+ "llama-3.2-3b-instruct",
+ "llama-3.3-70b-instruct",
+ "ministral-8b-2410",
+ "mistral-7b-instruct-v0.2",
+ "mistral-large-2402",
+ "mistral-large-2407",
+ "mistral-large-2411",
+ "mistral-medium",
+ "mixtral-8x22b-instruct-v0.1",
+ "mixtral-8x7b-instruct-v0.1",
+ "nemotron-4-340b-instruct",
+ "o1-2024-12-17",
+ "o1-mini",
+ "o1-preview",
+ "phi-3-medium-4k-instruct",
+ "phi-3-mini-128k-instruct",
+ "phi-3-mini-4k-instruct",
+ "phi-3-mini-4k-instruct-june-2024",
+ "phi-3-small-8k-instruct",
+ "qwen-max-0428",
+ "qwen-max-0919",
+ "qwen-plus-0828",
+ "qwen1.5-110b-chat",
+ "qwen1.5-14b-chat",
+ "qwen1.5-32b-chat",
+ "qwen1.5-72b-chat",
+ "qwen2-72b-instruct",
+ "qwen2.5-72b-instruct",
+ "qwen2.5-coder-32b-instruct",
+ "qwen2.5-plus-1127",
+ "qwq-32b-preview",
+ "reka-core-20240501",
+ "reka-core-20240722",
+ "reka-core-20240904",
+ "reka-flash-20240722",
+ "reka-flash-20240904",
+ "reka-flash-21b-20240226",
+ "reka-flash-21b-20240226-online",
+ "reka-flash-preview-20240611",
+ "smollm2-1.7b-instruct",
+ "snowflake-arctic-instruct",
+ "yi-1.5-34b-chat",
+ "yi-34b-chat",
+ "yi-large",
+ "yi-large-preview",
+ "yi-lightning",
+ "yi-lightning-lite"
+]

special_tokens_map.json

@@ -0,0 +1,41 @@
+{
+  "additional_special_tokens": [
+    "<|im_start|>",
+    "<|im_end|>"
+  ],
+  "bos_token": {
+    "content": "<|im_start|>",
+    "lstrip": false,
+    "normalized": false,
+    "rstrip": false,
+    "single_word": false
+  },
+  "cls_token": {
+    "content": "<|cls|>",
+    "lstrip": false,
+    "normalized": false,
+    "rstrip": false,
+    "single_word": false
+  },
+  "eos_token": {
+    "content": "<|im_end|>",
+    "lstrip": false,
+    "normalized": false,
+    "rstrip": false,
+    "single_word": false
+  },
+  "pad_token": {
+    "content": "<|im_end|>",
+    "lstrip": false,
+    "normalized": false,
+    "rstrip": false,
+    "single_word": false
+  },
+  "unk_token": {
+    "content": "<|endoftext|>",
+    "lstrip": false,
+    "normalized": false,
+    "rstrip": false,
+    "single_word": false
+  }
+}

tokenizer_config.json

@@ -0,0 +1,164 @@
+{
+  "add_prefix_space": false,
+  "added_tokens_decoder": {
+    "0": {
+      "content": "<|endoftext|>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "1": {
+      "content": "<|im_start|>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "2": {
+      "content": "<|im_end|>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "3": {
+      "content": "<repo_name>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "4": {
+      "content": "<reponame>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "5": {
+      "content": "<file_sep>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "6": {
+      "content": "<filename>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "7": {
+      "content": "<gh_stars>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "8": {
+      "content": "<issue_start>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "9": {
+      "content": "<issue_comment>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "10": {
+      "content": "<issue_closed>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "11": {
+      "content": "<jupyter_start>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "12": {
+      "content": "<jupyter_text>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "13": {
+      "content": "<jupyter_code>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "14": {
+      "content": "<jupyter_output>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "15": {
+      "content": "<jupyter_script>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "16": {
+      "content": "<empty_output>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "49152": {
+      "content": "<|cls|>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    }
+  },
+  "additional_special_tokens": [
+    "<|im_start|>",
+    "<|im_end|>"
+  ],
+  "bos_token": "<|im_start|>",
+  "chat_template": "{% for message in messages %}{{'<|im_start|>' + message['role'] + '\n' + message['content'] + '<|im_end|>' + '\n'}}{% endfor %}{% if add_generation_prompt %}{{ '<|im_start|>assistant\n' }}{% endif %}",
+  "clean_up_tokenization_spaces": false,
+  "cls_token": "<|cls|>",
+  "eos_token": "<|im_end|>",
+  "extra_special_tokens": {},
+  "model_max_length": 8192,
+  "pad_token": "<|im_end|>",
+  "tokenizer_class": "GPT2Tokenizer",
+  "unk_token": "<|endoftext|>",
+  "vocab_size": 49152
+}

training_args.bin

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:11b33b8c24a28647f2443096916ba266d817102d164a04313abdadfaddaacd89
+size 7032

training_config.json

@@ -0,0 +1,21 @@
+{
+ "proj_name": "SmolLM2-360M-Instruct-bag-full-train-half-batch",
+ "learning_rate": 8e-06,
+ "adam_epsilon": 1e-08,
+ "batch_size": 4,
+ "max_length": 8192,
+ "num_train_epochs": 1,
+ "train_data_path": "full-p2l-bag-data-01082025",
+ "val_data_path": "p2el/canonical_bt_val_data_11092024",
+ "output_dir": "training_outputs",
+ "pretrain_model_name": "HuggingFaceTB/SmolLM2-360M-Instruct",
+ "gradient_accumulation_steps": 16,
+ "chat_template": "{% for message in messages %}{{'<|im_start|>' + message['role'] + '\n' + message['content'] + '<|im_end|>' + '\n'}}{% endfor %}{% if add_generation_prompt %}{{ '<|im_start|>assistant\n' }}{% endif %}",
+ "model_type": "llama",
+ "head_type": "rk",
+ "loss_type": "bag",
+ "weighted_loss": false,
+ "deepspeed_config_path": "deepspeed/zero1.json",
+ "init_type": "reset_params",
+ "load_train_data_from_disk": true
+}