Upload StudentPRM

README.md

@@ -0,0 +1,199 @@
+---
+library_name: transformers
+tags: []
+---
+
+# Model Card for Model ID
+
+<!-- Provide a quick summary of what the model is/does. -->
+
+
+
+## Model Details
+
+### Model Description
+
+<!-- Provide a longer summary of what this model is. -->
+
+This is the model card of a 🤗 transformers model that has been pushed on the Hub. This model card has been automatically generated.
+
+- **Developed by:** [More Information Needed]
+- **Funded by [optional]:** [More Information Needed]
+- **Shared by [optional]:** [More Information Needed]
+- **Model type:** [More Information Needed]
+- **Language(s) (NLP):** [More Information Needed]
+- **License:** [More Information Needed]
+- **Finetuned from model [optional]:** [More Information Needed]
+
+### Model Sources [optional]
+
+<!-- Provide the basic links for the model. -->
+
+- **Repository:** [More Information Needed]
+- **Paper [optional]:** [More Information Needed]
+- **Demo [optional]:** [More Information Needed]
+
+## Uses
+
+<!-- Address questions around how the model is intended to be used, including the foreseeable users of the model and those affected by the model. -->
+
+### Direct Use
+
+<!-- This section is for the model use without fine-tuning or plugging into a larger ecosystem/app. -->
+
+[More Information Needed]
+
+### Downstream Use [optional]
+
+<!-- This section is for the model use when fine-tuned for a task, or when plugged into a larger ecosystem/app -->
+
+[More Information Needed]
+
+### Out-of-Scope Use
+
+<!-- This section addresses misuse, malicious use, and uses that the model will not work well for. -->
+
+[More Information Needed]
+
+## Bias, Risks, and Limitations
+
+<!-- This section is meant to convey both technical and sociotechnical limitations. -->
+
+[More Information Needed]
+
+### Recommendations
+
+<!-- This section is meant to convey recommendations with respect to the bias, risk, and technical limitations. -->
+
+Users (both direct and downstream) should be made aware of the risks, biases and limitations of the model. More information needed for further recommendations.
+
+## How to Get Started with the Model
+
+Use the code below to get started with the model.
+
+[More Information Needed]
+
+## Training Details
+
+### Training Data
+
+<!-- This should link to a Dataset Card, perhaps with a short stub of information on what the training data is all about as well as documentation related to data pre-processing or additional filtering. -->
+
+[More Information Needed]
+
+### Training Procedure
+
+<!-- This relates heavily to the Technical Specifications. Content here should link to that section when it is relevant to the training procedure. -->
+
+#### Preprocessing [optional]
+
+[More Information Needed]
+
+
+#### Training Hyperparameters
+
+- **Training regime:** [More Information Needed] <!--fp32, fp16 mixed precision, bf16 mixed precision, bf16 non-mixed precision, fp16 non-mixed precision, fp8 mixed precision -->
+
+#### Speeds, Sizes, Times [optional]
+
+<!-- This section provides information about throughput, start/end time, checkpoint size if relevant, etc. -->
+
+[More Information Needed]
+
+## Evaluation
+
+<!-- This section describes the evaluation protocols and provides the results. -->
+
+### Testing Data, Factors & Metrics
+
+#### Testing Data
+
+<!-- This should link to a Dataset Card if possible. -->
+
+[More Information Needed]
+
+#### Factors
+
+<!-- These are the things the evaluation is disaggregating by, e.g., subpopulations or domains. -->
+
+[More Information Needed]
+
+#### Metrics
+
+<!-- These are the evaluation metrics being used, ideally with a description of why. -->
+
+[More Information Needed]
+
+### Results
+
+[More Information Needed]
+
+#### Summary
+
+
+
+## Model Examination [optional]
+
+<!-- Relevant interpretability work for the model goes here -->
+
+[More Information Needed]
+
+## Environmental Impact
+
+<!-- Total emissions (in grams of CO2eq) and additional considerations, such as electricity usage, go here. Edit the suggested text below accordingly -->
+
+Carbon emissions can be estimated using the [Machine Learning Impact calculator](https://mlco2.github.io/impact#compute) presented in [Lacoste et al. (2019)](https://arxiv.org/abs/1910.09700).
+
+- **Hardware Type:** [More Information Needed]
+- **Hours used:** [More Information Needed]
+- **Cloud Provider:** [More Information Needed]
+- **Compute Region:** [More Information Needed]
+- **Carbon Emitted:** [More Information Needed]
+
+## Technical Specifications [optional]
+
+### Model Architecture and Objective
+
+[More Information Needed]
+
+### Compute Infrastructure
+
+[More Information Needed]
+
+#### Hardware
+
+[More Information Needed]
+
+#### Software
+
+[More Information Needed]
+
+## Citation [optional]
+
+<!-- If there is a paper or blog post introducing the model, the APA and Bibtex information for that should go in this section. -->
+
+**BibTeX:**
+
+[More Information Needed]
+
+**APA:**
+
+[More Information Needed]
+
+## Glossary [optional]
+
+<!-- If relevant, include terms and calculations in this section that can help readers understand the model or model card. -->
+
+[More Information Needed]
+
+## More Information [optional]
+
+[More Information Needed]
+
+## Model Card Authors [optional]
+
+[More Information Needed]
+
+## Model Card Contact
+
+[More Information Needed]

config.json

@@ -0,0 +1,75 @@
+{
+  "_name_or_path": "omrisap/Qwen2.5-Math-PRM-1.5B",
+  "add_cross_attention": false,
+  "architectures": [
+    "StudentPRM"
+  ],
+  "auto_map": {
+    "AutoConfig": "modeling_student_prm.StudentPRMConfig",
+    "AutoModel": "modeling_student_prm.StudentPRM"
+  },
+  "bad_words_ids": null,
+  "base_model_name": "Qwen/Qwen2.5-Math-1.5B",
+  "begin_suppress_tokens": null,
+  "bos_token_id": null,
+  "chunk_size_feed_forward": 0,
+  "cross_attention_hidden_size": null,
+  "decoder_start_token_id": null,
+  "diversity_penalty": 0.0,
+  "do_sample": false,
+  "dtype": "bfloat16",
+  "early_stopping": false,
+  "encoder_no_repeat_ngram_size": 0,
+  "eos_token_id": null,
+  "exponential_decay_length_penalty": null,
+  "finetuning_task": null,
+  "forced_bos_token_id": null,
+  "forced_eos_token_id": null,
+  "has_no_defaults_at_init": true,
+  "hidden_size": 1536,
+  "id2label": {
+    "0": "LABEL_0",
+    "1": "LABEL_1"
+  },
+  "is_decoder": false,
+  "is_encoder_decoder": false,
+  "label2id": {
+    "LABEL_0": 0,
+    "LABEL_1": 1
+  },
+  "length_penalty": 1.0,
+  "max_length": 20,
+  "min_length": 0,
+  "model_type": "student_prm",
+  "no_repeat_ngram_size": 0,
+  "num_beam_groups": 1,
+  "num_beams": 1,
+  "num_return_sequences": 1,
+  "output_attentions": false,
+  "output_hidden_states": false,
+  "output_scores": false,
+  "pad_token_id": null,
+  "pool_token": "</think>",
+  "prefix": null,
+  "problem_type": null,
+  "pruned_heads": {},
+  "remove_invalid_values": false,
+  "repetition_penalty": 1.0,
+  "return_dict": true,
+  "return_dict_in_generate": false,
+  "sep_token_id": null,
+  "suppress_tokens": null,
+  "task_specific_params": null,
+  "temperature": 1.0,
+  "tf_legacy_loss": false,
+  "tie_encoder_decoder": false,
+  "tie_word_embeddings": true,
+  "tokenizer_class": null,
+  "top_k": 50,
+  "top_p": 1.0,
+  "torchscript": false,
+  "transformers_version": "4.57.0",
+  "typical_p": 1.0,
+  "use_bfloat16": false,
+  "vocab_size": 151666
+}

model.safetensors

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:94bcba76178a03dc3894fd7dd6918677a78042cb3c827795ab6c1b288e41978f
+size 3086643684

modeling_student_prm.py

@@ -0,0 +1,121 @@
+#!/usr/bin/env python3
+"""Hugging Face compatible StudentPRM model.
+
+Provides StudentPRMConfig + StudentPRM so that after training you can
+push to the hub and later load with:
+
+    from transformers import AutoTokenizer, AutoModel
+    tok = AutoTokenizer.from_pretrained(repo_id, trust_remote_code=True)
+    model = AutoModel.from_pretrained(repo_id, trust_remote_code=True)
+
+The 2-logit PRM head is included. The model pools the final hidden state
+at the last occurrence of the configured pool token (e.g. </think>)."""
+from typing import Optional
+import torch
+import torch.nn as nn
+from transformers import (
+    AutoModel,
+    PreTrainedModel,
+    PretrainedConfig,
+)
+from transformers.modeling_outputs import SequenceClassifierOutput
+
+
+def last_token_index(input_ids: torch.Tensor, token_id: int) -> torch.Tensor:
+    mask = (input_ids == token_id)
+    flipped = torch.flip(mask, dims=[1]).int().argmax(dim=1)
+    return (input_ids.shape[1] - 1) - flipped
+
+
+class StudentPRMConfig(PretrainedConfig):
+    model_type = "student_prm"
+
+    def __init__(
+        self,
+        base_model_name: str,
+        pool_token: str,
+        hidden_size: int,
+        vocab_size: int,
+        num_labels: int = 2,
+        **kwargs,
+    ):
+        super().__init__(**kwargs)
+        self.base_model_name = base_model_name
+        self.pool_token = pool_token
+        self.hidden_size = hidden_size
+        self.vocab_size = vocab_size
+        self.num_labels = num_labels
+        self.auto_map = {"AutoModel": "modeling_student_prm.StudentPRM", "AutoConfig": "modeling_student_prm.StudentPRMConfig"}
+        self.architectures = ["StudentPRM"]
+        # Prevent default-diff logic from re-instantiating without required args
+        self.has_no_defaults_at_init = True
+
+    def _get_non_default_generation_parameters(self):
+        # Classification model; no generation params to diff
+        # Returning empty dict prevents transformers from calling self.__class__() with missing args
+        return {}
+
+    def to_diff_dict(self):  # override to bypass default instantiation logic
+        return self.to_dict()
+
+
+class StudentPRM(PreTrainedModel):
+    config_class = StudentPRMConfig
+
+    def __init__(self, config: StudentPRMConfig, base: Optional[PreTrainedModel] = None, tokenizer=None):
+        super().__init__(config)
+        if base is None:
+            # Load base model; rely on remote code if needed
+            base = AutoModel.from_pretrained(config.base_model_name, trust_remote_code=True)
+        # Resize embeddings if vocab changed due to added special tokens
+        try:
+            current_vocab = base.get_input_embeddings().weight.shape[0]
+            if current_vocab != config.vocab_size:
+                base.resize_token_embeddings(config.vocab_size)
+        except Exception:
+            pass
+        self.base = base
+        self.head = nn.Linear(config.hidden_size, config.num_labels)
+        self.tokenizer = tokenizer  # optional, only needed for pool id resolution when provided
+        if tokenizer is not None and config.pool_token in tokenizer.get_vocab():
+            self.pool_id = tokenizer.convert_tokens_to_ids(config.pool_token)
+        else:
+            # Will be resolved later if tokenizer added special token dynamically
+            self.pool_id = None
+        self.post_init()
+
+    def _resolve_pool_id(self, input_ids: torch.Tensor):
+        if self.pool_id is None and self.tokenizer is not None:
+            self.pool_id = self.tokenizer.convert_tokens_to_ids(self.config.pool_token)
+        if self.pool_id is None:
+            raise ValueError("pool_id not set and tokenizer unavailable to resolve it.")
+        return self.pool_id
+
+    def forward(
+        self,
+        input_ids: torch.Tensor,
+        attention_mask: Optional[torch.Tensor] = None,
+        **kwargs,
+    ) -> SequenceClassifierOutput:
+        pool_id = self._resolve_pool_id(input_ids)
+        out = self.base(
+            input_ids=input_ids,
+            attention_mask=attention_mask,
+            output_hidden_states=True,
+            use_cache=False,
+        )
+        hidden = out.hidden_states[-1]
+        pos = last_token_index(input_ids, pool_id)
+        pooled = hidden[torch.arange(len(input_ids), device=input_ids.device), pos]
+        # Match head weight dtype for safety (bfloat16 training etc.)
+        if pooled.dtype != self.head.weight.dtype:
+            pooled = pooled.to(self.head.weight.dtype)
+        logits = self.head(pooled)
+        return SequenceClassifierOutput(logits=logits)
+
+    def save_pretrained(self, save_directory: str, *args, **kwargs):
+        if not getattr(self.config, "auto_map", None):
+            self.config.auto_map = {"AutoModel": "modeling_student_prm.StudentPRM", "AutoConfig": "modeling_student_prm.StudentPRMConfig"}
+        if not getattr(self.config, "architectures", None):
+            self.config.architectures = ["StudentPRM"]
+        super().save_pretrained(save_directory, *args, **kwargs)