Upload AveyForCausalLM

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 +1,17 @@
+{
+  "architectures": [
+    "AveyForCausalLM"
+  ],
+  "auto_map": {
+    "AutoConfig": "configuration_avey.AveyConfig",
+    "AutoModelForCausalLM": "modeling_avey.AveyForCausalLM"
+  },
+  "context_len": 256,
+  "d_embed": 768,
+  "expansion_factor": 4,
+  "model_type": "avey",
+  "n_blocks": 22,
+  "torch_dtype": "float32",
+  "transformers_version": "4.49.0",
+  "vocab_size": 50281
+}

configuration_avey.py

@@ -0,0 +1,21 @@
+from transformers import PretrainedConfig
+
+
+class AveyConfig(PretrainedConfig):
+    model_type = "avey"
+
+    def __init__(
+        self,
+        vocab_size: int = 50281,
+        d_embed: int = 768,
+        n_blocks: int = 22,
+        expansion_factor: int = 4,
+        context_len: int = 256,
+        **kwargs
+    ):
+        self.vocab_size = vocab_size
+        self.d_embed = d_embed
+        self.n_blocks = n_blocks
+        self.expansion_factor = expansion_factor
+        self.context_len = context_len
+        super().__init__(**kwargs)

generation_config.json

@@ -0,0 +1,4 @@
+{
+  "_from_model_config": true,
+  "transformers_version": "4.49.0"
+}

model.safetensors

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:ce4801c145645ad98e2bfa923a8e498d8d2461bf245ad92f0a564f7bdb6bb49b
+size 498031496

modeling_avey.py

@@ -0,0 +1,99 @@
+import torch
+import torch.nn as nn
+import torch.nn.functional as F
+from transformers import PreTrainedModel, GenerationMixin
+from transformers.modeling_outputs import CausalLMOutput
+from configuration_avey import AveyConfig
+
+
+class SGU(nn.Module):
+    def __init__(self, config: AveyConfig):
+        super().__init__()
+        self.ctxt_mat = nn.Parameter(torch.empty(config.context_len, config.context_len))
+        nn.init.xavier_normal_(self.ctxt_mat)
+
+    def cosim(self, embeddings: torch.Tensor) -> torch.Tensor:
+        norm = torch.sqrt(torch.sum(embeddings ** 2, dim=-1, keepdim=True) + 1e-8)
+        normalized = embeddings / norm
+        cosim = torch.matmul(normalized, normalized.transpose(-1, -2))
+        return cosim
+
+    def forward(self, x: torch.Tensor) -> torch.Tensor:
+        x0, x1 = x.chunk(2, dim=-1)
+        c = torch.tril(self.cosim(x0)) * torch.tril(self.ctxt_mat)
+        x0 = c @ x0
+        output = x0 * x1
+        return output
+
+
+class NeuralContextualizerLayer(nn.Module):
+    def __init__(self, config: AveyConfig):
+        super().__init__()
+        self.split_factor = [
+            int(config.d_embed * config.expansion_factor * 0.75),
+            int(config.d_embed * config.expansion_factor * 0.25)
+        ]
+        self.enricher = nn.Linear(config.d_embed, config.d_embed * config.expansion_factor)
+        self.sgu = SGU(config)
+        proj_in_features = int(
+            config.d_embed * config.expansion_factor * 0.5 + config.d_embed * 0.5
+        )
+        self.fuser = nn.Linear(proj_in_features, config.d_embed)
+
+    def forward(self, x: torch.Tensor) -> torch.Tensor:
+        x_proj = F.gelu(self.enricher(x))
+        x0, x1 = x_proj.split(self.split_factor, dim=-1)
+        x0 = self.sgu(x0)
+        combined = torch.cat([x0, x1], dim=-1)
+        return self.fuser(combined)
+
+
+class AveyBlock(nn.Module):
+    def __init__(self, config: AveyConfig):
+        super().__init__()
+        self.rms_norm = nn.RMSNorm(config.d_embed, eps=1e-10)
+        self.ctxt = NeuralContextualizerLayer(config)
+
+    def forward(self, x: torch.Tensor) -> torch.Tensor:
+        return x + self.ctxt(self.rms_norm(x))
+
+
+class AveyForCausalLM(PreTrainedModel, GenerationMixin):
+    config_class = AveyConfig
+
+    def __init__(self, config):
+        super().__init__(config)
+        self.config = config
+
+        self.wte = nn.Embedding(config.vocab_size, config.d_embed)
+        nn.init.xavier_normal_(self.wte.weight)
+
+        self.blocks = nn.ModuleList([AveyBlock(config) for _ in range(config.n_blocks)])
+        self.ln_f = nn.RMSNorm(config.d_embed, eps=1e-10)
+
+    def forward(self, input_ids: torch.Tensor, labels: torch.Tensor = None, **kwargs):
+        x = self.wte(input_ids)
+        B, T, E = x.shape
+
+        padded = False
+        orig_T = T
+        if T % self.config.context_len != 0:
+            pad_length = self.config.context_len - (T % self.config.context_len)
+            pad_tensor = torch.zeros(B, pad_length, E, device=x.device, dtype=x.dtype)
+            x = torch.cat([x, pad_tensor], dim=1)
+            T = x.shape[1]
+            padded = True
+
+        for block in self.blocks:
+            x = block(x)
+
+        logits = F.linear(self.ln_f(x), self.wte.weight)
+
+        if padded:
+            logits = logits[:, :orig_T, :]
+
+        if labels is not None:
+            loss = F.cross_entropy(logits.view(-1, logits.size(-1)), labels.view(-1))
+            return CausalLMOutput(logits=logits, loss=loss)
+
+        return CausalLMOutput(logits=logits)