Upload AuriStream Parallel checkpoint

README.md

@@ -0,0 +1,54 @@
+---
+license: apache-2.0
+tags:
+- audio
+- speech
+- language-model
+- auristream
+- discrete-diffusion
+library_name: transformers
+---
+
+# AuriStreamParallel100M_Group4_BigAudioDataset_440k
+
+**AuriStream Parallel** is a discrete diffusion speech language model by **Greta Tuckute** and **Klemen Kotar**.
+
+
+
+## Model Details
+
+| Parameter | Value |
+|-----------|-------|
+| Parameters | ~0.12B |
+| Layers | 12 |
+| Hidden Size | 768 |
+| Attention Heads | 12 |
+| Vocab Size | 8193 |
+| Group Size | 4 |
+| Mask Schedule | linear_text_prime |
+
+## Architecture
+
+- Bidirectional transformer attention
+- Grouped token latent projection
+- Parallel token heads for group-wise prediction
+- Partial masking diffusion objective
+
+## Usage
+
+```python
+from transformers import AutoModel
+
+model = AutoModel.from_pretrained(
+    "TuKoResearch/AuriStreamParallel100M_Group4_BigAudioDataset_440k",
+    trust_remote_code=True,
+)
+```
+
+## Base Model Code
+
+This checkpoint uses shared model code from [TuKoResearch/AuriStreamParallel-base](https://huggingface.co/TuKoResearch/AuriStreamParallel-base).
+
+## Tokenizer
+
+This model is intended for cochlear tokens, e.g. from [WavCochCausalV8192](https://huggingface.co/TuKoResearch/WavCochCausalV8192).

config.json

@@ -0,0 +1,27 @@
+{
+  "model_type": "AuriStreamParallel",
+  "architectures": [
+    "AuriStreamModel"
+  ],
+  "auto_map": {
+    "AutoConfig": "TuKoResearch/AuriStreamParallel-base--configuration_auristream_parallel.AuriStreamParallelConfig",
+    "AutoModel": "TuKoResearch/AuriStreamParallel-base--modeling_auristream_parallel.AuriStreamModel",
+    "AutoModelForCausalLM": "TuKoResearch/AuriStreamParallel-base--modeling_auristream_parallel.AuriStreamModel"
+  },
+  "torch_dtype": "bfloat16",
+  "transformers_version": "4.40.0",
+  "vocab_size": 8193,
+  "base_vocab_size": 8192,
+  "mask_token_id": 8192,
+  "ignore_index": -100,
+  "n_embd": 768,
+  "n_layer": 12,
+  "n_head": 12,
+  "dropout": 0.0,
+  "bias": false,
+  "use_rope": true,
+  "group_size": 4,
+  "seq_len": 4096,
+  "skip_connections": false,
+  "mask_schedule": "linear_text_prime"
+}

model.safetensors

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:80c128564d981737c59c65a52b36e1e9d6ca60592f96aa8760143faaa68d5153
+size 475107024