You need to agree to share your Hugging Face information to access this model

This repository is publicly accessible, but you have to accept the conditions to access its files and content.

AuriStream100M_1Pred_BigAudioDataset_500k_seed1110-randinit

AuriStream is a speech language model by Greta Tuckute and Klemen Kotar.

This model predicts cochlear tokens from a tokenizer such as WavCochCausalV8192.

Native training step-zero initialization for the 100M 1-prediction model. This exactly uses origin seed 1110 and historical source commit 36b3287e5d6a59c81d49eb92769e4a9fb571d8e8, matching the start of W&B run rvvplus3. The weights are untrained FP32 values produced before XLA/FSDP wrapping.

Model Details

Parameter	Value
Parameters	~0.10B
Layers	12
Hidden Size	768
Attention Heads	12
Vocab Size	8192
Prediction Steps	1

Usage

from transformers import AutoModel, AutoConfig

# Load with trust_remote_code for custom model
model = AutoModel.from_pretrained(
    "TuKoResearch/AuriStream100M_1Pred_BigAudioDataset_500k_seed1110-randinit",
    trust_remote_code=True,
)

# Or load config first
config = AutoConfig.from_pretrained("TuKoResearch/AuriStream100M_1Pred_BigAudioDataset_500k_seed1110-randinit", trust_remote_code=True)

Base Model Code

This checkpoint uses shared model code from TuKoResearch/AuriStream-base.

Tokenizer

This model uses cochlear tokens from WavCochCausalV8192.

Seed and counterpart

Initialization seed: 1110. Untrained step-zero control. Matched trained model: TuKoResearch/AuriStream100M_1Pred_BigAudioDataset_500k_seed1110.

Downloads last month: 3

Safetensors

Model size

97.5M params

Tensor type

F32