Modelpublic

estrogen/SomehowMinistralReturnedButCommunist

sourceHugging Faceapache-2.0updated 8mo agoView on Hugging Face

1likes30downloads

Model Card

<img src="https://raw.githubusercontent.com/axolotl-ai-cloud/axolotl/main/image/axolotl-badge-web.png" alt="Built with Axolotl" width="200" height="32"/> <details><summary>See axolotl config</summary>

axolotl version: 0.15.0.dev0

yaml

# === Model Configuration ===
base_model: Aratako/Ministral-3-14B-Instruct-2512-BF16-TextOnly
load_in_8bit: false
load_in_4bit: false

# === Training Setup ===
num_epochs: 2
micro_batch_size: 2
gradient_accumulation_steps: 1
sequence_len: 2048
sample_packing: true
pad_to_sequence_len: true

# === Hyperparameter Configuration ===
optimizer: adamw_torch_8bit
learning_rate: 1e-5
lr_scheduler: constant
weight_decay: 0.001
max_grad_norm: 0.1
warmup_ratio: 0.05
cosine_min_lr_ratio: 0.1

# === Data Configuration ===
datasets:
  - path: WokeAI/polititune-tankie-warmup-3
    type: chat_template
    split: train
chat_template: tokenizer_default

dataset_prepared_path: last_run_prepared

# === Hardware Optimization ===
gradient_checkpointing: offload

# === Wandb Tracking ===
wandb_project: polititune-3-mstr14b-warmup

# === Checkpointing ===
saves_per_epoch: 1

# === Advanced Settings ===
output_dir: ./model-output
bf16: auto
flash_attention: true
train_on_inputs: false
group_by_length: false
logging_steps: 1
trust_remote_code: false

plugins:
  - axolotl.integrations.liger.LigerPlugin
  - axolotl.integrations.cut_cross_entropy.CutCrossEntropyPlugin

fsdp:
  - auto_wrap
  - full_shard
fsdp_config:
  fsdp_version: 2
  fsdp_offload_params: false
  fsdp_cpu_ram_efficient_loading: true
  fsdp_auto_wrap_policy: TRANSFORMER_BASED_WRAP
  fsdp_transformer_layer_cls_to_wrap: Ministral3DecoderLayer
  fsdp_state_dict_type: SHARDED_STATE_DICT
  fsdp_sharding_strategy: FULL_SHARD
  fsdp_reshard_after_forward: true
  fsdp_activation_checkpointing: true # will disable if doesnt work

</details><br>

model-output

This model is a fine-tuned version of Aratako/Ministral-3-14B-Instruct-2512-BF16-TextOnly on the WokeAI/polititune-tankie-warmup-3 dataset.

Model description

More information needed

Intended uses & limitations

More information needed

Training and evaluation data

More information needed

Training procedure

Training hyperparameters

The following hyperparameters were used during training:

—learning_rate: 1e-05
—trainbatchsize: 2
—evalbatchsize: 2
—seed: 42
—distributed_type: multi-GPU
—num_devices: 8
—totaltrainbatch_size: 16
—totalevalbatch_size: 16
—optimizer: Use OptimizerNames.ADAMWTORCH8BIT with betas=(0.9,0.999) and epsilon=1e-08 and optimizer_args=No additional optimizer arguments
—lrschedulertype: constant
—lrschedulerwarmup_steps: 3
—training_steps: 60

Training results

Framework versions

—Transformers 5.0.0
—Pytorch 2.8.0+cu128
—Datasets 4.5.0
—Tokenizers 0.22.2