Hugging Face's logo

FourOhFour
/
Virgil_9B

Text Generation
Transformers
Safetensors
gemma2
Generated from Trainer
conversational
text-generation-inference
Inference Endpoints
Model card Files Community
Virgil_9B / README.md
jeiku's picture
jeiku
Upload folder using huggingface_hub
4e1c36e verified
preview code
|
raw
history blame contribute delete
3.42 kB
metadata
library_name: transformers
license: gemma
base_model: jeiku/Dante_9B
tags:
  - generated_from_trainer
model-index:
  - name: outputs/out
    results: []

Built with Axolotl

See axolotl config

axolotl version: 0.4.1

base_model: jeiku/Dante_9B
model_type: AutoModelForCausalLM
tokenizer_type: AutoTokenizer

load_in_8bit: false
load_in_4bit: false
strict: false

datasets:
  - path: FourOhFour/RP_Phase
    type: sharegpt
    conversation: chatml

chat_template: chatml

val_set_size: 0.0025
output_dir: ./outputs/out

adapter:
lora_r:
lora_alpha:
lora_dropout:
lora_target_linear:

sequence_len: 8192
sample_packing: true
eval_sample_packing: false
pad_to_sequence_len: true

plugins:
  - axolotl.integrations.liger.LigerPlugin
liger_rope: true
liger_rms_norm: false
liger_swiglu: true
liger_fused_linear_cross_entropy: false

wandb_project: chatml9B
wandb_entity:
wandb_watch:
wandb_name: chatml9B
wandb_log_model:

gradient_accumulation_steps: 32
micro_batch_size: 1
num_epochs: 2
optimizer: adamw_bnb_8bit
lr_scheduler: cosine
learning_rate: 0.000008
weight_decay: 0.05

train_on_inputs: false
group_by_length: false
bf16: auto
fp16:
tf32: true

gradient_checkpointing: true
early_stopping_patience:
resume_from_checkpoint:
local_rank:
logging_steps: 1
xformers_attention:
flash_attention: true

warmup_ratio: 0.1
evals_per_epoch: 4
eval_table_size:
eval_max_new_tokens: 128
saves_per_epoch: 2

debug:
deepspeed: deepspeed_configs/zero3_bf16.json
fsdp:
fsdp_config:

special_tokens:
  pad_token: <pad>

outputs/out

This model is a fine-tuned version of jeiku/Dante_9B on the None dataset. It achieves the following results on the evaluation set:

  • Loss: 0.7075

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: 8e-06
  • train_batch_size: 1
  • eval_batch_size: 1
  • seed: 42
  • distributed_type: multi-GPU
  • num_devices: 4
  • gradient_accumulation_steps: 32
  • total_train_batch_size: 128
  • total_eval_batch_size: 4
  • optimizer: Adam with betas=(0.9,0.999) and epsilon=1e-08
  • lr_scheduler_type: cosine
  • lr_scheduler_warmup_steps: 14
  • num_epochs: 2

Training results

Training Loss Epoch Step Validation Loss
1.7474 0.0135 1 1.7996
1.6968 0.2570 19 0.9551
1.6583 0.5139 38 0.8805
1.5418 0.7709 57 0.7926
1.3997 1.0271 76 0.7500
1.3921 1.2847 95 0.7168
1.4141 1.5424 114 0.7155
1.4139 1.8 133 0.7075

Framework versions

  • Transformers 4.46.0.dev0
  • Pytorch 2.4.0+cu121
  • Datasets 2.21.0
  • Tokenizers 0.20.0