See axolotl config

axolotl version: 0.4.1

adapter: lora
base_model: NousResearch/Yarn-Mistral-7b-64k
bf16: auto
chat_template: llama3
dataset_prepared_path: null
datasets:
- data_files:
  - e1f1767c4221a0d6_train_data.json
  ds_type: json
  format: custom
  path: /workspace/input_data/e1f1767c4221a0d6_train_data.json
  type:
    field_instruction: en_US
    field_output: de_DE
    format: '{instruction}'
    no_input_format: '{instruction}'
    system_format: '{system}'
    system_prompt: ''
debug: null
deepspeed: null
early_stopping_patience: null
eval_max_new_tokens: 128
eval_table_size: null
evals_per_epoch: 4
flash_attention: false
fp16: null
fsdp: null
fsdp_config: null
gradient_accumulation_steps: 4
gradient_checkpointing: true
group_by_length: false
hub_model_id: clarxus/94f2ab45-5583-479e-b77d-dae633639fcf
hub_repo: null
hub_strategy: checkpoint
hub_token: null
learning_rate: 5.0e-05
load_in_4bit: false
load_in_8bit: false
local_rank: 0
logging_steps: 3
lora_alpha: 32
lora_dropout: 0.05
lora_fan_in_fan_out: null
lora_model_dir: null
lora_r: 16
lora_target_linear: true
lr_scheduler: cosine
max_steps: 100
micro_batch_size: 8
mlflow_experiment_name: /tmp/e1f1767c4221a0d6_train_data.json
model_type: AutoModelForCausalLM
num_epochs: 3
optimizer: adamw_bnb_8bit
output_dir: miner_id_24
pad_to_sequence_len: true
resume_from_checkpoint: null
s2_attention: null
sample_packing: false
saves_per_epoch: 4
sequence_len: 1024
special_tokens:
  pad_token: </s>
strict: false
tf32: false
tokenizer_type: AutoTokenizer
train_on_inputs: false
trust_remote_code: true
val_set_size: 0.05
wandb_entity: techspear-hub
wandb_mode: online
wandb_name: 593f218d-08e8-48a1-99c3-08b90210f9be
wandb_project: Gradients-On-Seven
wandb_run: your_name
wandb_runid: 593f218d-08e8-48a1-99c3-08b90210f9be
warmup_steps: 10
weight_decay: 0.0
xformers_attention: null

94f2ab45-5583-479e-b77d-dae633639fcf

This model is a fine-tuned version of NousResearch/Yarn-Mistral-7b-64k on the None dataset. It achieves the following results on the evaluation set:

• Loss: 0.8181

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: 5e-05
• train_batch_size: 8
• eval_batch_size: 8
• seed: 42
• gradient_accumulation_steps: 4
• total_train_batch_size: 32
• optimizer: Use OptimizerNames.ADAMW_BNB with betas=(0.9,0.999) and epsilon=1e-08 and optimizer_args=No additional optimizer arguments
• lr_scheduler_type: cosine
• lr_scheduler_warmup_steps: 10
• training_steps: 100

Training results

Training Loss	Epoch	Step	Validation Loss
No log	0.0021	1	3.6666
12.0429	0.0185	9	2.3610
5.2257	0.0369	18	1.2831
3.8667	0.0554	27	0.9862
3.7173	0.0738	36	0.9269
3.7293	0.0923	45	0.8883
3.4031	0.1108	54	0.8580
3.1709	0.1292	63	0.8416
3.7806	0.1477	72	0.8304
3.3891	0.1662	81	0.8223
3.2509	0.1846	90	0.8187
3.2285	0.2031	99	0.8181

Framework versions

• PEFT 0.13.2
• Transformers 4.46.0
• Pytorch 2.5.0+cu124
• Datasets 3.0.1
• Tokenizers 0.20.1