See axolotl config

axolotl version: 0.4.1

adapter: lora
base_model: NousResearch/Yarn-Mistral-7b-128k
bf16: true
chat_template: llama3
dataset_prepared_path: null
datasets:
- data_files:
  - 8c4ff8cee4cec4f4_train_data.json
  ds_type: json
  format: custom
  path: /workspace/input_data/8c4ff8cee4cec4f4_train_data.json
  type:
    field_instruction: title
    field_output: text
    format: '{instruction}'
    no_input_format: '{instruction}'
    system_format: '{system}'
    system_prompt: ''
debug: null
deepspeed: null
early_stopping_patience: 2
eval_max_new_tokens: 128
eval_steps: 100
eval_table_size: null
flash_attention: false
fp16: null
fsdp: null
fsdp_config: null
gradient_accumulation_steps: 8
gradient_checkpointing: true
group_by_length: false
hub_model_id: Romain-XV/8ce2d2b2-df8e-49c2-991a-52d0e53001e8
hub_repo: null
hub_strategy: checkpoint
hub_token: null
learning_rate: 0.0002
load_best_model_at_end: true
load_in_4bit: false
load_in_8bit: false
local_rank: null
logging_steps: 1
lora_alpha: 128
lora_dropout: 0.3
lora_fan_in_fan_out: null
lora_model_dir: null
lora_r: 64
lora_target_linear: true
lr_scheduler: cosine
max_grad_norm: 1.0
max_steps: 714
micro_batch_size: 4
mlflow_experiment_name: /tmp/8c4ff8cee4cec4f4_train_data.json
model_type: AutoModelForCausalLM
num_epochs: 2
optimizer: adamw_bnb_8bit
output_dir: miner_id_24
pad_to_sequence_len: true
resume_from_checkpoint: null
s2_attention: null
sample_packing: false
save_steps: 100
sequence_len: 2048
special_tokens:
  pad_token: </s>
strict: false
tf32: true
tokenizer_type: AutoTokenizer
train_on_inputs: false
trust_remote_code: true
use_rslora: true
val_set_size: 0.021045896891941945
wandb_entity: null
wandb_mode: online
wandb_name: 15bd6132-1c7e-43e2-afd4-1aee8666ab1b
wandb_project: Gradients-On-Demand
wandb_run: your_name
wandb_runid: 15bd6132-1c7e-43e2-afd4-1aee8666ab1b
warmup_steps: 10
weight_decay: 0.0
xformers_attention: null

8ce2d2b2-df8e-49c2-991a-52d0e53001e8

This model is a fine-tuned version of NousResearch/Yarn-Mistral-7b-128k on the None dataset. It achieves the following results on the evaluation set:

• Loss: 1.3366

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: 0.0002
• train_batch_size: 4
• eval_batch_size: 4
• seed: 42
• gradient_accumulation_steps: 8
• 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: 714

Training results

Training Loss	Epoch	Step	Validation Loss
12.4572	0.0001	1	1.5985
10.964	0.0138	100	1.3773
9.1479	0.0275	200	1.3809
10.1482	0.0413	300	1.3753
12.1973	0.0550	400	1.3625
13.0282	0.0688	500	1.3484
8.9636	0.0826	600	1.3389
9.1362	0.0963	700	1.3366

Framework versions

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