See axolotl config

axolotl version: 0.4.1

adapter: lora
base_model: NousResearch/Yarn-Llama-2-7b-64k
bf16: true
chat_template: llama3
datasets:
- data_files:
  - ed1483b1fcb57061_train_data.json
  ds_type: json
  format: custom
  path: /workspace/input_data/ed1483b1fcb57061_train_data.json
  type:
    field_input: path
    field_instruction: content
    field_output: language
    format: '{instruction} {input}'
    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: false
fsdp: null
fsdp_config: null
gradient_accumulation_steps: 2
gradient_checkpointing: true
group_by_length: false
hub_model_id: lesso04/148ce154-2bf7-4be7-b7f9-1df59f0d20a4
hub_repo: null
hub_strategy: checkpoint
hub_token: null
learning_rate: 0.0001
load_in_4bit: false
load_in_8bit: false
local_rank: null
logging_steps: 1
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_memory:
  0: 77GiB
max_steps: 100
micro_batch_size: 8
mlflow_experiment_name: /tmp/ed1483b1fcb57061_train_data.json
model_type: AutoModelForCausalLM
num_epochs: 3
optimizer: adamw_torch
output_dir: miner_id_24
pad_to_sequence_len: true
resume_from_checkpoint: null
s2_attention: null
sample_packing: false
save_steps: 25
save_strategy: steps
sequence_len: 1024
strict: false
tf32: false
tokenizer_type: AutoTokenizer
train_on_inputs: false
trust_remote_code: true
val_set_size: 0.05
wandb_entity: null
wandb_mode: online
wandb_name: 148ce154-2bf7-4be7-b7f9-1df59f0d20a4
wandb_project: Gradients-On-Demand
wandb_run: your_name
wandb_runid: 148ce154-2bf7-4be7-b7f9-1df59f0d20a4
warmup_steps: 10
weight_decay: 0.01
xformers_attention: false

148ce154-2bf7-4be7-b7f9-1df59f0d20a4

This model is a fine-tuned version of NousResearch/Yarn-Llama-2-7b-64k on the None dataset. It achieves the following results on the evaluation set:

• Loss: 0.0305

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: 0.0001
• train_batch_size: 8
• eval_batch_size: 8
• seed: 42
• gradient_accumulation_steps: 2
• total_train_batch_size: 16
• optimizer: Use OptimizerNames.ADAMW_TORCH 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
26.79	0.0003	1	10.0719
14.6411	0.0024	9	7.5024
6.3552	0.0048	18	2.5800
0.7602	0.0072	27	0.2037
0.1166	0.0096	36	0.0715
0.1665	0.0120	45	0.0766
0.0217	0.0143	54	0.0466
0.1841	0.0167	63	0.0346
0.2604	0.0191	72	0.0362
0.0035	0.0215	81	0.0307
0.0099	0.0239	90	0.0310
0.0065	0.0263	99	0.0305

Framework versions

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