See axolotl config

axolotl version: 0.4.1

adapter: lora
base_model: katuni4ka/tiny-random-olmo-hf
bf16: auto
chat_template: llama3
dataset_prepared_path: null
datasets:
- data_files:
  - 49ba5517891ceace_train_data.json
  ds_type: json
  format: custom
  path: /workspace/input_data/49ba5517891ceace_train_data.json
  type:
    field_input: text
    field_instruction: query
    field_output: answer
    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: true
fp16: null
fsdp: null
fsdp_config: null
gradient_accumulation_steps: 4
gradient_checkpointing: true
gradient_clipping: 1.0
group_by_length: false
hub_model_id: oldiday/a64fb4ac-f18b-44fc-b738-ee5073c65319
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/49ba5517891ceace_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
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: 7e9f1933-adfb-49de-a9cf-c22cd95696e5
wandb_project: Gradients-On-Six
wandb_run: your_name
wandb_runid: 7e9f1933-adfb-49de-a9cf-c22cd95696e5
warmup_steps: 10
weight_decay: 0.01
xformers_attention: null

a64fb4ac-f18b-44fc-b738-ee5073c65319

This model is a fine-tuned version of katuni4ka/tiny-random-olmo-hf on the None dataset. It achieves the following results on the evaluation set:

• Loss: 10.8295

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	0.0002	1	10.8399
10.8384	0.0019	9	10.8393
10.8374	0.0039	18	10.8378
10.8376	0.0058	27	10.8362
10.8406	0.0078	36	10.8346
10.8333	0.0097	45	10.8332
10.8345	0.0117	54	10.8319
10.8291	0.0136	63	10.8309
10.8266	0.0156	72	10.8301
10.8289	0.0175	81	10.8297
10.8253	0.0194	90	10.8296
10.83	0.0214	99	10.8295

Framework versions

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