See axolotl config

axolotl version: 0.4.1

adapter: lora
base_model: unsloth/mistral-7b-instruct-v0.2
bf16: true
chat_template: llama3
dataset_prepared_path: null
datasets:
- data_files:
  - 251398662c103075_train_data.json
  ds_type: json
  format: custom
  path: /workspace/input_data/251398662c103075_train_data.json
  type:
    field_input: Company Name
    field_instruction: Position
    field_output: Long Description
    format: '{instruction} {input}'
    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: true
fp16: null
fsdp: null
fsdp_config: null
gradient_accumulation_steps: 8
gradient_checkpointing: true
group_by_length: false
hub_model_id: Romain-XV/8aa54993-e0e8-449e-ab3b-34de6da402bc
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: 840
micro_batch_size: 4
mlflow_experiment_name: /tmp/251398662c103075_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
strict: false
tf32: true
tokenizer_type: AutoTokenizer
train_on_inputs: false
trust_remote_code: true
use_rslora: true
val_set_size: 0.03536142916752123
wandb_entity: null
wandb_mode: online
wandb_name: d3902dd9-ea96-4286-a166-f9b0d4a1cca4
wandb_project: Gradients-On-Demand
wandb_run: your_name
wandb_runid: d3902dd9-ea96-4286-a166-f9b0d4a1cca4
warmup_steps: 10
weight_decay: 0.0
xformers_attention: null

8aa54993-e0e8-449e-ab3b-34de6da402bc

This model is a fine-tuned version of unsloth/mistral-7b-instruct-v0.2 on the None dataset. It achieves the following results on the evaluation set:

• Loss: 1.7108

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: 840

Training results

Training Loss	Epoch	Step	Validation Loss
17.5256	0.0002	1	2.2722
14.3098	0.0235	100	1.9101
14.5653	0.0469	200	1.8682
14.6911	0.0704	300	1.8234
14.2559	0.0938	400	1.7895
15.0114	0.1173	500	1.7570
13.5731	0.1408	600	1.7312
14.1723	0.1642	700	1.7160
14.1182	0.1877	800	1.7108

Framework versions

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