axolotl version: 0.5.2
base_model: mistralai/Mistral-7B-v0.1
model_type: AutoModelForCausalLM
tokenizer_config: Open-Orca/Mistral-7B-OpenOrca
tokenizer_type: AutoTokenizer
tokenizer_use_fast: false
resize_token_embeddings_to_32x: false
plugins:
- axolotl.integrations.liger.LigerPlugin
liger_rope: true
liger_rms_norm: true
liger_glu_activation: true
liger_layer_norm: true
liger_fused_linear_cross_entropy: true
flash_attention: true
xformers_attention:
load_in_8bit: false
load_in_4bit: false
strict: false
chat_template: chatml
datasets:
- path: skymizer/Sonnet3.5-SlimOrcaDedupCleaned-train
type: chat_template
field_messages: messages
test_datasets:
- path: skymizer/Sonnet3.5-SlimOrcaDedupCleaned-test
type: chat_template
field_messages: messages
split: train
hf_use_auth_token: true
dataset_prepared_path: pretokenized/slim-orca
output_dir: ./exp_output_artifacts
sequence_len: 2048
sample_packing: true
pad_to_sequence_len: true
eval_sample_packing: false
# eval_causal_lm_metrics: ["perplexity"]
wandb_project: "axolotl_mistral_sft"
wandb_entity:
wandb_watch:
wandb_name: "mistral-7B-v0.1-sft-slim-orca-sonnet-3.5-v4"
wandb_log_model:
gradient_accumulation_steps: 2
micro_batch_size: 16
eval_batch_size: 1
num_epochs: 1
optimizer: adamw_torch
lr_scheduler: cosine
learning_rate: 0.000003
weight_decay: 0.0
adam_beta1: 0.9
adam_beta2: 0.95
adam_eps: 0.000001
max_grad_norm: 1.0
train_on_inputs: false
group_by_length: false
bf16: true
fp16:
tf32: false
hub_model_id: "skymizer/mistral-7B-v0.1-sft-slim-orca-sonnet-3.5-v4"
save_strategy: "epoch"
saves_per_epoch: 1
gradient_checkpointing: true
gradient_checkpointing_kwargs:
use_reentrant: false
early_stopping_patience:
resume_from_checkpoint:
local_rank:
logging_steps: 1
warmup_ratio: 0.03
eval_steps: 50
eval_table_size:
eval_max_new_tokens: 2048
debug:
deepspeed: deepspeed_configs/zero3_bf16.json
fsdp:
fsdp_config:
seed: 42
mistral-7B-v0.1-sft-slim-orca-sonnet-3.5-v4
This model is a fine-tuned version of mistralai/Mistral-7B-v0.1 on the None dataset. It achieves the following results on the evaluation set:
- Loss: 0.5315
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: 3e-06
- train_batch_size: 16
- eval_batch_size: 1
- seed: 42
- distributed_type: multi-GPU
- num_devices: 4
- gradient_accumulation_steps: 2
- total_train_batch_size: 128
- total_eval_batch_size: 4
- optimizer: Use OptimizerNames.ADAMW_TORCH with betas=(0.9,0.95) and epsilon=1e-08 and optimizer_args=No additional optimizer arguments
- lr_scheduler_type: cosine
- lr_scheduler_warmup_steps: 11
- num_epochs: 1
Training results
| Training Loss | Epoch | Step | Validation Loss |
|---|---|---|---|
| 1.2346 | 0.0026 | 1 | 1.3762 |
| 0.6186 | 0.1277 | 50 | 0.6218 |
| 0.5905 | 0.2554 | 100 | 0.5852 |
| 0.5684 | 0.3831 | 150 | 0.5648 |
| 0.5621 | 0.5109 | 200 | 0.5513 |
| 0.551 | 0.6386 | 250 | 0.5415 |
| 0.5489 | 0.7663 | 300 | 0.5345 |
| 0.5434 | 0.8940 | 350 | 0.5315 |
Framework versions
- Transformers 4.46.3
- Pytorch 2.5.1+cu124
- Datasets 3.1.0
- Tokenizers 0.20.3