Downloads · 30 days
3
10% of all-time downloads
hw-infi-qlora-10-epoch-16384-context
bshada/hw-infi-qlora-10-epoch-16384-context
hw-infi-qlora-10-epoch-16384-context is a machine learning model from bshada. Use it for the machine learning task on the model card, and read the license before you ship it in a product. It is set up for peft. The card lists the license as apache-2.0.
should probably proofread and complete it, then remove this comment. --
All-time downloads
29
Public
Repo size
1.5 GB
Likes
0
Public
Repo makeup
Click a slice to open those files.
.pt675 MB · 44%
At a glance
- Library
- peft
- License
- apache-2.0
- Model type
- mistral
- Access
- Public
- Created
- Aug 11, 2024
- Updated
- Aug 11, 2024
- SHA
- 25335212
Base models
Model card
From the Hugging Face model README
<!-- This model card has been generated automatically according to the information the Trainer had access to. You
should probably proofread and complete it, then remove this comment. -->
<details><summary>See axolotl config</summary>
axolotl version: 0.4.1
base_model: mistralai/Mistral-7B-Instruct-v0.3
model_type: MistralForCausalLM
tokenizer_type: LlamaTokenizer
load_in_8bit: false
load_in_4bit: true
strict: false
datasets:
- path: bshada/hw-infi
type: alpaca
dataset_prepared_path: last_run_prepared
val_set_size: 0.1
output_dir: ./adapters/hw-infi-qlora-10-epoch-16384-contex
adapter: qlora
lora_model_dir:
sequence_len: 16384
sample_packing: false
pad_to_sequence_len: true
lora_r: 32
lora_alpha: 16
lora_dropout: 0.05
lora_target_linear: true
lora_fan_in_fan_out:
lora_target_modules:
- gate_proj
- down_proj
- up_proj
- q_proj
- v_proj
- k_proj
- o_proj
wandb_project:
wandb_entity:
wandb_watch:
wandb_name:
wandb_log_model:
gradient_accumulation_steps: 1
micro_batch_size: 1
num_epochs: 10
optimizer: adamw_bnb_8bit
lr_scheduler: cosine
learning_rate: 0.0002
train_on_inputs: false
group_by_length: false
bf16: auto
fp16:
tf32: false
gradient_checkpointing: true
early_stopping_patience:
resume_from_checkpoint:
local_rank:
logging_steps: 1
xformers_attention:
flash_attention: true
loss_watchdog_threshold: 5.0
loss_watchdog_patience: 3
warmup_steps: 0
evals_per_epoch: 4
eval_table_size:
eval_max_new_tokens: 128
saves_per_epoch: 1
debug:
deepspeed:
weight_decay: 0.0
fsdp:
fsdp_config:
special_tokens:
adapters/hw-infi-qlora-10-epoch-16384-contex
This model is a fine-tuned version of mistralai/Mistral-7B-Instruct-v0.3 on the None dataset. It achieves the following results on the evaluation set:
- Loss: 0.2161
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: 1
- eval_batch_size: 1
- seed: 42
- optimizer: Adam with betas=(0.9,0.999) and epsilon=1e-08
- lr_scheduler_type: cosine
- num_epochs: 10
Training results
| Training Loss | Epoch | Step | Validation Loss |
|---|---|---|---|
| 0.2155 | 0.5 | 1 | 0.2359 |
| 0.2104 | 1.0 | 2 | 0.2150 |
| 0.1625 | 1.5 | 3 | 0.1978 |
| 0.1496 | 2.0 | 4 | 0.1850 |
| 0.1264 | 2.5 | 5 | 0.1786 |
| 0.11 | 3.0 | 6 | 0.1812 |
| 0.091 | 3.5 | 7 | 0.1830 |
| 0.1007 | 4.0 | 8 | 0.1835 |
| 0.0623 | 4.5 | 9 | 0.1880 |
| 0.082 | 5.0 | 10 | 0.1919 |
| 0.0405 | 5.5 | 11 | 0.1971 |
| 0.0636 | 6.0 | 12 | 0.2002 |
| 0.054 | 6.5 | 13 | 0.2056 |
| 0.0293 | 7.0 | 14 | 0.2076 |
| 0.0406 | 7.5 | 15 | 0.2110 |
| 0.0192 | 8.0 | 16 | 0.2140 |
| 0.034 | 8.5 | 17 | 0.2148 |
| 0.0153 | 9.0 | 18 | 0.2158 |
| 0.0314 | 9.5 | 19 | 0.2161 |
| 0.0142 | 10.0 | 20 | 0.2161 |
Framework versions
- PEFT 0.11.1
- Transformers 4.42.3
- Pytorch 2.3.1+cu121
- Datasets 2.19.1
- Tokenizers 0.19.1
- Library
- peft
- Type
- mistral
- License
- apache-2.0
- Created
- Aug 11, 2024
- Updated
- Aug 11, 2024