|
--- |
|
library_name: transformers |
|
license: apache-2.0 |
|
base_model: mistralai/Mistral-7B-v0.1 |
|
tags: |
|
- axolotl |
|
- generated_from_trainer |
|
datasets: |
|
- AiAF/pretraining.jsonl |
|
model-index: |
|
- name: UFOs-Pretraining-V1.1 |
|
results: [] |
|
--- |
|
|
|
<!-- 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. --> |
|
|
|
[<img src="https://raw.githubusercontent.com/axolotl-ai-cloud/axolotl/main/image/axolotl-badge-web.png" alt="Built with Axolotl" width="200" height="32"/>](https://github.com/axolotl-ai-cloud/axolotl) |
|
<details><summary>See axolotl config</summary> |
|
|
|
axolotl version: `0.6.0` |
|
```yaml |
|
base_model: mistralai/Mistral-7B-v0.1 |
|
# optionally might have model_type or tokenizer_type |
|
model_type: MistralForCausalLM |
|
tokenizer_type: LlamaTokenizer |
|
# Automatically upload checkpoint and final model to HF |
|
hub_model_id: AiAF/UFOs-Pretraining-V1.1 |
|
|
|
load_in_8bit: false |
|
load_in_4bit: false |
|
strict: false |
|
|
|
datasets: |
|
- path: AiAF/pretraining.jsonl |
|
type: completion |
|
|
|
dataset_prepared_path: |
|
val_set_size: 0.05 |
|
output_dir: ./outputs/out/v1.1 |
|
|
|
sequence_len: 8192 |
|
sample_packing: true |
|
pad_to_sequence_len: true |
|
eval_sample_packing: false |
|
|
|
max_steps: 100000 |
|
|
|
wandb_project: "UFO_LLM_Pretraining" |
|
wandb_entity: |
|
wandb_watch: "all" |
|
wandb_name: "UFO_LLM_Pretraining-V1.1" |
|
wandb_log_model: "false" |
|
|
|
gradient_accumulation_steps: 4 |
|
micro_batch_size: 2 |
|
num_epochs: 10 |
|
optimizer: adamw_bnb_8bit |
|
lr_scheduler: cosine |
|
learning_rate: 0.000005 |
|
|
|
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 |
|
|
|
warmup_steps: 10 |
|
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: |
|
|
|
|
|
``` |
|
|
|
</details><br> |
|
|
|
# UFOs-Pretraining-V1.1 |
|
|
|
This model is a fine-tuned version of [mistralai/Mistral-7B-v0.1](https://huggingface.co/mistralai/Mistral-7B-v0.1) on the AiAF/pretraining.jsonl dataset. |
|
It achieves the following results on the evaluation set: |
|
- Loss: 1.7822 |
|
|
|
## 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-06 |
|
- train_batch_size: 2 |
|
- eval_batch_size: 2 |
|
- seed: 42 |
|
- gradient_accumulation_steps: 4 |
|
- total_train_batch_size: 8 |
|
- 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: 90 |
|
|
|
### Training results |
|
|
|
| Training Loss | Epoch | Step | Validation Loss | |
|
|:-------------:|:------:|:----:|:---------------:| |
|
| 1.7686 | 0.1111 | 1 | 1.6895 | |
|
| 2.0582 | 0.3333 | 3 | 1.6884 | |
|
| 1.9134 | 0.6667 | 6 | 1.6791 | |
|
| 1.8262 | 1.0 | 9 | 1.6672 | |
|
| 1.875 | 1.3333 | 12 | 1.6578 | |
|
| 1.8751 | 1.6667 | 15 | 1.6501 | |
|
| 1.8375 | 2.0 | 18 | 1.6471 | |
|
| 1.7018 | 2.3333 | 21 | 1.6587 | |
|
| 1.398 | 2.6667 | 24 | 1.6508 | |
|
| 1.6955 | 3.0 | 27 | 1.6577 | |
|
| 1.4222 | 3.3333 | 30 | 1.6812 | |
|
| 1.264 | 3.6667 | 33 | 1.6664 | |
|
| 1.4261 | 4.0 | 36 | 1.6827 | |
|
| 1.2406 | 4.3333 | 39 | 1.7099 | |
|
| 1.2105 | 4.6667 | 42 | 1.7099 | |
|
| 1.3733 | 5.0 | 45 | 1.7162 | |
|
| 1.2441 | 5.3333 | 48 | 1.7490 | |
|
| 1.1755 | 5.6667 | 51 | 1.7440 | |
|
| 1.2253 | 6.0 | 54 | 1.7394 | |
|
| 1.1223 | 6.3333 | 57 | 1.7542 | |
|
| 1.1837 | 6.6667 | 60 | 1.7679 | |
|
| 0.9838 | 7.0 | 63 | 1.7670 | |
|
| 1.1613 | 7.3333 | 66 | 1.7693 | |
|
| 1.1775 | 7.6667 | 69 | 1.7753 | |
|
| 0.8999 | 8.0 | 72 | 1.7796 | |
|
| 1.1617 | 8.3333 | 75 | 1.7813 | |
|
| 1.1119 | 8.6667 | 78 | 1.7819 | |
|
| 1.1191 | 9.0 | 81 | 1.7825 | |
|
| 1.0606 | 9.3333 | 84 | 1.7821 | |
|
| 1.1476 | 9.6667 | 87 | 1.7820 | |
|
| 1.0837 | 10.0 | 90 | 1.7822 | |
|
|
|
|
|
### Framework versions |
|
|
|
- Transformers 4.48.3 |
|
- Pytorch 2.5.1+cu124 |
|
- Datasets 3.2.0 |
|
- Tokenizers 0.21.0 |
|
|
