File size: 1,382 Bytes
da42161
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
11d68fe
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
da42161
 
 
1
2
3
4
5
6
7
8
9
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
25
26
27
28
29
30
31
32
33
34
35
36
37
38
39
40
41
42
43
44
45
46
47
48
49
50
51
---
base_model: Onuii/DAMI-base-merge-0408
tags:
- text-generation-inference
- transformers
- unsloth
- llama
- trl
license: apache-2.0
language:
- en
---

# Uploaded  model

- **Developed by:** Onuii
- **License:** apache-2.0
- **Finetuned from model :** Onuii/DAMI-base-merge-0408

    --run_name "DAMI-lora-sft-digging-0409" \
    --model_name "Onuii/DAMI-base-merge-0408" \
    --dataset_name "Onuii/DAMI-lecture-DB-Dataset-0408-digging" \
    --max_seq_length 8192 \
    --dtype "bfloat16" \
    --load_in_4bit False \
    --r 16 \
    --lora_alpha 32 \
    --lora_dropout 0.1 \
    --use_rslora True \
    --batch_size 16 \
    --gradient_accumulation_steps 2 \
    --learning_rate 1e-5 \
    --warmup_ratio 0.05 \
    --num_train_epochs 2 \
    --optimizer "adamw_8bit" \
    --weight_decay 0.01 \
    --lr_scheduler_type "linear" \
    --output_dir "../outputs" \
    --logging_dir "../logs" \
    --logging_steps 1 \
    --save_steps 200 \
    --save_total_limit 2 \
    --report_to "wandb" \
    --use_gradient_checkpointing False \
    --save_strategy "steps" \
    --random_seed 562

This llama model was trained 2x faster with [Unsloth](https://github.com/unslothai/unsloth) and Huggingface's TRL library.

[<img src="https://raw.githubusercontent.com/unslothai/unsloth/main/images/unsloth%20made%20with%20love.png" width="200"/>](https://github.com/unslothai/unsloth)