-
Notifications
You must be signed in to change notification settings - Fork 50
/
1.5B_full_single_device.yaml
81 lines (68 loc) · 2.11 KB
/
1.5B_full_single_device.yaml
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
52
53
54
55
56
57
58
59
60
61
62
63
64
65
66
67
68
69
70
71
72
73
74
75
76
77
78
79
80
81
# Config for single device full finetuning in full_finetune_single_device.py
# using a Qwen2 1.5B
#
# This config assumes that you've run the following command before launching
# this run:
# tune download Qwen/Qwen2-1.5B-Instruct --output-dir /tmp/Qwen2-1.5B-Instruct --ignore-patterns None
#
# The default config uses an optimizer from bitsandbytes. If you do not have it installed,
# you can install it with
# pip install bitsandbytes
#
# To launch on a single device, run the following command from root:
# tune run full_finetune_single_device --config qwen2/1.5B_full_single_device
#
# You can add specific overrides through the command line. For example
# to override the checkpointer directory while launching training
# you can run:
# tune run full_finetune_single_device --config qwen2/1.5B_full_single_device checkpointer.checkpoint_dir=<YOUR_CHECKPOINT_DIR>
#
# This config works only for training on single device.
# Tokenizer
tokenizer:
_component_: torchtune.models.qwen2.qwen2_tokenizer
path: /tmp/Qwen2-1.5B-Instruct/vocab.json
merges_file: /tmp/Qwen2-1.5B-Instruct/merges.txt
# Dataset
dataset:
_component_: torchtune.datasets.alpaca_cleaned_dataset
seed: null
shuffle: True
# Model Arguments
model:
_component_: torchtune.models.qwen2.qwen2_1_5b
checkpointer:
_component_: torchtune.utils.FullModelHFCheckpointer
checkpoint_dir: /tmp/Qwen2-1.5B-Instruct
checkpoint_files: [
model.safetensors
]
recipe_checkpoint: null
output_dir: /tmp/Qwen2-1.5B-Instruct-finetune
model_type: QWEN2
resume_from_checkpoint: False
# Fine-tuning arguments
batch_size: 2
epochs: 1
optimizer:
_component_: bitsandbytes.optim.PagedAdamW
lr: 2e-5
optimizer_in_bwd: True
loss:
_component_: torch.nn.CrossEntropyLoss
max_steps_per_epoch: null
gradient_accumulation_steps: 1
compile: False
# Training environment
device: cuda
# Memory management
enable_activation_checkpointing: True
# Reduced precision
dtype: bf16
# Logging
metric_logger:
_component_: torchtune.utils.metric_logging.DiskLogger
log_dir: ${output_dir}
output_dir: /tmp/Qwen2-1.5B-Instruct-finetune
log_every_n_steps: 1
log_peak_memory_stats: False