Delete README.md
Browse files
README.md
DELETED
@@ -1,168 +0,0 @@
|
|
1 |
-
---
|
2 |
-
license: other
|
3 |
-
base_model: stabilityai/stablelm-2-1_6b
|
4 |
-
tags:
|
5 |
-
- generated_from_trainer
|
6 |
-
model-index:
|
7 |
-
- name: stablelm_1-6b_ContextSplitter
|
8 |
-
results: []
|
9 |
-
---
|
10 |
-
|
11 |
-
<!-- This model card has been generated automatically according to the information the Trainer had access to. You
|
12 |
-
should probably proofread and complete it, then remove this comment. -->
|
13 |
-
|
14 |
-
[<img src="https://raw.githubusercontent.com/OpenAccess-AI-Collective/axolotl/main/image/axolotl-badge-web.png" alt="Built with Axolotl" width="200" height="32"/>](https://github.com/OpenAccess-AI-Collective/axolotl)
|
15 |
-
<details><summary>See axolotl config</summary>
|
16 |
-
|
17 |
-
axolotl version: `0.4.0`
|
18 |
-
```yaml
|
19 |
-
base_model: stabilityai/stablelm-2-1_6b
|
20 |
-
base_model_config: stabilityai/stablelm-2-1_6b
|
21 |
-
model_type: StableLMEpochForCausalLM
|
22 |
-
tokenizer_type: AutoTokenizer
|
23 |
-
|
24 |
-
trust_remote_code: true
|
25 |
-
|
26 |
-
load_in_8bit: false
|
27 |
-
load_in_4bit: false
|
28 |
-
strict: false
|
29 |
-
|
30 |
-
datasets:
|
31 |
-
- path: /run/media/username/Storage/datasets/repo/alpaca/context-aware-splits-english_new.json
|
32 |
-
type: alpaca
|
33 |
-
|
34 |
-
dataset_prepared_path: stablelm_1-6b_ContextSplitter_data
|
35 |
-
val_set_size: 0.02
|
36 |
-
output_dir: ./stablelm_1-6b_ContextSplitter
|
37 |
-
|
38 |
-
sequence_len: 4096
|
39 |
-
sample_packing: true
|
40 |
-
pad_to_sequence_len: true
|
41 |
-
|
42 |
-
adapter:
|
43 |
-
lora_model_dir:
|
44 |
-
lora_r:
|
45 |
-
lora_alpha:
|
46 |
-
lora_dropout:
|
47 |
-
lora_target_linear:
|
48 |
-
lora_fan_in_fan_out:
|
49 |
-
|
50 |
-
wandb_project: stablelm_1-6b_ContextSplitter
|
51 |
-
wandb_entity:
|
52 |
-
wandb_watch:
|
53 |
-
wandb_name:
|
54 |
-
wandb_log_model:
|
55 |
-
|
56 |
-
gradient_accumulation_steps: 1
|
57 |
-
micro_batch_size: 1
|
58 |
-
num_epochs: 1
|
59 |
-
optimizer: paged_adamw_32bit
|
60 |
-
lr_scheduler: cosine
|
61 |
-
learning_rate: 0.00001
|
62 |
-
|
63 |
-
train_on_inputs: false
|
64 |
-
group_by_length: false
|
65 |
-
bf16: true
|
66 |
-
fp16: false
|
67 |
-
tf32: false
|
68 |
-
|
69 |
-
gradient_checkpointing: true
|
70 |
-
early_stopping_patience:
|
71 |
-
resume_from_checkpoint:
|
72 |
-
local_rank:
|
73 |
-
logging_steps: 1
|
74 |
-
xformers_attention:
|
75 |
-
flash_attention: true
|
76 |
-
flash_attn_cross_entropy: false
|
77 |
-
flash_attn_rms_norm: true
|
78 |
-
flash_attn_fuse_qkv: false
|
79 |
-
flash_attn_fuse_mlp: true
|
80 |
-
|
81 |
-
warmup_steps: 100
|
82 |
-
evals_per_epoch: 30
|
83 |
-
eval_table_size:
|
84 |
-
saves_per_epoch: 4
|
85 |
-
debug:
|
86 |
-
deepspeed: #deepspeed_configs/zero2.json # multi-gpu only
|
87 |
-
weight_decay: 0.1
|
88 |
-
fsdp:
|
89 |
-
fsdp_config:
|
90 |
-
special_tokens:
|
91 |
-
```
|
92 |
-
|
93 |
-
</details><br>
|
94 |
-
|
95 |
-
# stablelm_1-6b_ContextSplitter
|
96 |
-
|
97 |
-
This model is a fine-tuned version of [stabilityai/stablelm-2-1_6b](https://huggingface.co/stabilityai/stablelm-2-1_6b) on the None dataset.
|
98 |
-
It achieves the following results on the evaluation set:
|
99 |
-
- Loss: 0.0377
|
100 |
-
|
101 |
-
## Model description
|
102 |
-
|
103 |
-
More information needed
|
104 |
-
|
105 |
-
## Intended uses & limitations
|
106 |
-
|
107 |
-
More information needed
|
108 |
-
|
109 |
-
## Training and evaluation data
|
110 |
-
|
111 |
-
More information needed
|
112 |
-
|
113 |
-
## Training procedure
|
114 |
-
|
115 |
-
### Training hyperparameters
|
116 |
-
|
117 |
-
The following hyperparameters were used during training:
|
118 |
-
- learning_rate: 1e-05
|
119 |
-
- train_batch_size: 1
|
120 |
-
- eval_batch_size: 1
|
121 |
-
- seed: 42
|
122 |
-
- optimizer: Adam with betas=(0.9,0.999) and epsilon=1e-08
|
123 |
-
- lr_scheduler_type: cosine
|
124 |
-
- lr_scheduler_warmup_steps: 100
|
125 |
-
- num_epochs: 1
|
126 |
-
|
127 |
-
### Training results
|
128 |
-
|
129 |
-
| Training Loss | Epoch | Step | Validation Loss |
|
130 |
-
|:-------------:|:-----:|:----:|:---------------:|
|
131 |
-
| 0.1781 | 0.0 | 1 | 0.2283 |
|
132 |
-
| 0.0709 | 0.03 | 248 | 0.0589 |
|
133 |
-
| 0.0274 | 0.07 | 496 | 0.0512 |
|
134 |
-
| 0.0614 | 0.1 | 744 | 0.0480 |
|
135 |
-
| 0.0266 | 0.13 | 992 | 0.0466 |
|
136 |
-
| 0.0471 | 0.17 | 1240 | 0.0440 |
|
137 |
-
| 0.0425 | 0.2 | 1488 | 0.0435 |
|
138 |
-
| 0.1172 | 0.23 | 1736 | 0.0423 |
|
139 |
-
| 0.0322 | 0.27 | 1984 | 0.0415 |
|
140 |
-
| 0.0529 | 0.3 | 2232 | 0.0413 |
|
141 |
-
| 0.0296 | 0.33 | 2480 | 0.0409 |
|
142 |
-
| 0.0357 | 0.37 | 2728 | 0.0398 |
|
143 |
-
| 0.0242 | 0.4 | 2976 | 0.0394 |
|
144 |
-
| 0.0266 | 0.43 | 3224 | 0.0391 |
|
145 |
-
| 0.0292 | 0.47 | 3472 | 0.0386 |
|
146 |
-
| 0.0261 | 0.5 | 3720 | 0.0386 |
|
147 |
-
| 0.0382 | 0.53 | 3968 | 0.0383 |
|
148 |
-
| 0.0378 | 0.57 | 4216 | 0.0383 |
|
149 |
-
| 0.0345 | 0.6 | 4464 | 0.0379 |
|
150 |
-
| 0.0467 | 0.64 | 4712 | 0.0379 |
|
151 |
-
| 0.0542 | 0.67 | 4960 | 0.0378 |
|
152 |
-
| 0.0317 | 0.7 | 5208 | 0.0378 |
|
153 |
-
| 0.0363 | 0.74 | 5456 | 0.0377 |
|
154 |
-
| 0.054 | 0.77 | 5704 | 0.0377 |
|
155 |
-
| 0.0207 | 0.8 | 5952 | 0.0377 |
|
156 |
-
| 0.0302 | 0.84 | 6200 | 0.0377 |
|
157 |
-
| 0.0427 | 0.87 | 6448 | 0.0377 |
|
158 |
-
| 0.0278 | 0.9 | 6696 | 0.0377 |
|
159 |
-
| 0.0648 | 0.94 | 6944 | 0.0377 |
|
160 |
-
| 0.0497 | 0.97 | 7192 | 0.0377 |
|
161 |
-
|
162 |
-
|
163 |
-
### Framework versions
|
164 |
-
|
165 |
-
- Transformers 4.38.0.dev0
|
166 |
-
- Pytorch 2.0.1+cu117
|
167 |
-
- Datasets 2.15.0
|
168 |
-
- Tokenizers 0.15.0
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|