Model save
Browse files- README.md +59 -0
- tokenizer.json +1 -6
- trainer_state.json +182 -0
README.md
ADDED
@@ -0,0 +1,59 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
1 |
+
---
|
2 |
+
license: apache-2.0
|
3 |
+
library_name: peft
|
4 |
+
tags:
|
5 |
+
- trl
|
6 |
+
- sft
|
7 |
+
- generated_from_trainer
|
8 |
+
base_model: PY007/TinyLlama-1.1B-step-50K-105b
|
9 |
+
model-index:
|
10 |
+
- name: tinyllama
|
11 |
+
results: []
|
12 |
+
---
|
13 |
+
|
14 |
+
<!-- This model card has been generated automatically according to the information the Trainer had access to. You
|
15 |
+
should probably proofread and complete it, then remove this comment. -->
|
16 |
+
|
17 |
+
# tinyllama
|
18 |
+
|
19 |
+
This model is a fine-tuned version of [PY007/TinyLlama-1.1B-step-50K-105b](https://huggingface.co/PY007/TinyLlama-1.1B-step-50K-105b) on the None dataset.
|
20 |
+
|
21 |
+
## Model description
|
22 |
+
|
23 |
+
More information needed
|
24 |
+
|
25 |
+
## Intended uses & limitations
|
26 |
+
|
27 |
+
More information needed
|
28 |
+
|
29 |
+
## Training and evaluation data
|
30 |
+
|
31 |
+
More information needed
|
32 |
+
|
33 |
+
## Training procedure
|
34 |
+
|
35 |
+
### Training hyperparameters
|
36 |
+
|
37 |
+
The following hyperparameters were used during training:
|
38 |
+
- learning_rate: 0.002
|
39 |
+
- train_batch_size: 3
|
40 |
+
- eval_batch_size: 8
|
41 |
+
- seed: 42
|
42 |
+
- gradient_accumulation_steps: 2
|
43 |
+
- total_train_batch_size: 6
|
44 |
+
- optimizer: Adam with betas=(0.9,0.999) and epsilon=1e-08
|
45 |
+
- lr_scheduler_type: cosine
|
46 |
+
- lr_scheduler_warmup_ratio: 0.03
|
47 |
+
- training_steps: 200
|
48 |
+
|
49 |
+
### Training results
|
50 |
+
|
51 |
+
|
52 |
+
|
53 |
+
### Framework versions
|
54 |
+
|
55 |
+
- PEFT 0.11.2.dev0
|
56 |
+
- Transformers 4.41.2
|
57 |
+
- Pytorch 2.3.0+cu121
|
58 |
+
- Datasets 2.19.1
|
59 |
+
- Tokenizers 0.19.1
|
tokenizer.json
CHANGED
@@ -1,11 +1,6 @@
|
|
1 |
{
|
2 |
"version": "1.0",
|
3 |
-
"truncation":
|
4 |
-
"direction": "Right",
|
5 |
-
"max_length": 500,
|
6 |
-
"strategy": "LongestFirst",
|
7 |
-
"stride": 0
|
8 |
-
},
|
9 |
"padding": null,
|
10 |
"added_tokens": [
|
11 |
{
|
|
|
1 |
{
|
2 |
"version": "1.0",
|
3 |
+
"truncation": null,
|
|
|
|
|
|
|
|
|
|
|
4 |
"padding": null,
|
5 |
"added_tokens": [
|
6 |
{
|
trainer_state.json
ADDED
@@ -0,0 +1,182 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
1 |
+
{
|
2 |
+
"best_metric": null,
|
3 |
+
"best_model_checkpoint": null,
|
4 |
+
"epoch": 2.150537634408602,
|
5 |
+
"eval_steps": 500,
|
6 |
+
"global_step": 200,
|
7 |
+
"is_hyper_param_search": false,
|
8 |
+
"is_local_process_zero": true,
|
9 |
+
"is_world_process_zero": true,
|
10 |
+
"log_history": [
|
11 |
+
{
|
12 |
+
"epoch": 0.10752688172043011,
|
13 |
+
"grad_norm": 0.35747238993644714,
|
14 |
+
"learning_rate": 0.0019979028262377117,
|
15 |
+
"loss": 2.7578,
|
16 |
+
"step": 10
|
17 |
+
},
|
18 |
+
{
|
19 |
+
"epoch": 0.21505376344086022,
|
20 |
+
"grad_norm": 0.31179729104042053,
|
21 |
+
"learning_rate": 0.001974410524646926,
|
22 |
+
"loss": 2.6817,
|
23 |
+
"step": 20
|
24 |
+
},
|
25 |
+
{
|
26 |
+
"epoch": 0.3225806451612903,
|
27 |
+
"grad_norm": 0.29704058170318604,
|
28 |
+
"learning_rate": 0.0019254212296427042,
|
29 |
+
"loss": 2.472,
|
30 |
+
"step": 30
|
31 |
+
},
|
32 |
+
{
|
33 |
+
"epoch": 0.43010752688172044,
|
34 |
+
"grad_norm": 0.31029245257377625,
|
35 |
+
"learning_rate": 0.0018522168236559692,
|
36 |
+
"loss": 2.606,
|
37 |
+
"step": 40
|
38 |
+
},
|
39 |
+
{
|
40 |
+
"epoch": 0.5376344086021505,
|
41 |
+
"grad_norm": 0.2981964945793152,
|
42 |
+
"learning_rate": 0.0017567128158176952,
|
43 |
+
"loss": 2.5843,
|
44 |
+
"step": 50
|
45 |
+
},
|
46 |
+
{
|
47 |
+
"epoch": 0.6451612903225806,
|
48 |
+
"grad_norm": 0.3093157708644867,
|
49 |
+
"learning_rate": 0.00164140821963114,
|
50 |
+
"loss": 2.4672,
|
51 |
+
"step": 60
|
52 |
+
},
|
53 |
+
{
|
54 |
+
"epoch": 0.7526881720430108,
|
55 |
+
"grad_norm": 0.2779924273490906,
|
56 |
+
"learning_rate": 0.001509320162328763,
|
57 |
+
"loss": 2.4794,
|
58 |
+
"step": 70
|
59 |
+
},
|
60 |
+
{
|
61 |
+
"epoch": 0.8602150537634409,
|
62 |
+
"grad_norm": 0.2921682894229889,
|
63 |
+
"learning_rate": 0.0013639049369634877,
|
64 |
+
"loss": 2.5483,
|
65 |
+
"step": 80
|
66 |
+
},
|
67 |
+
{
|
68 |
+
"epoch": 0.967741935483871,
|
69 |
+
"grad_norm": 0.2675682008266449,
|
70 |
+
"learning_rate": 0.0012089675630312753,
|
71 |
+
"loss": 2.5437,
|
72 |
+
"step": 90
|
73 |
+
},
|
74 |
+
{
|
75 |
+
"epoch": 1.075268817204301,
|
76 |
+
"grad_norm": 0.3106231093406677,
|
77 |
+
"learning_rate": 0.0010485622221144484,
|
78 |
+
"loss": 2.5045,
|
79 |
+
"step": 100
|
80 |
+
},
|
81 |
+
{
|
82 |
+
"epoch": 1.1827956989247312,
|
83 |
+
"grad_norm": 0.3767760992050171,
|
84 |
+
"learning_rate": 0.0008868861738047158,
|
85 |
+
"loss": 2.3468,
|
86 |
+
"step": 110
|
87 |
+
},
|
88 |
+
{
|
89 |
+
"epoch": 1.2903225806451613,
|
90 |
+
"grad_norm": 0.382185161113739,
|
91 |
+
"learning_rate": 0.0007281699277636571,
|
92 |
+
"loss": 2.2433,
|
93 |
+
"step": 120
|
94 |
+
},
|
95 |
+
{
|
96 |
+
"epoch": 1.3978494623655915,
|
97 |
+
"grad_norm": 0.44693300127983093,
|
98 |
+
"learning_rate": 0.0005765665457425102,
|
99 |
+
"loss": 2.1934,
|
100 |
+
"step": 130
|
101 |
+
},
|
102 |
+
{
|
103 |
+
"epoch": 1.5053763440860215,
|
104 |
+
"grad_norm": 0.37442219257354736,
|
105 |
+
"learning_rate": 0.0004360429701490934,
|
106 |
+
"loss": 2.2445,
|
107 |
+
"step": 140
|
108 |
+
},
|
109 |
+
{
|
110 |
+
"epoch": 1.6129032258064515,
|
111 |
+
"grad_norm": 0.36704185605049133,
|
112 |
+
"learning_rate": 0.00031027622272189573,
|
113 |
+
"loss": 2.2128,
|
114 |
+
"step": 150
|
115 |
+
},
|
116 |
+
{
|
117 |
+
"epoch": 1.7204301075268817,
|
118 |
+
"grad_norm": 0.3427051901817322,
|
119 |
+
"learning_rate": 0.0002025571894372794,
|
120 |
+
"loss": 2.2675,
|
121 |
+
"step": 160
|
122 |
+
},
|
123 |
+
{
|
124 |
+
"epoch": 1.827956989247312,
|
125 |
+
"grad_norm": 0.3787896931171417,
|
126 |
+
"learning_rate": 0.00011570450926997656,
|
127 |
+
"loss": 2.2544,
|
128 |
+
"step": 170
|
129 |
+
},
|
130 |
+
{
|
131 |
+
"epoch": 1.935483870967742,
|
132 |
+
"grad_norm": 0.3463006317615509,
|
133 |
+
"learning_rate": 5.199082004372957e-05,
|
134 |
+
"loss": 2.1877,
|
135 |
+
"step": 180
|
136 |
+
},
|
137 |
+
{
|
138 |
+
"epoch": 2.043010752688172,
|
139 |
+
"grad_norm": 0.33030927181243896,
|
140 |
+
"learning_rate": 1.3083291266109298e-05,
|
141 |
+
"loss": 2.2705,
|
142 |
+
"step": 190
|
143 |
+
},
|
144 |
+
{
|
145 |
+
"epoch": 2.150537634408602,
|
146 |
+
"grad_norm": 0.33233529329299927,
|
147 |
+
"learning_rate": 0.0,
|
148 |
+
"loss": 2.0534,
|
149 |
+
"step": 200
|
150 |
+
},
|
151 |
+
{
|
152 |
+
"epoch": 2.150537634408602,
|
153 |
+
"step": 200,
|
154 |
+
"total_flos": 1207440721170432.0,
|
155 |
+
"train_loss": 2.3959710121154787,
|
156 |
+
"train_runtime": 104.1427,
|
157 |
+
"train_samples_per_second": 11.523,
|
158 |
+
"train_steps_per_second": 1.92
|
159 |
+
}
|
160 |
+
],
|
161 |
+
"logging_steps": 10,
|
162 |
+
"max_steps": 200,
|
163 |
+
"num_input_tokens_seen": 0,
|
164 |
+
"num_train_epochs": 3,
|
165 |
+
"save_steps": 10,
|
166 |
+
"stateful_callbacks": {
|
167 |
+
"TrainerControl": {
|
168 |
+
"args": {
|
169 |
+
"should_epoch_stop": false,
|
170 |
+
"should_evaluate": false,
|
171 |
+
"should_log": false,
|
172 |
+
"should_save": true,
|
173 |
+
"should_training_stop": true
|
174 |
+
},
|
175 |
+
"attributes": {}
|
176 |
+
}
|
177 |
+
},
|
178 |
+
"total_flos": 1207440721170432.0,
|
179 |
+
"train_batch_size": 3,
|
180 |
+
"trial_name": null,
|
181 |
+
"trial_params": null
|
182 |
+
}
|