Hugging Face's logo

khangmacon
/
cyllama3

PEFT
Safetensors
Generated from Trainer
Eval Results
Model card Files Community
khangmacon commited on
Commit
b665019
•
1 Parent(s): f778197

End of training

Browse files
Files changed (7) hide show
  1. README.md +121 -2
  2. adapter_model.safetensors +1 -1
  3. all_results.json +15 -0
  4. eval_results.json +10 -0
  5. pt_lora_model/adapter_config.json +5 -5
  6. train_results.json +8 -0
  7. trainer_state.json +0 -0
README.md CHANGED
@@ -4,9 +4,23 @@ library_name: peft
4
  tags:
5
  - generated_from_trainer
6
  base_model: meta-llama/Meta-Llama-3-8B
 
 
 
 
7
  model-index:
8
  - name: cyllama3
9
- results: []
 
 
 
 
 
 
 
 
 
 
10
  ---
11
 
12
  <!-- This model card has been generated automatically according to the information the Trainer had access to. You
@@ -14,7 +28,10 @@ should probably proofread and complete it, then remove this comment. -->
14
 
15
  # cyllama3
16
 
17
- This model is a fine-tuned version of [meta-llama/Meta-Llama-3-8B](https://huggingface.co/meta-llama/Meta-Llama-3-8B) on an unknown dataset.
 
 
 
18
 
19
  ## Model description
20
 
@@ -47,6 +64,108 @@ The following hyperparameters were used during training:
47
  - lr_scheduler_warmup_ratio: 0.05
48
  - num_epochs: 1.0
49
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
50
  ### Framework versions
51
 
52
  - PEFT 0.10.1.dev0
 
4
  tags:
5
  - generated_from_trainer
6
  base_model: meta-llama/Meta-Llama-3-8B
7
+ datasets:
8
+ - khangmacon/llmtrain
9
+ metrics:
10
+ - accuracy
11
  model-index:
12
  - name: cyllama3
13
+ results:
14
+ - task:
15
+ type: text-generation
16
+ name: Causal Language Modeling
17
+ dataset:
18
+ name: khangmacon/llmtrain
19
+ type: khangmacon/llmtrain
20
+ metrics:
21
+ - type: accuracy
22
+ value: 0.5590444975644216
23
+ name: Accuracy
24
  ---
25
 
26
  <!-- This model card has been generated automatically according to the information the Trainer had access to. You
 
28
 
29
  # cyllama3
30
 
31
+ This model is a fine-tuned version of [meta-llama/Meta-Llama-3-8B](https://huggingface.co/meta-llama/Meta-Llama-3-8B) on the khangmacon/llmtrain dataset.
32
+ It achieves the following results on the evaluation set:
33
+ - Loss: 1.9930
34
+ - Accuracy: 0.5590
35
 
36
  ## Model description
37
 
 
64
  - lr_scheduler_warmup_ratio: 0.05
65
  - num_epochs: 1.0
66
 
67
+ ### Training results
68
+
69
+ | Training Loss | Epoch | Step | Validation Loss | Accuracy |
70
+ |:-------------:|:-----:|:-----:|:---------------:|:--------:|
71
+ | 2.2432 | 0.01 | 500 | 2.1239 | 0.5358 |
72
+ | 2.209 | 0.02 | 1000 | 2.0922 | 0.5404 |
73
+ | 2.1988 | 0.03 | 1500 | 2.0742 | 0.5436 |
74
+ | 2.1877 | 0.04 | 2000 | 2.0615 | 0.5463 |
75
+ | 2.1743 | 0.05 | 2500 | 2.0514 | 0.5479 |
76
+ | 2.1885 | 0.06 | 3000 | 2.0427 | 0.5495 |
77
+ | 2.1883 | 0.07 | 3500 | 2.0355 | 0.5509 |
78
+ | 2.1954 | 0.08 | 4000 | 2.0298 | 0.5519 |
79
+ | 2.1597 | 0.09 | 4500 | 2.0254 | 0.5526 |
80
+ | 2.1763 | 0.1 | 5000 | 2.0222 | 0.5532 |
81
+ | 2.1413 | 0.11 | 5500 | 2.0195 | 0.5541 |
82
+ | 2.1812 | 0.12 | 6000 | 2.0169 | 0.5545 |
83
+ | 2.1526 | 0.14 | 6500 | 2.0148 | 0.5547 |
84
+ | 2.155 | 0.15 | 7000 | 2.0131 | 0.5554 |
85
+ | 2.1594 | 0.16 | 7500 | 2.0110 | 0.5558 |
86
+ | 2.1681 | 0.17 | 8000 | 2.0097 | 0.5559 |
87
+ | 2.1572 | 0.18 | 8500 | 2.0083 | 0.5562 |
88
+ | 2.0943 | 0.19 | 9000 | 2.0074 | 0.5566 |
89
+ | 2.1421 | 0.2 | 9500 | 2.0063 | 0.5566 |
90
+ | 2.1196 | 0.21 | 10000 | 2.0049 | 0.5568 |
91
+ | 2.1634 | 0.22 | 10500 | 2.0042 | 0.5568 |
92
+ | 2.1361 | 0.23 | 11000 | 2.0035 | 0.5573 |
93
+ | 2.1614 | 0.24 | 11500 | 2.0027 | 0.5572 |
94
+ | 2.1205 | 0.25 | 12000 | 2.0021 | 0.5576 |
95
+ | 2.0984 | 0.26 | 12500 | 2.0011 | 0.5576 |
96
+ | 2.1226 | 0.27 | 13000 | 2.0006 | 0.5575 |
97
+ | 2.1054 | 0.28 | 13500 | 2.0001 | 0.5577 |
98
+ | 2.1297 | 0.29 | 14000 | 1.9997 | 0.5578 |
99
+ | 2.1233 | 0.3 | 14500 | 1.9988 | 0.5581 |
100
+ | 2.1348 | 0.31 | 15000 | 1.9984 | 0.5581 |
101
+ | 2.1494 | 0.32 | 15500 | 1.9980 | 0.5582 |
102
+ | 2.0827 | 0.33 | 16000 | 1.9976 | 0.5584 |
103
+ | 2.0991 | 0.34 | 16500 | 1.9975 | 0.5582 |
104
+ | 2.1108 | 0.35 | 17000 | 1.9972 | 0.5582 |
105
+ | 2.1209 | 0.36 | 17500 | 1.9968 | 0.5583 |
106
+ | 2.1012 | 0.37 | 18000 | 1.9963 | 0.5584 |
107
+ | 2.1155 | 0.38 | 18500 | 1.9959 | 0.5585 |
108
+ | 2.1493 | 0.4 | 19000 | 1.9956 | 0.5585 |
109
+ | 2.1219 | 0.41 | 19500 | 1.9953 | 0.5587 |
110
+ | 2.1584 | 0.42 | 20000 | 1.9952 | 0.5588 |
111
+ | 2.1167 | 0.43 | 20500 | 1.9950 | 0.5587 |
112
+ | 2.1507 | 0.44 | 21000 | 1.9948 | 0.5586 |
113
+ | 2.1043 | 0.45 | 21500 | 1.9946 | 0.5587 |
114
+ | 2.0864 | 0.46 | 22000 | 1.9945 | 0.5587 |
115
+ | 2.1074 | 0.47 | 22500 | 1.9943 | 0.5587 |
116
+ | 2.0858 | 0.48 | 23000 | 1.9942 | 0.5590 |
117
+ | 2.1178 | 0.49 | 23500 | 1.9941 | 0.5588 |
118
+ | 2.1148 | 0.5 | 24000 | 1.9940 | 0.5588 |
119
+ | 2.1165 | 0.51 | 24500 | 1.9939 | 0.5588 |
120
+ | 2.1012 | 0.52 | 25000 | 1.9938 | 0.5590 |
121
+ | 2.1573 | 0.53 | 25500 | 1.9936 | 0.5590 |
122
+ | 2.1674 | 0.54 | 26000 | 1.9936 | 0.5589 |
123
+ | 2.1184 | 0.55 | 26500 | 1.9935 | 0.5590 |
124
+ | 2.1424 | 0.56 | 27000 | 1.9935 | 0.5590 |
125
+ | 2.1437 | 0.57 | 27500 | 1.9935 | 0.5590 |
126
+ | 2.1244 | 0.58 | 28000 | 1.9933 | 0.5591 |
127
+ | 2.0767 | 0.59 | 28500 | 1.9933 | 0.5589 |
128
+ | 2.1182 | 0.6 | 29000 | 1.9934 | 0.5591 |
129
+ | 2.1277 | 0.61 | 29500 | 1.9933 | 0.5591 |
130
+ | 2.1407 | 0.62 | 30000 | 1.9932 | 0.5591 |
131
+ | 2.1222 | 0.63 | 30500 | 1.9932 | 0.5591 |
132
+ | 2.1146 | 0.64 | 31000 | 1.9931 | 0.5591 |
133
+ | 2.1441 | 0.65 | 31500 | 1.9932 | 0.5591 |
134
+ | 2.1224 | 0.67 | 32000 | 1.9931 | 0.5590 |
135
+ | 2.0878 | 0.68 | 32500 | 1.9932 | 0.5591 |
136
+ | 2.1172 | 0.69 | 33000 | 1.9932 | 0.5590 |
137
+ | 2.1166 | 0.7 | 33500 | 1.9931 | 0.5592 |
138
+ | 2.1054 | 0.71 | 34000 | 1.9931 | 0.5591 |
139
+ | 2.0972 | 0.72 | 34500 | 1.9931 | 0.5590 |
140
+ | 2.1228 | 0.73 | 35000 | 1.9931 | 0.5590 |
141
+ | 2.1231 | 0.74 | 35500 | 1.9931 | 0.5592 |
142
+ | 2.0974 | 0.75 | 36000 | 1.9931 | 0.5590 |
143
+ | 2.1025 | 0.76 | 36500 | 1.9931 | 0.5591 |
144
+ | 2.1217 | 0.77 | 37000 | 1.9931 | 0.5590 |
145
+ | 2.1227 | 0.78 | 37500 | 1.9930 | 0.5591 |
146
+ | 2.1272 | 0.79 | 38000 | 1.9931 | 0.5592 |
147
+ | 2.117 | 0.8 | 38500 | 1.9931 | 0.5591 |
148
+ | 2.1325 | 0.81 | 39000 | 1.9931 | 0.5591 |
149
+ | 2.1046 | 0.82 | 39500 | 1.9930 | 0.5591 |
150
+ | 2.1096 | 0.83 | 40000 | 1.9930 | 0.5591 |
151
+ | 2.1149 | 0.84 | 40500 | 1.9931 | 0.5591 |
152
+ | 2.122 | 0.85 | 41000 | 1.9931 | 0.5591 |
153
+ | 2.1137 | 0.86 | 41500 | 1.9931 | 0.5591 |
154
+ | 2.0983 | 0.87 | 42000 | 1.9930 | 0.5590 |
155
+ | 2.1109 | 0.88 | 42500 | 1.9931 | 0.5591 |
156
+ | 2.172 | 0.89 | 43000 | 1.9930 | 0.5590 |
157
+ | 2.0882 | 0.9 | 43500 | 1.9930 | 0.5591 |
158
+ | 2.0646 | 0.91 | 44000 | 1.9930 | 0.5591 |
159
+ | 2.1223 | 0.93 | 44500 | 1.9930 | 0.5591 |
160
+ | 2.1342 | 0.94 | 45000 | 1.9930 | 0.5591 |
161
+ | 2.0991 | 0.95 | 45500 | 1.9930 | 0.5590 |
162
+ | 2.1431 | 0.96 | 46000 | 1.9930 | 0.5592 |
163
+ | 2.0965 | 0.97 | 46500 | 1.9931 | 0.5590 |
164
+ | 2.1377 | 0.98 | 47000 | 1.9931 | 0.5592 |
165
+ | 2.1118 | 0.99 | 47500 | 1.9931 | 0.5592 |
166
+ | 2.089 | 1.0 | 48000 | 1.9930 | 0.5590 |
167
+
168
+
169
  ### Framework versions
170
 
171
  - PEFT 0.10.1.dev0
adapter_model.safetensors CHANGED
@@ -1,3 +1,3 @@
1
  version https://git-lfs.github.com/spec/v1
2
- oid sha256:f8d2cde08e33ca2f842ae37ca875d9beef71974ca2f019b674992367e56fdf73
3
  size 2436952864
 
1
  version https://git-lfs.github.com/spec/v1
2
+ oid sha256:94493ffb396ca39e82a3e8069f1d5c4062cb624e6a7b0b6924dbc5816f4aab9f
3
  size 2436952864
all_results.json ADDED
@@ -0,0 +1,15 @@
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
1
+ {
2
+ "epoch": 1.0,
3
+ "eval_accuracy": 0.5590444975644216,
4
+ "eval_loss": 1.9930245876312256,
5
+ "eval_runtime": 16.3839,
6
+ "eval_samples": 595,
7
+ "eval_samples_per_second": 36.316,
8
+ "eval_steps_per_second": 1.16,
9
+ "perplexity": 7.337693730732541,
10
+ "train_loss": 2.1293136361829954,
11
+ "train_runtime": 442446.2037,
12
+ "train_samples": 6156843,
13
+ "train_samples_per_second": 13.915,
14
+ "train_steps_per_second": 0.109
15
+ }
eval_results.json ADDED
@@ -0,0 +1,10 @@
 
 
 
 
 
 
 
 
 
 
 
1
+ {
2
+ "epoch": 1.0,
3
+ "eval_accuracy": 0.5590444975644216,
4
+ "eval_loss": 1.9930245876312256,
5
+ "eval_runtime": 16.3839,
6
+ "eval_samples": 595,
7
+ "eval_samples_per_second": 36.316,
8
+ "eval_steps_per_second": 1.16,
9
+ "perplexity": 7.337693730732541
10
+ }
pt_lora_model/adapter_config.json CHANGED
@@ -23,13 +23,13 @@
23
  "rank_pattern": {},
24
  "revision": null,
25
  "target_modules": [
26
- "up_proj",
27
- "gate_proj",
28
- "q_proj",
29
- "v_proj",
30
  "o_proj",
 
31
  "down_proj",
32
- "k_proj"
 
 
33
  ],
34
  "task_type": "CAUSAL_LM",
35
  "use_dora": false,
 
23
  "rank_pattern": {},
24
  "revision": null,
25
  "target_modules": [
26
+ "k_proj",
 
 
 
27
  "o_proj",
28
+ "gate_proj",
29
  "down_proj",
30
+ "up_proj",
31
+ "q_proj",
32
+ "v_proj"
33
  ],
34
  "task_type": "CAUSAL_LM",
35
  "use_dora": false,
train_results.json ADDED
@@ -0,0 +1,8 @@
 
 
 
 
 
 
 
 
 
1
+ {
2
+ "epoch": 1.0,
3
+ "train_loss": 2.1293136361829954,
4
+ "train_runtime": 442446.2037,
5
+ "train_samples": 6156843,
6
+ "train_samples_per_second": 13.915,
7
+ "train_steps_per_second": 0.109
8
+ }
trainer_state.json ADDED
The diff for this file is too large to render. See raw diff