Hugging Face's logo

modulora-repoducibility
/
llama-7b-4bit-c4-quantized-samsum-adapter

PEFT
Model card Files Community
osbm commited on
Commit
1624aee
•
1 Parent(s): 10a489c

Upload 19 files

Browse files
Files changed (19) hide show
  1. README.md +9 -0
  2. adapter_config.json +21 -0
  3. adapter_model.bin +3 -0
  4. checkpoint-350/README.md +9 -0
  5. checkpoint-350/adapter_config.json +21 -0
  6. checkpoint-350/adapter_model.bin +3 -0
  7. checkpoint-350/optimizer.pt +3 -0
  8. checkpoint-350/rng_state.pth +3 -0
  9. checkpoint-350/scheduler.pt +3 -0
  10. checkpoint-350/trainer_state.json +117 -0
  11. checkpoint-350/training_args.bin +3 -0
  12. checkpoint-50/README.md +9 -0
  13. checkpoint-50/adapter_config.json +21 -0
  14. checkpoint-50/adapter_model.bin +3 -0
  15. checkpoint-50/optimizer.pt +3 -0
  16. checkpoint-50/rng_state.pth +3 -0
  17. checkpoint-50/scheduler.pt +3 -0
  18. checkpoint-50/trainer_state.json +33 -0
  19. checkpoint-50/training_args.bin +3 -0
README.md ADDED
@@ -0,0 +1,9 @@
 
 
 
 
 
 
 
 
 
 
1
+ ---
2
+ library_name: peft
3
+ ---
4
+ ## Training procedure
5
+
6
+ ### Framework versions
7
+
8
+
9
+ - PEFT 0.5.0
adapter_config.json ADDED
@@ -0,0 +1,21 @@
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
1
+ {
2
+ "auto_mapping": null,
3
+ "base_model_name_or_path": null,
4
+ "bias": "none",
5
+ "fan_in_fan_out": false,
6
+ "inference_mode": true,
7
+ "init_lora_weights": true,
8
+ "layers_pattern": null,
9
+ "layers_to_transform": null,
10
+ "lora_alpha": 32,
11
+ "lora_dropout": 0.1,
12
+ "modules_to_save": null,
13
+ "peft_type": "LORA",
14
+ "r": 8,
15
+ "revision": null,
16
+ "target_modules": [
17
+ "q_proj",
18
+ "v_proj"
19
+ ],
20
+ "task_type": "CAUSAL_LM"
21
+ }
adapter_model.bin ADDED
@@ -0,0 +1,3 @@
 
 
 
 
1
+ version https://git-lfs.github.com/spec/v1
2
+ oid sha256:9289d43fc53025a256dd97fd752b987808899f68e06ede3b9471fa125bf8662e
3
+ size 16824842
checkpoint-350/README.md ADDED
@@ -0,0 +1,9 @@
 
 
 
 
 
 
 
 
 
 
1
+ ---
2
+ library_name: peft
3
+ ---
4
+ ## Training procedure
5
+
6
+ ### Framework versions
7
+
8
+
9
+ - PEFT 0.5.0
checkpoint-350/adapter_config.json ADDED
@@ -0,0 +1,21 @@
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
1
+ {
2
+ "auto_mapping": null,
3
+ "base_model_name_or_path": null,
4
+ "bias": "none",
5
+ "fan_in_fan_out": false,
6
+ "inference_mode": true,
7
+ "init_lora_weights": true,
8
+ "layers_pattern": null,
9
+ "layers_to_transform": null,
10
+ "lora_alpha": 32,
11
+ "lora_dropout": 0.1,
12
+ "modules_to_save": null,
13
+ "peft_type": "LORA",
14
+ "r": 8,
15
+ "revision": null,
16
+ "target_modules": [
17
+ "q_proj",
18
+ "v_proj"
19
+ ],
20
+ "task_type": "CAUSAL_LM"
21
+ }
checkpoint-350/adapter_model.bin ADDED
@@ -0,0 +1,3 @@
 
 
 
 
1
+ version https://git-lfs.github.com/spec/v1
2
+ oid sha256:9289d43fc53025a256dd97fd752b987808899f68e06ede3b9471fa125bf8662e
3
+ size 16824842
checkpoint-350/optimizer.pt ADDED
@@ -0,0 +1,3 @@
 
 
 
 
1
+ version https://git-lfs.github.com/spec/v1
2
+ oid sha256:0dceb1ef0d4c4ae359f4b5da113a32ca1c088ec346cfee7cf9d6a72bf8aacec1
3
+ size 1512
checkpoint-350/rng_state.pth ADDED
@@ -0,0 +1,3 @@
 
 
 
 
1
+ version https://git-lfs.github.com/spec/v1
2
+ oid sha256:871d6aa2fce1d1e046894f25d33034a98e4e2632003973db55cd524ccc8613f4
3
+ size 14244
checkpoint-350/scheduler.pt ADDED
@@ -0,0 +1,3 @@
 
 
 
 
1
+ version https://git-lfs.github.com/spec/v1
2
+ oid sha256:d63c34d420be46c89620f598f59ed0336f83fa393605281d77882cb1a3b89375
3
+ size 1064
checkpoint-350/trainer_state.json ADDED
@@ -0,0 +1,117 @@
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
1
+ {
2
+ "best_metric": NaN,
3
+ "best_model_checkpoint": "outputllama7-4/checkpoint-50",
4
+ "epoch": 2.9932803909590717,
5
+ "eval_steps": 50,
6
+ "global_step": 350,
7
+ "is_hyper_param_search": false,
8
+ "is_local_process_zero": true,
9
+ "is_world_process_zero": true,
10
+ "log_history": [
11
+ {
12
+ "epoch": 0.43,
13
+ "learning_rate": 0.0,
14
+ "loss": 0.0,
15
+ "step": 50
16
+ },
17
+ {
18
+ "epoch": 0.43,
19
+ "eval_loss": NaN,
20
+ "eval_runtime": 27.7284,
21
+ "eval_samples_per_second": 29.537,
22
+ "eval_steps_per_second": 9.846,
23
+ "step": 50
24
+ },
25
+ {
26
+ "epoch": 0.86,
27
+ "learning_rate": 0.0,
28
+ "loss": 0.0,
29
+ "step": 100
30
+ },
31
+ {
32
+ "epoch": 0.86,
33
+ "eval_loss": NaN,
34
+ "eval_runtime": 27.9747,
35
+ "eval_samples_per_second": 29.276,
36
+ "eval_steps_per_second": 9.759,
37
+ "step": 100
38
+ },
39
+ {
40
+ "epoch": 1.28,
41
+ "learning_rate": 0.0,
42
+ "loss": 0.0,
43
+ "step": 150
44
+ },
45
+ {
46
+ "epoch": 1.28,
47
+ "eval_loss": NaN,
48
+ "eval_runtime": 28.2253,
49
+ "eval_samples_per_second": 29.016,
50
+ "eval_steps_per_second": 9.672,
51
+ "step": 150
52
+ },
53
+ {
54
+ "epoch": 1.71,
55
+ "learning_rate": 0.0,
56
+ "loss": 0.0,
57
+ "step": 200
58
+ },
59
+ {
60
+ "epoch": 1.71,
61
+ "eval_loss": NaN,
62
+ "eval_runtime": 27.4046,
63
+ "eval_samples_per_second": 29.886,
64
+ "eval_steps_per_second": 9.962,
65
+ "step": 200
66
+ },
67
+ {
68
+ "epoch": 2.14,
69
+ "learning_rate": 0.0,
70
+ "loss": 0.0,
71
+ "step": 250
72
+ },
73
+ {
74
+ "epoch": 2.14,
75
+ "eval_loss": NaN,
76
+ "eval_runtime": 27.3115,
77
+ "eval_samples_per_second": 29.987,
78
+ "eval_steps_per_second": 9.996,
79
+ "step": 250
80
+ },
81
+ {
82
+ "epoch": 2.57,
83
+ "learning_rate": 0.0,
84
+ "loss": 0.0,
85
+ "step": 300
86
+ },
87
+ {
88
+ "epoch": 2.57,
89
+ "eval_loss": NaN,
90
+ "eval_runtime": 27.4016,
91
+ "eval_samples_per_second": 29.889,
92
+ "eval_steps_per_second": 9.963,
93
+ "step": 300
94
+ },
95
+ {
96
+ "epoch": 2.99,
97
+ "learning_rate": 0.0,
98
+ "loss": 0.0,
99
+ "step": 350
100
+ },
101
+ {
102
+ "epoch": 2.99,
103
+ "eval_loss": NaN,
104
+ "eval_runtime": 27.3965,
105
+ "eval_samples_per_second": 29.894,
106
+ "eval_steps_per_second": 9.965,
107
+ "step": 350
108
+ }
109
+ ],
110
+ "logging_steps": 50,
111
+ "max_steps": 350,
112
+ "num_train_epochs": 4,
113
+ "save_steps": 50,
114
+ "total_flos": 0.0,
115
+ "trial_name": null,
116
+ "trial_params": null
117
+ }
checkpoint-350/training_args.bin ADDED
@@ -0,0 +1,3 @@
 
 
 
 
1
+ version https://git-lfs.github.com/spec/v1
2
+ oid sha256:4f4613b420540b363d2a86ae0776c2f8c7551e27c74090632a64931cd4dd4abb
3
+ size 4472
checkpoint-50/README.md ADDED
@@ -0,0 +1,9 @@
 
 
 
 
 
 
 
 
 
 
1
+ ---
2
+ library_name: peft
3
+ ---
4
+ ## Training procedure
5
+
6
+ ### Framework versions
7
+
8
+
9
+ - PEFT 0.5.0
checkpoint-50/adapter_config.json ADDED
@@ -0,0 +1,21 @@
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
1
+ {
2
+ "auto_mapping": null,
3
+ "base_model_name_or_path": null,
4
+ "bias": "none",
5
+ "fan_in_fan_out": false,
6
+ "inference_mode": true,
7
+ "init_lora_weights": true,
8
+ "layers_pattern": null,
9
+ "layers_to_transform": null,
10
+ "lora_alpha": 32,
11
+ "lora_dropout": 0.1,
12
+ "modules_to_save": null,
13
+ "peft_type": "LORA",
14
+ "r": 8,
15
+ "revision": null,
16
+ "target_modules": [
17
+ "q_proj",
18
+ "v_proj"
19
+ ],
20
+ "task_type": "CAUSAL_LM"
21
+ }
checkpoint-50/adapter_model.bin ADDED
@@ -0,0 +1,3 @@
 
 
 
 
1
+ version https://git-lfs.github.com/spec/v1
2
+ oid sha256:9289d43fc53025a256dd97fd752b987808899f68e06ede3b9471fa125bf8662e
3
+ size 16824842
checkpoint-50/optimizer.pt ADDED
@@ -0,0 +1,3 @@
 
 
 
 
1
+ version https://git-lfs.github.com/spec/v1
2
+ oid sha256:0dceb1ef0d4c4ae359f4b5da113a32ca1c088ec346cfee7cf9d6a72bf8aacec1
3
+ size 1512
checkpoint-50/rng_state.pth ADDED
@@ -0,0 +1,3 @@
 
 
 
 
1
+ version https://git-lfs.github.com/spec/v1
2
+ oid sha256:e91601dd7e71c9647411cd0190f59e192eb4db58a8ce1ca40db035bd1eb6c8f3
3
+ size 14244
checkpoint-50/scheduler.pt ADDED
@@ -0,0 +1,3 @@
 
 
 
 
1
+ version https://git-lfs.github.com/spec/v1
2
+ oid sha256:d63c34d420be46c89620f598f59ed0336f83fa393605281d77882cb1a3b89375
3
+ size 1064
checkpoint-50/trainer_state.json ADDED
@@ -0,0 +1,33 @@
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
1
+ {
2
+ "best_metric": NaN,
3
+ "best_model_checkpoint": "outputllama7-4/checkpoint-50",
4
+ "epoch": 0.4276114844227245,
5
+ "eval_steps": 50,
6
+ "global_step": 50,
7
+ "is_hyper_param_search": false,
8
+ "is_local_process_zero": true,
9
+ "is_world_process_zero": true,
10
+ "log_history": [
11
+ {
12
+ "epoch": 0.43,
13
+ "learning_rate": 0.0,
14
+ "loss": 0.0,
15
+ "step": 50
16
+ },
17
+ {
18
+ "epoch": 0.43,
19
+ "eval_loss": NaN,
20
+ "eval_runtime": 27.7284,
21
+ "eval_samples_per_second": 29.537,
22
+ "eval_steps_per_second": 9.846,
23
+ "step": 50
24
+ }
25
+ ],
26
+ "logging_steps": 50,
27
+ "max_steps": 350,
28
+ "num_train_epochs": 4,
29
+ "save_steps": 50,
30
+ "total_flos": 0.0,
31
+ "trial_name": null,
32
+ "trial_params": null
33
+ }
checkpoint-50/training_args.bin ADDED
@@ -0,0 +1,3 @@
 
 
 
 
1
+ version https://git-lfs.github.com/spec/v1
2
+ oid sha256:4f4613b420540b363d2a86ae0776c2f8c7551e27c74090632a64931cd4dd4abb
3
+ size 4472