Ciaranshu commited on
Commit
7833cf2
1 Parent(s): 1e29b48
README.md ADDED
@@ -0,0 +1,20 @@
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
1
+ ---
2
+ library_name: peft
3
+ ---
4
+ ## Training procedure
5
+
6
+
7
+ The following `bitsandbytes` quantization config was used during training:
8
+ - load_in_8bit: True
9
+ - load_in_4bit: False
10
+ - llm_int8_threshold: 6.0
11
+ - llm_int8_skip_modules: None
12
+ - llm_int8_enable_fp32_cpu_offload: False
13
+ - llm_int8_has_fp16_weight: False
14
+ - bnb_4bit_quant_type: fp4
15
+ - bnb_4bit_use_double_quant: False
16
+ - bnb_4bit_compute_dtype: float32
17
+ ### Framework versions
18
+
19
+
20
+ - PEFT 0.4.0
adapter_config.json ADDED
@@ -0,0 +1,21 @@
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
1
+ {
2
+ "auto_mapping": null,
3
+ "base_model_name_or_path": "/mnt/nas_home/cs2175/models/CodeLlama-7b-Instruct-hf",
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": 16,
11
+ "lora_dropout": 0.05,
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:e7b8cc13150e0aa54e442041891053c2d0fc97d92f93e2e36ab283a6c098902c
3
+ size 16822989
checkpoint-200/README.md ADDED
@@ -0,0 +1,20 @@
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
1
+ ---
2
+ library_name: peft
3
+ ---
4
+ ## Training procedure
5
+
6
+
7
+ The following `bitsandbytes` quantization config was used during training:
8
+ - load_in_8bit: True
9
+ - load_in_4bit: False
10
+ - llm_int8_threshold: 6.0
11
+ - llm_int8_skip_modules: None
12
+ - llm_int8_enable_fp32_cpu_offload: False
13
+ - llm_int8_has_fp16_weight: False
14
+ - bnb_4bit_quant_type: fp4
15
+ - bnb_4bit_use_double_quant: False
16
+ - bnb_4bit_compute_dtype: float32
17
+ ### Framework versions
18
+
19
+
20
+ - PEFT 0.4.0
checkpoint-200/adapter_config.json ADDED
@@ -0,0 +1,21 @@
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
1
+ {
2
+ "auto_mapping": null,
3
+ "base_model_name_or_path": "/mnt/nas_home/cs2175/models/CodeLlama-7b-Instruct-hf",
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": 16,
11
+ "lora_dropout": 0.05,
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-200/adapter_model.bin ADDED
@@ -0,0 +1,3 @@
 
 
 
 
1
+ version https://git-lfs.github.com/spec/v1
2
+ oid sha256:e7b8cc13150e0aa54e442041891053c2d0fc97d92f93e2e36ab283a6c098902c
3
+ size 16822989
checkpoint-200/optimizer.pt ADDED
@@ -0,0 +1,3 @@
 
 
 
 
1
+ version https://git-lfs.github.com/spec/v1
2
+ oid sha256:53f40d78b337416e504b05683731dd717d98decb6e1e74de603273c90a455bdf
3
+ size 33629765
checkpoint-200/rng_state.pth ADDED
@@ -0,0 +1,3 @@
 
 
 
 
1
+ version https://git-lfs.github.com/spec/v1
2
+ oid sha256:1692f2c1aad57ec78503ac58ad66fed77cd92f78e2b4642934bff98cd699e521
3
+ size 14575
checkpoint-200/scheduler.pt ADDED
@@ -0,0 +1,3 @@
 
 
 
 
1
+ version https://git-lfs.github.com/spec/v1
2
+ oid sha256:03c512da1dc86834c4bb76144aebc692f2e54d6024b2e620dbad6400a2851e7d
3
+ size 627
checkpoint-200/trainer_state.json ADDED
@@ -0,0 +1,144 @@
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
1
+ {
2
+ "best_metric": 0.7773334980010986,
3
+ "best_model_checkpoint": "../../models/lora/codellama-7b-multitask_cot_reflection/checkpoint-200",
4
+ "epoch": 12.5,
5
+ "global_step": 200,
6
+ "is_hyper_param_search": false,
7
+ "is_local_process_zero": true,
8
+ "is_world_process_zero": true,
9
+ "log_history": [
10
+ {
11
+ "epoch": 0.62,
12
+ "learning_rate": 2.9999999999999997e-05,
13
+ "loss": 2.4326,
14
+ "step": 10
15
+ },
16
+ {
17
+ "epoch": 1.25,
18
+ "learning_rate": 5.9999999999999995e-05,
19
+ "loss": 2.3071,
20
+ "step": 20
21
+ },
22
+ {
23
+ "epoch": 1.88,
24
+ "learning_rate": 8.699999999999999e-05,
25
+ "loss": 2.0883,
26
+ "step": 30
27
+ },
28
+ {
29
+ "epoch": 2.5,
30
+ "learning_rate": 0.000117,
31
+ "loss": 1.8582,
32
+ "step": 40
33
+ },
34
+ {
35
+ "epoch": 3.12,
36
+ "learning_rate": 0.00014399999999999998,
37
+ "loss": 1.4434,
38
+ "step": 50
39
+ },
40
+ {
41
+ "epoch": 3.75,
42
+ "learning_rate": 0.00017399999999999997,
43
+ "loss": 1.1487,
44
+ "step": 60
45
+ },
46
+ {
47
+ "epoch": 4.38,
48
+ "learning_rate": 0.000204,
49
+ "loss": 1.042,
50
+ "step": 70
51
+ },
52
+ {
53
+ "epoch": 5.0,
54
+ "learning_rate": 0.000234,
55
+ "loss": 1.0089,
56
+ "step": 80
57
+ },
58
+ {
59
+ "epoch": 5.62,
60
+ "learning_rate": 0.00026399999999999997,
61
+ "loss": 0.9878,
62
+ "step": 90
63
+ },
64
+ {
65
+ "epoch": 6.25,
66
+ "learning_rate": 0.000294,
67
+ "loss": 0.9677,
68
+ "step": 100
69
+ },
70
+ {
71
+ "epoch": 6.88,
72
+ "learning_rate": 0.0002936842105263158,
73
+ "loss": 0.9404,
74
+ "step": 110
75
+ },
76
+ {
77
+ "epoch": 7.5,
78
+ "learning_rate": 0.0002857894736842105,
79
+ "loss": 0.9229,
80
+ "step": 120
81
+ },
82
+ {
83
+ "epoch": 8.12,
84
+ "learning_rate": 0.00027789473684210523,
85
+ "loss": 0.9233,
86
+ "step": 130
87
+ },
88
+ {
89
+ "epoch": 8.75,
90
+ "learning_rate": 0.00027,
91
+ "loss": 0.8922,
92
+ "step": 140
93
+ },
94
+ {
95
+ "epoch": 9.38,
96
+ "learning_rate": 0.0002621052631578947,
97
+ "loss": 0.8933,
98
+ "step": 150
99
+ },
100
+ {
101
+ "epoch": 10.0,
102
+ "learning_rate": 0.00025421052631578945,
103
+ "loss": 0.8536,
104
+ "step": 160
105
+ },
106
+ {
107
+ "epoch": 10.62,
108
+ "learning_rate": 0.00024631578947368417,
109
+ "loss": 0.8694,
110
+ "step": 170
111
+ },
112
+ {
113
+ "epoch": 11.25,
114
+ "learning_rate": 0.00023842105263157895,
115
+ "loss": 0.8344,
116
+ "step": 180
117
+ },
118
+ {
119
+ "epoch": 11.88,
120
+ "learning_rate": 0.00023052631578947364,
121
+ "loss": 0.8342,
122
+ "step": 190
123
+ },
124
+ {
125
+ "epoch": 12.5,
126
+ "learning_rate": 0.0002226315789473684,
127
+ "loss": 0.8227,
128
+ "step": 200
129
+ },
130
+ {
131
+ "epoch": 12.5,
132
+ "eval_loss": 0.7773334980010986,
133
+ "eval_runtime": 0.9943,
134
+ "eval_samples_per_second": 21.12,
135
+ "eval_steps_per_second": 3.017,
136
+ "step": 200
137
+ }
138
+ ],
139
+ "max_steps": 480,
140
+ "num_train_epochs": 30,
141
+ "total_flos": 2.669105373043753e+17,
142
+ "trial_name": null,
143
+ "trial_params": null
144
+ }
checkpoint-200/training_args.bin ADDED
@@ -0,0 +1,3 @@
 
 
 
 
1
+ version https://git-lfs.github.com/spec/v1
2
+ oid sha256:e23f8efc0898dc6f2dbc18b3c7163dd6932c7f78062f3d4f8b8790631210e0fc
3
+ size 4027
checkpoint-400/README.md ADDED
@@ -0,0 +1,20 @@
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
1
+ ---
2
+ library_name: peft
3
+ ---
4
+ ## Training procedure
5
+
6
+
7
+ The following `bitsandbytes` quantization config was used during training:
8
+ - load_in_8bit: True
9
+ - load_in_4bit: False
10
+ - llm_int8_threshold: 6.0
11
+ - llm_int8_skip_modules: None
12
+ - llm_int8_enable_fp32_cpu_offload: False
13
+ - llm_int8_has_fp16_weight: False
14
+ - bnb_4bit_quant_type: fp4
15
+ - bnb_4bit_use_double_quant: False
16
+ - bnb_4bit_compute_dtype: float32
17
+ ### Framework versions
18
+
19
+
20
+ - PEFT 0.4.0
checkpoint-400/adapter_config.json ADDED
@@ -0,0 +1,21 @@
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
1
+ {
2
+ "auto_mapping": null,
3
+ "base_model_name_or_path": "/mnt/nas_home/cs2175/models/CodeLlama-7b-Instruct-hf",
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": 16,
11
+ "lora_dropout": 0.05,
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-400/adapter_model.bin ADDED
@@ -0,0 +1,3 @@
 
 
 
 
1
+ version https://git-lfs.github.com/spec/v1
2
+ oid sha256:d2e8484724521e9b466a19d09e2bdc5ae6ec21da96c0a1d66c5f7c5413faf3ca
3
+ size 16822989
checkpoint-400/optimizer.pt ADDED
@@ -0,0 +1,3 @@
 
 
 
 
1
+ version https://git-lfs.github.com/spec/v1
2
+ oid sha256:a3d951a54258b814f103e1b9e3ffafb1b19dff2223366fb0b4aef375391cc943
3
+ size 33629893
checkpoint-400/rng_state.pth ADDED
@@ -0,0 +1,3 @@
 
 
 
 
1
+ version https://git-lfs.github.com/spec/v1
2
+ oid sha256:743bf87ddc94d8ee3121dbadf7b4a640f749203f6f1e4f4f667d811a88eb54ae
3
+ size 14575
checkpoint-400/scheduler.pt ADDED
@@ -0,0 +1,3 @@
 
 
 
 
1
+ version https://git-lfs.github.com/spec/v1
2
+ oid sha256:8944a19145692ec3d40c8cde3bc6852e7450c71a57993cccc642cd173303150e
3
+ size 627
checkpoint-400/trainer_state.json ADDED
@@ -0,0 +1,272 @@
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
1
+ {
2
+ "best_metric": 0.7773334980010986,
3
+ "best_model_checkpoint": "../../models/lora/codellama-7b-multitask_cot_reflection/checkpoint-200",
4
+ "epoch": 25.0,
5
+ "global_step": 400,
6
+ "is_hyper_param_search": false,
7
+ "is_local_process_zero": true,
8
+ "is_world_process_zero": true,
9
+ "log_history": [
10
+ {
11
+ "epoch": 0.62,
12
+ "learning_rate": 2.9999999999999997e-05,
13
+ "loss": 2.4326,
14
+ "step": 10
15
+ },
16
+ {
17
+ "epoch": 1.25,
18
+ "learning_rate": 5.9999999999999995e-05,
19
+ "loss": 2.3071,
20
+ "step": 20
21
+ },
22
+ {
23
+ "epoch": 1.88,
24
+ "learning_rate": 8.699999999999999e-05,
25
+ "loss": 2.0883,
26
+ "step": 30
27
+ },
28
+ {
29
+ "epoch": 2.5,
30
+ "learning_rate": 0.000117,
31
+ "loss": 1.8582,
32
+ "step": 40
33
+ },
34
+ {
35
+ "epoch": 3.12,
36
+ "learning_rate": 0.00014399999999999998,
37
+ "loss": 1.4434,
38
+ "step": 50
39
+ },
40
+ {
41
+ "epoch": 3.75,
42
+ "learning_rate": 0.00017399999999999997,
43
+ "loss": 1.1487,
44
+ "step": 60
45
+ },
46
+ {
47
+ "epoch": 4.38,
48
+ "learning_rate": 0.000204,
49
+ "loss": 1.042,
50
+ "step": 70
51
+ },
52
+ {
53
+ "epoch": 5.0,
54
+ "learning_rate": 0.000234,
55
+ "loss": 1.0089,
56
+ "step": 80
57
+ },
58
+ {
59
+ "epoch": 5.62,
60
+ "learning_rate": 0.00026399999999999997,
61
+ "loss": 0.9878,
62
+ "step": 90
63
+ },
64
+ {
65
+ "epoch": 6.25,
66
+ "learning_rate": 0.000294,
67
+ "loss": 0.9677,
68
+ "step": 100
69
+ },
70
+ {
71
+ "epoch": 6.88,
72
+ "learning_rate": 0.0002936842105263158,
73
+ "loss": 0.9404,
74
+ "step": 110
75
+ },
76
+ {
77
+ "epoch": 7.5,
78
+ "learning_rate": 0.0002857894736842105,
79
+ "loss": 0.9229,
80
+ "step": 120
81
+ },
82
+ {
83
+ "epoch": 8.12,
84
+ "learning_rate": 0.00027789473684210523,
85
+ "loss": 0.9233,
86
+ "step": 130
87
+ },
88
+ {
89
+ "epoch": 8.75,
90
+ "learning_rate": 0.00027,
91
+ "loss": 0.8922,
92
+ "step": 140
93
+ },
94
+ {
95
+ "epoch": 9.38,
96
+ "learning_rate": 0.0002621052631578947,
97
+ "loss": 0.8933,
98
+ "step": 150
99
+ },
100
+ {
101
+ "epoch": 10.0,
102
+ "learning_rate": 0.00025421052631578945,
103
+ "loss": 0.8536,
104
+ "step": 160
105
+ },
106
+ {
107
+ "epoch": 10.62,
108
+ "learning_rate": 0.00024631578947368417,
109
+ "loss": 0.8694,
110
+ "step": 170
111
+ },
112
+ {
113
+ "epoch": 11.25,
114
+ "learning_rate": 0.00023842105263157895,
115
+ "loss": 0.8344,
116
+ "step": 180
117
+ },
118
+ {
119
+ "epoch": 11.88,
120
+ "learning_rate": 0.00023052631578947364,
121
+ "loss": 0.8342,
122
+ "step": 190
123
+ },
124
+ {
125
+ "epoch": 12.5,
126
+ "learning_rate": 0.0002226315789473684,
127
+ "loss": 0.8227,
128
+ "step": 200
129
+ },
130
+ {
131
+ "epoch": 12.5,
132
+ "eval_loss": 0.7773334980010986,
133
+ "eval_runtime": 0.9943,
134
+ "eval_samples_per_second": 21.12,
135
+ "eval_steps_per_second": 3.017,
136
+ "step": 200
137
+ },
138
+ {
139
+ "epoch": 13.12,
140
+ "learning_rate": 0.00021473684210526314,
141
+ "loss": 0.8147,
142
+ "step": 210
143
+ },
144
+ {
145
+ "epoch": 13.75,
146
+ "learning_rate": 0.0002068421052631579,
147
+ "loss": 0.7956,
148
+ "step": 220
149
+ },
150
+ {
151
+ "epoch": 14.38,
152
+ "learning_rate": 0.0001989473684210526,
153
+ "loss": 0.8037,
154
+ "step": 230
155
+ },
156
+ {
157
+ "epoch": 15.0,
158
+ "learning_rate": 0.00019105263157894736,
159
+ "loss": 0.7974,
160
+ "step": 240
161
+ },
162
+ {
163
+ "epoch": 15.62,
164
+ "learning_rate": 0.00018315789473684208,
165
+ "loss": 0.8115,
166
+ "step": 250
167
+ },
168
+ {
169
+ "epoch": 16.25,
170
+ "learning_rate": 0.00017526315789473683,
171
+ "loss": 0.7757,
172
+ "step": 260
173
+ },
174
+ {
175
+ "epoch": 16.88,
176
+ "learning_rate": 0.00016736842105263155,
177
+ "loss": 0.7783,
178
+ "step": 270
179
+ },
180
+ {
181
+ "epoch": 17.5,
182
+ "learning_rate": 0.0001594736842105263,
183
+ "loss": 0.755,
184
+ "step": 280
185
+ },
186
+ {
187
+ "epoch": 18.12,
188
+ "learning_rate": 0.00015157894736842105,
189
+ "loss": 0.7654,
190
+ "step": 290
191
+ },
192
+ {
193
+ "epoch": 18.75,
194
+ "learning_rate": 0.00014368421052631577,
195
+ "loss": 0.7366,
196
+ "step": 300
197
+ },
198
+ {
199
+ "epoch": 19.38,
200
+ "learning_rate": 0.00013578947368421052,
201
+ "loss": 0.7439,
202
+ "step": 310
203
+ },
204
+ {
205
+ "epoch": 20.0,
206
+ "learning_rate": 0.00012789473684210524,
207
+ "loss": 0.7442,
208
+ "step": 320
209
+ },
210
+ {
211
+ "epoch": 20.62,
212
+ "learning_rate": 0.00011999999999999999,
213
+ "loss": 0.75,
214
+ "step": 330
215
+ },
216
+ {
217
+ "epoch": 21.25,
218
+ "learning_rate": 0.00011210526315789472,
219
+ "loss": 0.7311,
220
+ "step": 340
221
+ },
222
+ {
223
+ "epoch": 21.88,
224
+ "learning_rate": 0.00010421052631578947,
225
+ "loss": 0.7194,
226
+ "step": 350
227
+ },
228
+ {
229
+ "epoch": 22.5,
230
+ "learning_rate": 9.63157894736842e-05,
231
+ "loss": 0.712,
232
+ "step": 360
233
+ },
234
+ {
235
+ "epoch": 23.12,
236
+ "learning_rate": 8.842105263157893e-05,
237
+ "loss": 0.7149,
238
+ "step": 370
239
+ },
240
+ {
241
+ "epoch": 23.75,
242
+ "learning_rate": 8.052631578947368e-05,
243
+ "loss": 0.6846,
244
+ "step": 380
245
+ },
246
+ {
247
+ "epoch": 24.38,
248
+ "learning_rate": 7.263157894736842e-05,
249
+ "loss": 0.7005,
250
+ "step": 390
251
+ },
252
+ {
253
+ "epoch": 25.0,
254
+ "learning_rate": 6.473684210526315e-05,
255
+ "loss": 0.696,
256
+ "step": 400
257
+ },
258
+ {
259
+ "epoch": 25.0,
260
+ "eval_loss": 0.8131171464920044,
261
+ "eval_runtime": 0.9869,
262
+ "eval_samples_per_second": 21.28,
263
+ "eval_steps_per_second": 3.04,
264
+ "step": 400
265
+ }
266
+ ],
267
+ "max_steps": 480,
268
+ "num_train_epochs": 30,
269
+ "total_flos": 5.3309496187079885e+17,
270
+ "trial_name": null,
271
+ "trial_params": null
272
+ }
checkpoint-400/training_args.bin ADDED
@@ -0,0 +1,3 @@
 
 
 
 
1
+ version https://git-lfs.github.com/spec/v1
2
+ oid sha256:e23f8efc0898dc6f2dbc18b3c7163dd6932c7f78062f3d4f8b8790631210e0fc
3
+ size 4027
runs/Sep21_07-30-54_ltl-gpu05/events.out.tfevents.1695277855.ltl-gpu05.57264.0 ADDED
@@ -0,0 +1,3 @@
 
 
 
 
1
+ version https://git-lfs.github.com/spec/v1
2
+ oid sha256:7b1433617a1474f57ef860c2b10cefe258c5dee1c810028562d4da4ff6e9e986
3
+ size 12955