madiedgar commited on
Commit
0a09c3f
·
1 Parent(s): 689d1ea

feat: consolidate all condition adapters into hub repo subdirectories (#4)

Browse files

- feat: consolidate all condition adapters into hub repo subdirectories (e669ee8fcee93de9ec766706d595e4d2a8739cb1)

condition-1-en-5k/adapter_config.json ADDED
@@ -0,0 +1,50 @@
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
1
+ {
2
+ "alora_invocation_tokens": null,
3
+ "alpha_pattern": {},
4
+ "arrow_config": null,
5
+ "auto_mapping": {
6
+ "base_model_class": "Cohere2ForCausalLM",
7
+ "parent_library": "transformers.models.cohere2.modeling_cohere2",
8
+ "unsloth_fixed": true
9
+ },
10
+ "base_model_name_or_path": "CohereLabs/tiny-aya-base",
11
+ "bias": "none",
12
+ "corda_config": null,
13
+ "ensure_weight_tying": false,
14
+ "eva_config": null,
15
+ "exclude_modules": null,
16
+ "fan_in_fan_out": false,
17
+ "inference_mode": true,
18
+ "init_lora_weights": true,
19
+ "layer_replication": null,
20
+ "layers_pattern": null,
21
+ "layers_to_transform": null,
22
+ "loftq_config": {},
23
+ "lora_alpha": 32,
24
+ "lora_bias": false,
25
+ "lora_dropout": 0.0,
26
+ "megatron_config": null,
27
+ "megatron_core": "megatron.core",
28
+ "modules_to_save": null,
29
+ "peft_type": "LORA",
30
+ "peft_version": "0.18.1",
31
+ "qalora_group_size": 16,
32
+ "r": 16,
33
+ "rank_pattern": {},
34
+ "revision": null,
35
+ "target_modules": [
36
+ "k_proj",
37
+ "q_proj",
38
+ "o_proj",
39
+ "up_proj",
40
+ "v_proj",
41
+ "gate_proj",
42
+ "down_proj"
43
+ ],
44
+ "target_parameters": null,
45
+ "task_type": "CAUSAL_LM",
46
+ "trainable_token_indices": null,
47
+ "use_dora": false,
48
+ "use_qalora": false,
49
+ "use_rslora": false
50
+ }
condition-1-en-5k/adapter_model.safetensors ADDED
@@ -0,0 +1,3 @@
 
 
 
 
1
+ version https://git-lfs.github.com/spec/v1
2
+ oid sha256:56d7139396d7a97a9b9e765f0d7c5e03ed114d88c52afbc648c4b87f9fd25832
3
+ size 120981200
condition-1-en-5k/training_metrics.json ADDED
@@ -0,0 +1,207 @@
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
1
+ [
2
+ {
3
+ "loss": 1.0164,
4
+ "grad_norm": 0.039307065308094025,
5
+ "learning_rate": 0.00012,
6
+ "epoch": 0.03546099290780142,
7
+ "step": 10
8
+ },
9
+ {
10
+ "loss": 1.1027,
11
+ "grad_norm": 0.03674926236271858,
12
+ "learning_rate": 0.00019988926445681492,
13
+ "epoch": 0.07092198581560284,
14
+ "step": 20
15
+ },
16
+ {
17
+ "loss": 1.1062,
18
+ "grad_norm": 0.03771359100937843,
19
+ "learning_rate": 0.0001986463043917528,
20
+ "epoch": 0.10638297872340426,
21
+ "step": 30
22
+ },
23
+ {
24
+ "loss": 1.0428,
25
+ "grad_norm": 0.03601245582103729,
26
+ "learning_rate": 0.00019603921063437793,
27
+ "epoch": 0.14184397163120568,
28
+ "step": 40
29
+ },
30
+ {
31
+ "loss": 1.0116,
32
+ "grad_norm": 0.04361054301261902,
33
+ "learning_rate": 0.0001921040354671897,
34
+ "epoch": 0.1773049645390071,
35
+ "step": 50
36
+ },
37
+ {
38
+ "loss": 1.0443,
39
+ "grad_norm": 0.04239841178059578,
40
+ "learning_rate": 0.00018689519659051467,
41
+ "epoch": 0.2127659574468085,
42
+ "step": 60
43
+ },
44
+ {
45
+ "loss": 1.0685,
46
+ "grad_norm": 0.03858834505081177,
47
+ "learning_rate": 0.00018048472460553257,
48
+ "epoch": 0.24822695035460993,
49
+ "step": 70
50
+ },
51
+ {
52
+ "loss": 1.0954,
53
+ "grad_norm": 0.058491941541433334,
54
+ "learning_rate": 0.00017296126693671884,
55
+ "epoch": 0.28368794326241137,
56
+ "step": 80
57
+ },
58
+ {
59
+ "loss": 1.0308,
60
+ "grad_norm": 0.04502752423286438,
61
+ "learning_rate": 0.00016442886196799464,
62
+ "epoch": 0.3191489361702128,
63
+ "step": 90
64
+ },
65
+ {
66
+ "loss": 1.0046,
67
+ "grad_norm": 0.03750888630747795,
68
+ "learning_rate": 0.00015500550034448413,
69
+ "epoch": 0.3546099290780142,
70
+ "step": 100
71
+ },
72
+ {
73
+ "loss": 1.0566,
74
+ "grad_norm": 0.04296468198299408,
75
+ "learning_rate": 0.00014482149333496454,
76
+ "epoch": 0.3900709219858156,
77
+ "step": 110
78
+ },
79
+ {
80
+ "loss": 1.0718,
81
+ "grad_norm": 0.04349366948008537,
82
+ "learning_rate": 0.0001340176708181637,
83
+ "epoch": 0.425531914893617,
84
+ "step": 120
85
+ },
86
+ {
87
+ "loss": 1.0893,
88
+ "grad_norm": 0.04348432272672653,
89
+ "learning_rate": 0.00012274343381211066,
90
+ "epoch": 0.46099290780141844,
91
+ "step": 130
92
+ },
93
+ {
94
+ "loss": 1.0306,
95
+ "grad_norm": 0.039543844759464264,
96
+ "learning_rate": 0.00011115468847720245,
97
+ "epoch": 0.49645390070921985,
98
+ "step": 140
99
+ },
100
+ {
101
+ "loss": 1.0367,
102
+ "grad_norm": 0.03936546668410301,
103
+ "learning_rate": 9.941169016269379e-05,
104
+ "epoch": 0.5319148936170213,
105
+ "step": 150
106
+ },
107
+ {
108
+ "loss": 1.0447,
109
+ "grad_norm": 0.03707996383309364,
110
+ "learning_rate": 8.767682731028415e-05,
111
+ "epoch": 0.5673758865248227,
112
+ "step": 160
113
+ },
114
+ {
115
+ "loss": 1.0971,
116
+ "grad_norm": 0.04127993807196617,
117
+ "learning_rate": 7.611237586016557e-05,
118
+ "epoch": 0.6028368794326241,
119
+ "step": 170
120
+ },
121
+ {
122
+ "loss": 1.0177,
123
+ "grad_norm": 0.04585001990199089,
124
+ "learning_rate": 6.487825521280109e-05,
125
+ "epoch": 0.6382978723404256,
126
+ "step": 180
127
+ },
128
+ {
129
+ "loss": 1.0081,
130
+ "grad_norm": 0.04047301039099693,
131
+ "learning_rate": 5.4129816778190936e-05,
132
+ "epoch": 0.6737588652482269,
133
+ "step": 190
134
+ },
135
+ {
136
+ "loss": 1.1236,
137
+ "grad_norm": 0.03668837621808052,
138
+ "learning_rate": 4.401569569374668e-05,
139
+ "epoch": 0.7092198581560284,
140
+ "step": 200
141
+ },
142
+ {
143
+ "loss": 1.1207,
144
+ "grad_norm": 0.039949752390384674,
145
+ "learning_rate": 3.467575541836305e-05,
146
+ "epoch": 0.7446808510638298,
147
+ "step": 210
148
+ },
149
+ {
150
+ "loss": 1.0138,
151
+ "grad_norm": 0.03711694851517677,
152
+ "learning_rate": 2.6239153625937784e-05,
153
+ "epoch": 0.7801418439716312,
154
+ "step": 220
155
+ },
156
+ {
157
+ "loss": 1.0105,
158
+ "grad_norm": 0.05135485529899597,
159
+ "learning_rate": 1.882255614419376e-05,
160
+ "epoch": 0.8156028368794326,
161
+ "step": 230
162
+ },
163
+ {
164
+ "loss": 1.0054,
165
+ "grad_norm": 0.04557114839553833,
166
+ "learning_rate": 1.2528523637410838e-05,
167
+ "epoch": 0.851063829787234,
168
+ "step": 240
169
+ },
170
+ {
171
+ "loss": 1.1106,
172
+ "grad_norm": 0.04462781921029091,
173
+ "learning_rate": 7.4440933428779e-06,
174
+ "epoch": 0.8865248226950354,
175
+ "step": 250
176
+ },
177
+ {
178
+ "loss": 1.0755,
179
+ "grad_norm": 0.03919963911175728,
180
+ "learning_rate": 3.6395754735699894e-06,
181
+ "epoch": 0.9219858156028369,
182
+ "step": 260
183
+ },
184
+ {
185
+ "loss": 1.0508,
186
+ "grad_norm": 0.03482827916741371,
187
+ "learning_rate": 1.1675809310361497e-06,
188
+ "epoch": 0.9574468085106383,
189
+ "step": 270
190
+ },
191
+ {
192
+ "loss": 1.062,
193
+ "grad_norm": 0.03561776503920555,
194
+ "learning_rate": 6.229377380218005e-08,
195
+ "epoch": 0.9929078014184397,
196
+ "step": 280
197
+ },
198
+ {
199
+ "train_runtime": 4962.1033,
200
+ "train_samples_per_second": 0.907,
201
+ "train_steps_per_second": 0.057,
202
+ "total_flos": 7.86793932867502e+16,
203
+ "train_loss": 1.0566039296752172,
204
+ "epoch": 1.0,
205
+ "step": 282
206
+ }
207
+ ]
condition-2-es-5k/adapter_config.json ADDED
@@ -0,0 +1,50 @@
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
1
+ {
2
+ "alora_invocation_tokens": null,
3
+ "alpha_pattern": {},
4
+ "arrow_config": null,
5
+ "auto_mapping": {
6
+ "base_model_class": "Cohere2ForCausalLM",
7
+ "parent_library": "transformers.models.cohere2.modeling_cohere2",
8
+ "unsloth_fixed": true
9
+ },
10
+ "base_model_name_or_path": "CohereLabs/tiny-aya-base",
11
+ "bias": "none",
12
+ "corda_config": null,
13
+ "ensure_weight_tying": false,
14
+ "eva_config": null,
15
+ "exclude_modules": null,
16
+ "fan_in_fan_out": false,
17
+ "inference_mode": true,
18
+ "init_lora_weights": true,
19
+ "layer_replication": null,
20
+ "layers_pattern": null,
21
+ "layers_to_transform": null,
22
+ "loftq_config": {},
23
+ "lora_alpha": 32,
24
+ "lora_bias": false,
25
+ "lora_dropout": 0.0,
26
+ "megatron_config": null,
27
+ "megatron_core": "megatron.core",
28
+ "modules_to_save": null,
29
+ "peft_type": "LORA",
30
+ "peft_version": "0.18.1",
31
+ "qalora_group_size": 16,
32
+ "r": 16,
33
+ "rank_pattern": {},
34
+ "revision": null,
35
+ "target_modules": [
36
+ "up_proj",
37
+ "k_proj",
38
+ "q_proj",
39
+ "gate_proj",
40
+ "v_proj",
41
+ "o_proj",
42
+ "down_proj"
43
+ ],
44
+ "target_parameters": null,
45
+ "task_type": "CAUSAL_LM",
46
+ "trainable_token_indices": null,
47
+ "use_dora": false,
48
+ "use_qalora": false,
49
+ "use_rslora": false
50
+ }
condition-2-es-5k/adapter_model.safetensors ADDED
@@ -0,0 +1,3 @@
 
 
 
 
1
+ version https://git-lfs.github.com/spec/v1
2
+ oid sha256:47972c3cbce9ff16b15a915a0614f82db3108406aab465c5bfd2f4450fa3ffd6
3
+ size 120981200
condition-2-es-5k/training_metrics.json ADDED
@@ -0,0 +1,207 @@
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
1
+ [
2
+ {
3
+ "loss": 1.289,
4
+ "grad_norm": 0.06134699657559395,
5
+ "learning_rate": 0.00012,
6
+ "epoch": 0.03546099290780142,
7
+ "step": 10
8
+ },
9
+ {
10
+ "loss": 1.2668,
11
+ "grad_norm": 0.08268137276172638,
12
+ "learning_rate": 0.00019988926445681492,
13
+ "epoch": 0.07092198581560284,
14
+ "step": 20
15
+ },
16
+ {
17
+ "loss": 1.2005,
18
+ "grad_norm": 0.08327318727970123,
19
+ "learning_rate": 0.0001986463043917528,
20
+ "epoch": 0.10638297872340426,
21
+ "step": 30
22
+ },
23
+ {
24
+ "loss": 1.1013,
25
+ "grad_norm": 0.06584063172340393,
26
+ "learning_rate": 0.00019603921063437793,
27
+ "epoch": 0.14184397163120568,
28
+ "step": 40
29
+ },
30
+ {
31
+ "loss": 1.0508,
32
+ "grad_norm": 0.0763179138302803,
33
+ "learning_rate": 0.0001921040354671897,
34
+ "epoch": 0.1773049645390071,
35
+ "step": 50
36
+ },
37
+ {
38
+ "loss": 1.0812,
39
+ "grad_norm": 0.07645361870527267,
40
+ "learning_rate": 0.00018689519659051467,
41
+ "epoch": 0.2127659574468085,
42
+ "step": 60
43
+ },
44
+ {
45
+ "loss": 1.0913,
46
+ "grad_norm": 0.06947488337755203,
47
+ "learning_rate": 0.00018048472460553257,
48
+ "epoch": 0.24822695035460993,
49
+ "step": 70
50
+ },
51
+ {
52
+ "loss": 1.1241,
53
+ "grad_norm": 0.10595150291919708,
54
+ "learning_rate": 0.00017296126693671884,
55
+ "epoch": 0.28368794326241137,
56
+ "step": 80
57
+ },
58
+ {
59
+ "loss": 1.0546,
60
+ "grad_norm": 0.07947733253240585,
61
+ "learning_rate": 0.00016442886196799464,
62
+ "epoch": 0.3191489361702128,
63
+ "step": 90
64
+ },
65
+ {
66
+ "loss": 1.0223,
67
+ "grad_norm": 0.07233906537294388,
68
+ "learning_rate": 0.00015500550034448413,
69
+ "epoch": 0.3546099290780142,
70
+ "step": 100
71
+ },
72
+ {
73
+ "loss": 1.0651,
74
+ "grad_norm": 0.0785745233297348,
75
+ "learning_rate": 0.00014482149333496454,
76
+ "epoch": 0.3900709219858156,
77
+ "step": 110
78
+ },
79
+ {
80
+ "loss": 1.0878,
81
+ "grad_norm": 0.07600134611129761,
82
+ "learning_rate": 0.0001340176708181637,
83
+ "epoch": 0.425531914893617,
84
+ "step": 120
85
+ },
86
+ {
87
+ "loss": 1.1027,
88
+ "grad_norm": 0.08883103728294373,
89
+ "learning_rate": 0.00012274343381211066,
90
+ "epoch": 0.46099290780141844,
91
+ "step": 130
92
+ },
93
+ {
94
+ "loss": 1.0414,
95
+ "grad_norm": 0.07369714975357056,
96
+ "learning_rate": 0.00011115468847720245,
97
+ "epoch": 0.49645390070921985,
98
+ "step": 140
99
+ },
100
+ {
101
+ "loss": 1.0445,
102
+ "grad_norm": 0.07386789470911026,
103
+ "learning_rate": 9.941169016269379e-05,
104
+ "epoch": 0.5319148936170213,
105
+ "step": 150
106
+ },
107
+ {
108
+ "loss": 1.0534,
109
+ "grad_norm": 0.06285598129034042,
110
+ "learning_rate": 8.767682731028415e-05,
111
+ "epoch": 0.5673758865248227,
112
+ "step": 160
113
+ },
114
+ {
115
+ "loss": 1.1014,
116
+ "grad_norm": 0.07704062014818192,
117
+ "learning_rate": 7.611237586016557e-05,
118
+ "epoch": 0.6028368794326241,
119
+ "step": 170
120
+ },
121
+ {
122
+ "loss": 1.0246,
123
+ "grad_norm": 0.0698208138346672,
124
+ "learning_rate": 6.487825521280109e-05,
125
+ "epoch": 0.6382978723404256,
126
+ "step": 180
127
+ },
128
+ {
129
+ "loss": 1.0135,
130
+ "grad_norm": 0.0755171999335289,
131
+ "learning_rate": 5.4129816778190936e-05,
132
+ "epoch": 0.6737588652482269,
133
+ "step": 190
134
+ },
135
+ {
136
+ "loss": 1.1256,
137
+ "grad_norm": 0.06397763639688492,
138
+ "learning_rate": 4.401569569374668e-05,
139
+ "epoch": 0.7092198581560284,
140
+ "step": 200
141
+ },
142
+ {
143
+ "loss": 1.1244,
144
+ "grad_norm": 0.07543677091598511,
145
+ "learning_rate": 3.467575541836305e-05,
146
+ "epoch": 0.7446808510638298,
147
+ "step": 210
148
+ },
149
+ {
150
+ "loss": 1.0198,
151
+ "grad_norm": 0.06617226451635361,
152
+ "learning_rate": 2.6239153625937784e-05,
153
+ "epoch": 0.7801418439716312,
154
+ "step": 220
155
+ },
156
+ {
157
+ "loss": 1.0186,
158
+ "grad_norm": 0.09004773944616318,
159
+ "learning_rate": 1.882255614419376e-05,
160
+ "epoch": 0.8156028368794326,
161
+ "step": 230
162
+ },
163
+ {
164
+ "loss": 1.0071,
165
+ "grad_norm": 0.07530700415372849,
166
+ "learning_rate": 1.2528523637410838e-05,
167
+ "epoch": 0.851063829787234,
168
+ "step": 240
169
+ },
170
+ {
171
+ "loss": 1.1137,
172
+ "grad_norm": 0.07999535650014877,
173
+ "learning_rate": 7.4440933428779e-06,
174
+ "epoch": 0.8865248226950354,
175
+ "step": 250
176
+ },
177
+ {
178
+ "loss": 1.0764,
179
+ "grad_norm": 0.06478076428174973,
180
+ "learning_rate": 3.6395754735699894e-06,
181
+ "epoch": 0.9219858156028369,
182
+ "step": 260
183
+ },
184
+ {
185
+ "loss": 1.0549,
186
+ "grad_norm": 0.05741119384765625,
187
+ "learning_rate": 1.1675809310361497e-06,
188
+ "epoch": 0.9574468085106383,
189
+ "step": 270
190
+ },
191
+ {
192
+ "loss": 1.0645,
193
+ "grad_norm": 0.05959143117070198,
194
+ "learning_rate": 6.229377380218005e-08,
195
+ "epoch": 0.9929078014184397,
196
+ "step": 280
197
+ },
198
+ {
199
+ "train_runtime": 4558.3131,
200
+ "train_samples_per_second": 0.987,
201
+ "train_steps_per_second": 0.062,
202
+ "total_flos": 7.869358305338982e+16,
203
+ "train_loss": 1.0875117643505123,
204
+ "epoch": 1.0,
205
+ "step": 282
206
+ }
207
+ ]
condition-2-ur-5k/adapter_config.json ADDED
@@ -0,0 +1,50 @@
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
1
+ {
2
+ "alora_invocation_tokens": null,
3
+ "alpha_pattern": {},
4
+ "arrow_config": null,
5
+ "auto_mapping": {
6
+ "base_model_class": "Cohere2ForCausalLM",
7
+ "parent_library": "transformers.models.cohere2.modeling_cohere2",
8
+ "unsloth_fixed": true
9
+ },
10
+ "base_model_name_or_path": "CohereLabs/tiny-aya-base",
11
+ "bias": "none",
12
+ "corda_config": null,
13
+ "ensure_weight_tying": false,
14
+ "eva_config": null,
15
+ "exclude_modules": null,
16
+ "fan_in_fan_out": false,
17
+ "inference_mode": true,
18
+ "init_lora_weights": true,
19
+ "layer_replication": null,
20
+ "layers_pattern": null,
21
+ "layers_to_transform": null,
22
+ "loftq_config": {},
23
+ "lora_alpha": 32,
24
+ "lora_bias": false,
25
+ "lora_dropout": 0.0,
26
+ "megatron_config": null,
27
+ "megatron_core": "megatron.core",
28
+ "modules_to_save": null,
29
+ "peft_type": "LORA",
30
+ "peft_version": "0.18.1",
31
+ "qalora_group_size": 16,
32
+ "r": 16,
33
+ "rank_pattern": {},
34
+ "revision": null,
35
+ "target_modules": [
36
+ "k_proj",
37
+ "v_proj",
38
+ "o_proj",
39
+ "q_proj",
40
+ "up_proj",
41
+ "down_proj",
42
+ "gate_proj"
43
+ ],
44
+ "target_parameters": null,
45
+ "task_type": "CAUSAL_LM",
46
+ "trainable_token_indices": null,
47
+ "use_dora": false,
48
+ "use_qalora": false,
49
+ "use_rslora": false
50
+ }
condition-2-ur-5k/adapter_model.safetensors ADDED
@@ -0,0 +1,3 @@
 
 
 
 
1
+ version https://git-lfs.github.com/spec/v1
2
+ oid sha256:178b078bde04793fc012225157db82f0cba9df8baca7989842583f02fc7c1ee2
3
+ size 120981200
condition-2-zh-5k/adapter_config.json ADDED
@@ -0,0 +1,50 @@
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
1
+ {
2
+ "alora_invocation_tokens": null,
3
+ "alpha_pattern": {},
4
+ "arrow_config": null,
5
+ "auto_mapping": {
6
+ "base_model_class": "Cohere2ForCausalLM",
7
+ "parent_library": "transformers.models.cohere2.modeling_cohere2",
8
+ "unsloth_fixed": true
9
+ },
10
+ "base_model_name_or_path": "CohereLabs/tiny-aya-base",
11
+ "bias": "none",
12
+ "corda_config": null,
13
+ "ensure_weight_tying": false,
14
+ "eva_config": null,
15
+ "exclude_modules": null,
16
+ "fan_in_fan_out": false,
17
+ "inference_mode": true,
18
+ "init_lora_weights": true,
19
+ "layer_replication": null,
20
+ "layers_pattern": null,
21
+ "layers_to_transform": null,
22
+ "loftq_config": {},
23
+ "lora_alpha": 32,
24
+ "lora_bias": false,
25
+ "lora_dropout": 0.0,
26
+ "megatron_config": null,
27
+ "megatron_core": "megatron.core",
28
+ "modules_to_save": null,
29
+ "peft_type": "LORA",
30
+ "peft_version": "0.18.1",
31
+ "qalora_group_size": 16,
32
+ "r": 16,
33
+ "rank_pattern": {},
34
+ "revision": null,
35
+ "target_modules": [
36
+ "v_proj",
37
+ "up_proj",
38
+ "o_proj",
39
+ "k_proj",
40
+ "down_proj",
41
+ "q_proj",
42
+ "gate_proj"
43
+ ],
44
+ "target_parameters": null,
45
+ "task_type": "CAUSAL_LM",
46
+ "trainable_token_indices": null,
47
+ "use_dora": false,
48
+ "use_qalora": false,
49
+ "use_rslora": false
50
+ }
condition-2-zh-5k/adapter_model.safetensors ADDED
@@ -0,0 +1,3 @@
 
 
 
 
1
+ version https://git-lfs.github.com/spec/v1
2
+ oid sha256:d53fe2a8738acf06a0315971137dcdca0d5b26e1b93d410d5ae7cfb34545bf49
3
+ size 120981200
condition-2-zh-5k/training_metrics.json ADDED
@@ -0,0 +1,207 @@
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
1
+ [
2
+ {
3
+ "loss": 1.2934,
4
+ "grad_norm": 0.06680645048618317,
5
+ "learning_rate": 0.00012,
6
+ "epoch": 0.03546099290780142,
7
+ "step": 10
8
+ },
9
+ {
10
+ "loss": 1.2413,
11
+ "grad_norm": 0.07441197335720062,
12
+ "learning_rate": 0.00019988926445681492,
13
+ "epoch": 0.07092198581560284,
14
+ "step": 20
15
+ },
16
+ {
17
+ "loss": 1.1801,
18
+ "grad_norm": 0.07900094240903854,
19
+ "learning_rate": 0.0001986463043917528,
20
+ "epoch": 0.10638297872340426,
21
+ "step": 30
22
+ },
23
+ {
24
+ "loss": 1.0804,
25
+ "grad_norm": 0.06095115840435028,
26
+ "learning_rate": 0.00019603921063437793,
27
+ "epoch": 0.14184397163120568,
28
+ "step": 40
29
+ },
30
+ {
31
+ "loss": 1.0369,
32
+ "grad_norm": 0.06544255465269089,
33
+ "learning_rate": 0.0001921040354671897,
34
+ "epoch": 0.1773049645390071,
35
+ "step": 50
36
+ },
37
+ {
38
+ "loss": 1.0632,
39
+ "grad_norm": 0.06714016199111938,
40
+ "learning_rate": 0.00018689519659051467,
41
+ "epoch": 0.2127659574468085,
42
+ "step": 60
43
+ },
44
+ {
45
+ "loss": 1.0759,
46
+ "grad_norm": 0.061957791447639465,
47
+ "learning_rate": 0.00018048472460553257,
48
+ "epoch": 0.24822695035460993,
49
+ "step": 70
50
+ },
51
+ {
52
+ "loss": 1.1097,
53
+ "grad_norm": 0.09282051771879196,
54
+ "learning_rate": 0.00017296126693671884,
55
+ "epoch": 0.28368794326241137,
56
+ "step": 80
57
+ },
58
+ {
59
+ "loss": 1.0426,
60
+ "grad_norm": 0.07323622703552246,
61
+ "learning_rate": 0.00016442886196799464,
62
+ "epoch": 0.3191489361702128,
63
+ "step": 90
64
+ },
65
+ {
66
+ "loss": 1.0081,
67
+ "grad_norm": 0.05898305028676987,
68
+ "learning_rate": 0.00015500550034448413,
69
+ "epoch": 0.3546099290780142,
70
+ "step": 100
71
+ },
72
+ {
73
+ "loss": 1.0533,
74
+ "grad_norm": 0.0711919292807579,
75
+ "learning_rate": 0.00014482149333496454,
76
+ "epoch": 0.3900709219858156,
77
+ "step": 110
78
+ },
79
+ {
80
+ "loss": 1.0723,
81
+ "grad_norm": 0.06663306057453156,
82
+ "learning_rate": 0.0001340176708181637,
83
+ "epoch": 0.425531914893617,
84
+ "step": 120
85
+ },
86
+ {
87
+ "loss": 1.0894,
88
+ "grad_norm": 0.07218194752931595,
89
+ "learning_rate": 0.00012274343381211066,
90
+ "epoch": 0.46099290780141844,
91
+ "step": 130
92
+ },
93
+ {
94
+ "loss": 1.0284,
95
+ "grad_norm": 0.06280335038900375,
96
+ "learning_rate": 0.00011115468847720245,
97
+ "epoch": 0.49645390070921985,
98
+ "step": 140
99
+ },
100
+ {
101
+ "loss": 1.0339,
102
+ "grad_norm": 0.06357726454734802,
103
+ "learning_rate": 9.941169016269379e-05,
104
+ "epoch": 0.5319148936170213,
105
+ "step": 150
106
+ },
107
+ {
108
+ "loss": 1.0422,
109
+ "grad_norm": 0.05644501745700836,
110
+ "learning_rate": 8.767682731028415e-05,
111
+ "epoch": 0.5673758865248227,
112
+ "step": 160
113
+ },
114
+ {
115
+ "loss": 1.0903,
116
+ "grad_norm": 0.06660397350788116,
117
+ "learning_rate": 7.611237586016557e-05,
118
+ "epoch": 0.6028368794326241,
119
+ "step": 170
120
+ },
121
+ {
122
+ "loss": 1.0108,
123
+ "grad_norm": 0.0638660416007042,
124
+ "learning_rate": 6.487825521280109e-05,
125
+ "epoch": 0.6382978723404256,
126
+ "step": 180
127
+ },
128
+ {
129
+ "loss": 1.0012,
130
+ "grad_norm": 0.06482306122779846,
131
+ "learning_rate": 5.4129816778190936e-05,
132
+ "epoch": 0.6737588652482269,
133
+ "step": 190
134
+ },
135
+ {
136
+ "loss": 1.1147,
137
+ "grad_norm": 0.05531209707260132,
138
+ "learning_rate": 4.401569569374668e-05,
139
+ "epoch": 0.7092198581560284,
140
+ "step": 200
141
+ },
142
+ {
143
+ "loss": 1.1137,
144
+ "grad_norm": 0.06445838510990143,
145
+ "learning_rate": 3.467575541836305e-05,
146
+ "epoch": 0.7446808510638298,
147
+ "step": 210
148
+ },
149
+ {
150
+ "loss": 1.0079,
151
+ "grad_norm": 0.05836805701255798,
152
+ "learning_rate": 2.6239153625937784e-05,
153
+ "epoch": 0.7801418439716312,
154
+ "step": 220
155
+ },
156
+ {
157
+ "loss": 1.0046,
158
+ "grad_norm": 0.0849972814321518,
159
+ "learning_rate": 1.882255614419376e-05,
160
+ "epoch": 0.8156028368794326,
161
+ "step": 230
162
+ },
163
+ {
164
+ "loss": 0.9942,
165
+ "grad_norm": 0.06769896298646927,
166
+ "learning_rate": 1.2528523637410838e-05,
167
+ "epoch": 0.851063829787234,
168
+ "step": 240
169
+ },
170
+ {
171
+ "loss": 1.1035,
172
+ "grad_norm": 0.07061105966567993,
173
+ "learning_rate": 7.4440933428779e-06,
174
+ "epoch": 0.8865248226950354,
175
+ "step": 250
176
+ },
177
+ {
178
+ "loss": 1.0659,
179
+ "grad_norm": 0.05728829279541969,
180
+ "learning_rate": 3.6395754735699894e-06,
181
+ "epoch": 0.9219858156028369,
182
+ "step": 260
183
+ },
184
+ {
185
+ "loss": 1.0457,
186
+ "grad_norm": 0.052473634481430054,
187
+ "learning_rate": 1.1675809310361497e-06,
188
+ "epoch": 0.9574468085106383,
189
+ "step": 270
190
+ },
191
+ {
192
+ "loss": 1.0523,
193
+ "grad_norm": 0.05180735141038895,
194
+ "learning_rate": 6.229377380218005e-08,
195
+ "epoch": 0.9929078014184397,
196
+ "step": 280
197
+ },
198
+ {
199
+ "train_runtime": 5040.1276,
200
+ "train_samples_per_second": 0.893,
201
+ "train_steps_per_second": 0.056,
202
+ "total_flos": 7.869481087716557e+16,
203
+ "train_loss": 1.0745156301674268,
204
+ "epoch": 1.0,
205
+ "step": 282
206
+ }
207
+ ]
condition-3-zh-5k/adapter_config.json ADDED
@@ -0,0 +1,50 @@
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
1
+ {
2
+ "alora_invocation_tokens": null,
3
+ "alpha_pattern": {},
4
+ "arrow_config": null,
5
+ "auto_mapping": {
6
+ "base_model_class": "Cohere2ForCausalLM",
7
+ "parent_library": "transformers.models.cohere2.modeling_cohere2",
8
+ "unsloth_fixed": true
9
+ },
10
+ "base_model_name_or_path": "CohereLabs/tiny-aya-base",
11
+ "bias": "none",
12
+ "corda_config": null,
13
+ "ensure_weight_tying": false,
14
+ "eva_config": null,
15
+ "exclude_modules": null,
16
+ "fan_in_fan_out": false,
17
+ "inference_mode": true,
18
+ "init_lora_weights": true,
19
+ "layer_replication": null,
20
+ "layers_pattern": null,
21
+ "layers_to_transform": null,
22
+ "loftq_config": {},
23
+ "lora_alpha": 32,
24
+ "lora_bias": false,
25
+ "lora_dropout": 0.0,
26
+ "megatron_config": null,
27
+ "megatron_core": "megatron.core",
28
+ "modules_to_save": null,
29
+ "peft_type": "LORA",
30
+ "peft_version": "0.18.1",
31
+ "qalora_group_size": 16,
32
+ "r": 16,
33
+ "rank_pattern": {},
34
+ "revision": null,
35
+ "target_modules": [
36
+ "o_proj",
37
+ "up_proj",
38
+ "gate_proj",
39
+ "q_proj",
40
+ "k_proj",
41
+ "v_proj",
42
+ "down_proj"
43
+ ],
44
+ "target_parameters": null,
45
+ "task_type": "CAUSAL_LM",
46
+ "trainable_token_indices": null,
47
+ "use_dora": false,
48
+ "use_qalora": false,
49
+ "use_rslora": false
50
+ }
condition-3-zh-5k/adapter_model.safetensors ADDED
@@ -0,0 +1,3 @@
 
 
 
 
1
+ version https://git-lfs.github.com/spec/v1
2
+ oid sha256:793136a7a1478440e06697d23124679d66b0171e124e2bcdc81a68861a5f2d73
3
+ size 120981200
condition-3-zh-5k/training_metrics.json ADDED
@@ -0,0 +1,207 @@
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
1
+ [
2
+ {
3
+ "loss": 1.3593,
4
+ "grad_norm": 0.07132186740636826,
5
+ "learning_rate": 0.00012,
6
+ "epoch": 0.03546099290780142,
7
+ "step": 10
8
+ },
9
+ {
10
+ "loss": 1.2212,
11
+ "grad_norm": 0.06576087325811386,
12
+ "learning_rate": 0.00019988926445681492,
13
+ "epoch": 0.07092198581560284,
14
+ "step": 20
15
+ },
16
+ {
17
+ "loss": 1.4382,
18
+ "grad_norm": 0.064721018075943,
19
+ "learning_rate": 0.0001986463043917528,
20
+ "epoch": 0.10638297872340426,
21
+ "step": 30
22
+ },
23
+ {
24
+ "loss": 1.1636,
25
+ "grad_norm": 0.06784526258707047,
26
+ "learning_rate": 0.00019603921063437793,
27
+ "epoch": 0.14184397163120568,
28
+ "step": 40
29
+ },
30
+ {
31
+ "loss": 1.1023,
32
+ "grad_norm": 0.06855457276105881,
33
+ "learning_rate": 0.0001921040354671897,
34
+ "epoch": 0.1773049645390071,
35
+ "step": 50
36
+ },
37
+ {
38
+ "loss": 1.159,
39
+ "grad_norm": 0.0872117206454277,
40
+ "learning_rate": 0.00018689519659051467,
41
+ "epoch": 0.2127659574468085,
42
+ "step": 60
43
+ },
44
+ {
45
+ "loss": 1.2097,
46
+ "grad_norm": 0.06708938628435135,
47
+ "learning_rate": 0.00018048472460553257,
48
+ "epoch": 0.24822695035460993,
49
+ "step": 70
50
+ },
51
+ {
52
+ "loss": 1.1172,
53
+ "grad_norm": 0.07587900757789612,
54
+ "learning_rate": 0.00017296126693671884,
55
+ "epoch": 0.28368794326241137,
56
+ "step": 80
57
+ },
58
+ {
59
+ "loss": 1.0976,
60
+ "grad_norm": 0.07327847182750702,
61
+ "learning_rate": 0.00016442886196799464,
62
+ "epoch": 0.3191489361702128,
63
+ "step": 90
64
+ },
65
+ {
66
+ "loss": 1.1342,
67
+ "grad_norm": 0.07166396826505661,
68
+ "learning_rate": 0.00015500550034448413,
69
+ "epoch": 0.3546099290780142,
70
+ "step": 100
71
+ },
72
+ {
73
+ "loss": 1.1876,
74
+ "grad_norm": 0.08132865279912949,
75
+ "learning_rate": 0.00014482149333496454,
76
+ "epoch": 0.3900709219858156,
77
+ "step": 110
78
+ },
79
+ {
80
+ "loss": 1.2186,
81
+ "grad_norm": 0.08324161171913147,
82
+ "learning_rate": 0.0001340176708181637,
83
+ "epoch": 0.425531914893617,
84
+ "step": 120
85
+ },
86
+ {
87
+ "loss": 1.2672,
88
+ "grad_norm": 0.08210298418998718,
89
+ "learning_rate": 0.00012274343381211066,
90
+ "epoch": 0.46099290780141844,
91
+ "step": 130
92
+ },
93
+ {
94
+ "loss": 1.1838,
95
+ "grad_norm": 0.06662478297948837,
96
+ "learning_rate": 0.00011115468847720245,
97
+ "epoch": 0.49645390070921985,
98
+ "step": 140
99
+ },
100
+ {
101
+ "loss": 1.161,
102
+ "grad_norm": 0.08010110259056091,
103
+ "learning_rate": 9.941169016269379e-05,
104
+ "epoch": 0.5319148936170213,
105
+ "step": 150
106
+ },
107
+ {
108
+ "loss": 1.2318,
109
+ "grad_norm": 0.09503161162137985,
110
+ "learning_rate": 8.767682731028415e-05,
111
+ "epoch": 0.5673758865248227,
112
+ "step": 160
113
+ },
114
+ {
115
+ "loss": 1.1064,
116
+ "grad_norm": 0.06613818556070328,
117
+ "learning_rate": 7.611237586016557e-05,
118
+ "epoch": 0.6028368794326241,
119
+ "step": 170
120
+ },
121
+ {
122
+ "loss": 1.1023,
123
+ "grad_norm": 0.08144712448120117,
124
+ "learning_rate": 6.487825521280109e-05,
125
+ "epoch": 0.6382978723404256,
126
+ "step": 180
127
+ },
128
+ {
129
+ "loss": 1.0777,
130
+ "grad_norm": 0.07447583973407745,
131
+ "learning_rate": 5.4129816778190936e-05,
132
+ "epoch": 0.6737588652482269,
133
+ "step": 190
134
+ },
135
+ {
136
+ "loss": 1.0299,
137
+ "grad_norm": 0.08357255160808563,
138
+ "learning_rate": 4.401569569374668e-05,
139
+ "epoch": 0.7092198581560284,
140
+ "step": 200
141
+ },
142
+ {
143
+ "loss": 1.1504,
144
+ "grad_norm": 0.07654684036970139,
145
+ "learning_rate": 3.467575541836305e-05,
146
+ "epoch": 0.7446808510638298,
147
+ "step": 210
148
+ },
149
+ {
150
+ "loss": 1.0784,
151
+ "grad_norm": 0.07995031774044037,
152
+ "learning_rate": 2.6239153625937784e-05,
153
+ "epoch": 0.7801418439716312,
154
+ "step": 220
155
+ },
156
+ {
157
+ "loss": 1.0866,
158
+ "grad_norm": 0.08516442775726318,
159
+ "learning_rate": 1.882255614419376e-05,
160
+ "epoch": 0.8156028368794326,
161
+ "step": 230
162
+ },
163
+ {
164
+ "loss": 1.1881,
165
+ "grad_norm": 0.06854885816574097,
166
+ "learning_rate": 1.2528523637410838e-05,
167
+ "epoch": 0.851063829787234,
168
+ "step": 240
169
+ },
170
+ {
171
+ "loss": 1.1803,
172
+ "grad_norm": 0.07308827340602875,
173
+ "learning_rate": 7.4440933428779e-06,
174
+ "epoch": 0.8865248226950354,
175
+ "step": 250
176
+ },
177
+ {
178
+ "loss": 1.252,
179
+ "grad_norm": 0.082114577293396,
180
+ "learning_rate": 3.6395754735699894e-06,
181
+ "epoch": 0.9219858156028369,
182
+ "step": 260
183
+ },
184
+ {
185
+ "loss": 0.9981,
186
+ "grad_norm": 0.08104278147220612,
187
+ "learning_rate": 1.1675809310361497e-06,
188
+ "epoch": 0.9574468085106383,
189
+ "step": 270
190
+ },
191
+ {
192
+ "loss": 1.0975,
193
+ "grad_norm": 0.1148340180516243,
194
+ "learning_rate": 6.229377380218005e-08,
195
+ "epoch": 0.9929078014184397,
196
+ "step": 280
197
+ },
198
+ {
199
+ "train_runtime": 5043.2773,
200
+ "train_samples_per_second": 0.892,
201
+ "train_steps_per_second": 0.056,
202
+ "total_flos": 7.853598969744589e+16,
203
+ "train_loss": 1.1630526971309743,
204
+ "epoch": 1.0,
205
+ "step": 282
206
+ }
207
+ ]