Hugging Face's logo

legesher
/
language-decoded-lora

Text Generation
Transformers
Safetensors
lora
aya
tiny-aya
multilingual
code
legesher
tiny-aya-expedition
language-decoded
unsloth
arxiv:2603.11510
arxiv:2211.15533
arxiv:2510.09591
arxiv:1809.05053
arxiv:2308.16884
arxiv:2106.06937
arxiv:2210.03057
Model card Files
xet
 Community
23

open condition-5-zh-5k-c4ai-aya-expanse-32b-seed123

#15
by rafaym - opened
base: refs/heads/main
←
from: refs/pr/15
Discussion Files changed
+257
-0
Files changed (3) hide show
  1. tiny-aya-base/condition-5-zh-5k-c4ai-aya-expanse-32b-seed123/adapter_config.json +54 -0
  2. tiny-aya-base/condition-5-zh-5k-c4ai-aya-expanse-32b-seed123/adapter_model.safetensors +3 -0
  3. tiny-aya-base/condition-5-zh-5k-c4ai-aya-expanse-32b-seed123/training_metrics.json +200 -0
tiny-aya-base/condition-5-zh-5k-c4ai-aya-expanse-32b-seed123/adapter_config.json ADDED
@@ -0,0 +1,54 @@
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
1
+ {
2
+ "alora_invocation_tokens": null,
3
+ "alpha_pattern": {},
4
+ "arrow_config": null,
5
+ "auto_mapping": {
6
+ "base_model_class": "Cohere2ForCausalLM",
7
+ "parent_library": "unsloth_compiled_module_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
+ "lora_ga_config": null,
27
+ "megatron_config": null,
28
+ "megatron_core": "megatron.core",
29
+ "modules_to_save": null,
30
+ "monteclora_config": null,
31
+ "peft_type": "LORA",
32
+ "peft_version": "0.20.0",
33
+ "qalora_group_size": 16,
34
+ "r": 16,
35
+ "rank_pattern": {},
36
+ "revision": null,
37
+ "target_modules": [
38
+ "up_proj",
39
+ "q_proj",
40
+ "o_proj",
41
+ "gate_proj",
42
+ "k_proj",
43
+ "v_proj",
44
+ "down_proj"
45
+ ],
46
+ "target_parameters": null,
47
+ "task_type": "CAUSAL_LM",
48
+ "trainable_token_indices": null,
49
+ "use_bdlora": null,
50
+ "use_dora": false,
51
+ "use_qalora": false,
52
+ "use_rslora": false,
53
+ "velora_config": null
54
+ }
tiny-aya-base/condition-5-zh-5k-c4ai-aya-expanse-32b-seed123/adapter_model.safetensors ADDED
@@ -0,0 +1,3 @@
 
 
 
 
1
+ version https://git-lfs.github.com/spec/v1
2
+ oid sha256:318940f1c4867fa5c8f5a75c6b01dbb90b4e8e1ca41dd8582f90c098535e9569
3
+ size 120981200
tiny-aya-base/condition-5-zh-5k-c4ai-aya-expanse-32b-seed123/training_metrics.json ADDED
@@ -0,0 +1,200 @@
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
1
+ {
2
+ "model_config_name": "tiny-aya-base",
3
+ "condition_name": "condition-5-zh-5k-c4ai-aya-expanse-32b",
4
+ "seed": 123,
5
+ "output_name": "condition-5-zh-5k-c4ai-aya-expanse-32b-seed123",
6
+ "train_result": {
7
+ "train_runtime": 3211.5992,
8
+ "train_samples_per_second": 1.262,
9
+ "train_steps_per_second": 0.079,
10
+ "total_flos": 7.097712441609421e+16,
11
+ "train_loss": 1.146669057410533,
12
+ "epoch": 1.0
13
+ },
14
+ "log_history": [
15
+ {
16
+ "loss": 1.8412,
17
+ "grad_norm": 0.16717135906219482,
18
+ "learning_rate": 0.00013846153846153847,
19
+ "epoch": 0.03937007874015748,
20
+ "step": 10
21
+ },
22
+ {
23
+ "loss": 1.5275,
24
+ "grad_norm": 0.19819766283035278,
25
+ "learning_rate": 0.0001996942849934735,
26
+ "epoch": 0.07874015748031496,
27
+ "step": 20
28
+ },
29
+ {
30
+ "loss": 1.4178,
31
+ "grad_norm": 0.18672265112400055,
32
+ "learning_rate": 0.00019783279156711022,
33
+ "epoch": 0.11811023622047244,
34
+ "step": 30
35
+ },
36
+ {
37
+ "loss": 1.2861,
38
+ "grad_norm": 0.20970652997493744,
39
+ "learning_rate": 0.00019431119516492726,
40
+ "epoch": 0.15748031496062992,
41
+ "step": 40
42
+ },
43
+ {
44
+ "loss": 1.2381,
45
+ "grad_norm": 0.21465376019477844,
46
+ "learning_rate": 0.00018918925297644416,
47
+ "epoch": 0.1968503937007874,
48
+ "step": 50
49
+ },
50
+ {
51
+ "loss": 1.1321,
52
+ "grad_norm": 0.2263137251138687,
53
+ "learning_rate": 0.00018255387810444448,
54
+ "epoch": 0.23622047244094488,
55
+ "step": 60
56
+ },
57
+ {
58
+ "loss": 1.154,
59
+ "grad_norm": 0.22450079023838043,
60
+ "learning_rate": 0.0001745176647557809,
61
+ "epoch": 0.2755905511811024,
62
+ "step": 70
63
+ },
64
+ {
65
+ "loss": 1.1461,
66
+ "grad_norm": 0.23569053411483765,
67
+ "learning_rate": 0.0001652169776552123,
68
+ "epoch": 0.31496062992125984,
69
+ "step": 80
70
+ },
71
+ {
72
+ "loss": 1.0944,
73
+ "grad_norm": 0.24625685811042786,
74
+ "learning_rate": 0.00015480963810258613,
75
+ "epoch": 0.3543307086614173,
76
+ "step": 90
77
+ },
78
+ {
79
+ "loss": 1.0856,
80
+ "grad_norm": 0.2760292589664459,
81
+ "learning_rate": 0.00014347224593814944,
82
+ "epoch": 0.3937007874015748,
83
+ "step": 100
84
+ },
85
+ {
86
+ "loss": 1.0947,
87
+ "grad_norm": 0.2629387676715851,
88
+ "learning_rate": 0.00013139718285896655,
89
+ "epoch": 0.4330708661417323,
90
+ "step": 110
91
+ },
92
+ {
93
+ "loss": 1.0694,
94
+ "grad_norm": 0.23124784231185913,
95
+ "learning_rate": 0.00011878934793650273,
96
+ "epoch": 0.47244094488188976,
97
+ "step": 120
98
+ },
99
+ {
100
+ "loss": 1.0125,
101
+ "grad_norm": 0.2516206204891205,
102
+ "learning_rate": 0.00010586268072965396,
103
+ "epoch": 0.5118110236220472,
104
+ "step": 130
105
+ },
106
+ {
107
+ "loss": 1.0837,
108
+ "grad_norm": 0.24598746001720428,
109
+ "learning_rate": 9.283653099174704e-05,
110
+ "epoch": 0.5511811023622047,
111
+ "step": 140
112
+ },
113
+ {
114
+ "loss": 1.0627,
115
+ "grad_norm": 0.23725923895835876,
116
+ "learning_rate": 7.993193657314875e-05,
117
+ "epoch": 0.5905511811023622,
118
+ "step": 150
119
+ },
120
+ {
121
+ "loss": 1.0748,
122
+ "grad_norm": 0.26233577728271484,
123
+ "learning_rate": 6.736787267892991e-05,
124
+ "epoch": 0.6299212598425197,
125
+ "step": 160
126
+ },
127
+ {
128
+ "loss": 1.0552,
129
+ "grad_norm": 0.2711500823497772,
130
+ "learning_rate": 5.5357536127100904e-05,
131
+ "epoch": 0.6692913385826772,
132
+ "step": 170
133
+ },
134
+ {
135
+ "loss": 1.0402,
136
+ "grad_norm": 0.25870391726493835,
137
+ "learning_rate": 4.410472765901755e-05,
138
+ "epoch": 0.7086614173228346,
139
+ "step": 180
140
+ },
141
+ {
142
+ "loss": 1.0465,
143
+ "grad_norm": 0.2572452127933502,
144
+ "learning_rate": 3.3800393689731146e-05,
145
+ "epoch": 0.7480314960629921,
146
+ "step": 190
147
+ },
148
+ {
149
+ "loss": 1.0791,
150
+ "grad_norm": 0.2789197564125061,
151
+ "learning_rate": 2.461938618055478e-05,
152
+ "epoch": 0.7874015748031497,
153
+ "step": 200
154
+ },
155
+ {
156
+ "loss": 1.0435,
157
+ "grad_norm": 0.29483678936958313,
158
+ "learning_rate": 1.6717495614851652e-05,
159
+ "epoch": 0.8267716535433071,
160
+ "step": 210
161
+ },
162
+ {
163
+ "loss": 1.0383,
164
+ "grad_norm": 0.21859194338321686,
165
+ "learning_rate": 1.0228807423821263e-05,
166
+ "epoch": 0.8661417322834646,
167
+ "step": 220
168
+ },
169
+ {
170
+ "loss": 1.0105,
171
+ "grad_norm": 0.23072956502437592,
172
+ "learning_rate": 5.263426720509468e-06,
173
+ "epoch": 0.905511811023622,
174
+ "step": 230
175
+ },
176
+ {
177
+ "loss": 1.077,
178
+ "grad_norm": 0.28396376967430115,
179
+ "learning_rate": 1.9056099505273427e-06,
180
+ "epoch": 0.9448818897637795,
181
+ "step": 240
182
+ },
183
+ {
184
+ "loss": 0.9869,
185
+ "grad_norm": 0.23137874901294708,
186
+ "learning_rate": 2.1233516308323264e-07,
187
+ "epoch": 0.984251968503937,
188
+ "step": 250
189
+ },
190
+ {
191
+ "train_runtime": 3211.5992,
192
+ "train_samples_per_second": 1.262,
193
+ "train_steps_per_second": 0.079,
194
+ "total_flos": 7.097712441609421e+16,
195
+ "train_loss": 1.146669057410533,
196
+ "epoch": 1.0,
197
+ "step": 254
198
+ }
199
+ ]
200
+ }