Pierluigi Cassotti commited on
Commit
e690c95
1 Parent(s): 72da74c

upload model

Browse files
.gitattributes CHANGED
@@ -33,3 +33,5 @@ saved_model/**/* filter=lfs diff=lfs merge=lfs -text
33
  *.zip filter=lfs diff=lfs merge=lfs -text
34
  *.zst filter=lfs diff=lfs merge=lfs -text
35
  *tfevents* filter=lfs diff=lfs merge=lfs -text
 
 
 
33
  *.zip filter=lfs diff=lfs merge=lfs -text
34
  *.zst filter=lfs diff=lfs merge=lfs -text
35
  *tfevents* filter=lfs diff=lfs merge=lfs -text
36
+ .safetensors filter=lfs diff=lfs merge=lfs -text
37
+ adapter_model.safetensors filter=lfs diff=lfs merge=lfs -text
README.md CHANGED
@@ -1,4 +1,21 @@
1
  ---
2
- language:
3
- - en
4
- ---
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
1
  ---
2
+ library_name: peft
3
+ ---
4
+ ## Training procedure
5
+
6
+
7
+ The following `bitsandbytes` quantization config was used during training:
8
+ - quant_method: bitsandbytes
9
+ - load_in_8bit: False
10
+ - load_in_4bit: True
11
+ - llm_int8_threshold: 6.0
12
+ - llm_int8_skip_modules: None
13
+ - llm_int8_enable_fp32_cpu_offload: False
14
+ - llm_int8_has_fp16_weight: False
15
+ - bnb_4bit_quant_type: nf4
16
+ - bnb_4bit_use_double_quant: False
17
+ - bnb_4bit_compute_dtype: float16
18
+ ### Framework versions
19
+
20
+
21
+ - PEFT 0.5.0
adapter_config.json ADDED
@@ -0,0 +1,24 @@
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
1
+ {
2
+ "auto_mapping": null,
3
+ "base_model_name_or_path": "meta-llama/Llama-2-7b-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": 0.1,
12
+ "modules_to_save": [
13
+ "embed_tokens",
14
+ "lm_head"
15
+ ],
16
+ "peft_type": "LORA",
17
+ "r": 8,
18
+ "revision": null,
19
+ "target_modules": [
20
+ "q_proj",
21
+ "v_proj"
22
+ ],
23
+ "task_type": "CAUSAL_LM"
24
+ }
adapter_model.safetensors ADDED
@@ -0,0 +1,3 @@
 
 
 
 
1
+ version https://git-lfs.github.com/spec/v1
2
+ oid sha256:88e5dcbe6b45623e118bc0f64438327c339ab069fc5b393fbf8fcc23a58982a1
3
+ size 1065502176
added_tokens.json ADDED
@@ -0,0 +1,6 @@
 
 
 
 
 
 
 
1
+ {
2
+ "<|answer|>": 32000,
3
+ "<|end|>": 32001,
4
+ "<|s|>": 32002,
5
+ "[PAD]": 32003
6
+ }
rng_state.pth ADDED
@@ -0,0 +1,3 @@
 
 
 
 
1
+ version https://git-lfs.github.com/spec/v1
2
+ oid sha256:e3245d2bf3edd382271dc07af01b6d1338987fd020fb5bb97bd6ed874f9f69ec
3
+ size 14244
scheduler.pt ADDED
@@ -0,0 +1,3 @@
 
 
 
 
1
+ version https://git-lfs.github.com/spec/v1
2
+ oid sha256:70e59687e7dcfb372e763610b5a1830f3c778014c52fc6b1727a6c0902f13f98
3
+ size 1064
special_tokens_map.json ADDED
@@ -0,0 +1,53 @@
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
1
+ {
2
+ "additional_special_tokens": [
3
+ {
4
+ "content": "<|answer|>",
5
+ "lstrip": false,
6
+ "normalized": true,
7
+ "rstrip": false,
8
+ "single_word": false
9
+ },
10
+ {
11
+ "content": "<|end|>",
12
+ "lstrip": false,
13
+ "normalized": true,
14
+ "rstrip": false,
15
+ "single_word": false
16
+ },
17
+ {
18
+ "content": "<|s|>",
19
+ "lstrip": false,
20
+ "normalized": true,
21
+ "rstrip": false,
22
+ "single_word": false
23
+ }
24
+ ],
25
+ "bos_token": {
26
+ "content": "<s>",
27
+ "lstrip": false,
28
+ "normalized": false,
29
+ "rstrip": false,
30
+ "single_word": false
31
+ },
32
+ "eos_token": {
33
+ "content": "</s>",
34
+ "lstrip": false,
35
+ "normalized": false,
36
+ "rstrip": false,
37
+ "single_word": false
38
+ },
39
+ "pad_token": {
40
+ "content": "[PAD]",
41
+ "lstrip": false,
42
+ "normalized": false,
43
+ "rstrip": false,
44
+ "single_word": false
45
+ },
46
+ "unk_token": {
47
+ "content": "<unk>",
48
+ "lstrip": false,
49
+ "normalized": false,
50
+ "rstrip": false,
51
+ "single_word": false
52
+ }
53
+ }
tokenizer.model ADDED
@@ -0,0 +1,3 @@
 
 
 
 
1
+ version https://git-lfs.github.com/spec/v1
2
+ oid sha256:9e556afd44213b6bd1be2b850ebbbd98f5481437a8021afaf58ee7fb1818d347
3
+ size 499723
tokenizer_config.json ADDED
@@ -0,0 +1,79 @@
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
1
+ {
2
+ "add_bos_token": true,
3
+ "add_eos_token": false,
4
+ "added_tokens_decoder": {
5
+ "0": {
6
+ "content": "<unk>",
7
+ "lstrip": false,
8
+ "normalized": false,
9
+ "rstrip": false,
10
+ "single_word": false,
11
+ "special": true
12
+ },
13
+ "1": {
14
+ "content": "<s>",
15
+ "lstrip": false,
16
+ "normalized": false,
17
+ "rstrip": false,
18
+ "single_word": false,
19
+ "special": true
20
+ },
21
+ "2": {
22
+ "content": "</s>",
23
+ "lstrip": false,
24
+ "normalized": false,
25
+ "rstrip": false,
26
+ "single_word": false,
27
+ "special": true
28
+ },
29
+ "32000": {
30
+ "content": "<|answer|>",
31
+ "lstrip": false,
32
+ "normalized": true,
33
+ "rstrip": false,
34
+ "single_word": false,
35
+ "special": true
36
+ },
37
+ "32001": {
38
+ "content": "<|end|>",
39
+ "lstrip": false,
40
+ "normalized": true,
41
+ "rstrip": false,
42
+ "single_word": false,
43
+ "special": true
44
+ },
45
+ "32002": {
46
+ "content": "<|s|>",
47
+ "lstrip": false,
48
+ "normalized": true,
49
+ "rstrip": false,
50
+ "single_word": false,
51
+ "special": true
52
+ },
53
+ "32003": {
54
+ "content": "[PAD]",
55
+ "lstrip": false,
56
+ "normalized": false,
57
+ "rstrip": false,
58
+ "single_word": false,
59
+ "special": true
60
+ }
61
+ },
62
+ "additional_special_tokens": [
63
+ "<|answer|>",
64
+ "<|end|>",
65
+ "<|s|>"
66
+ ],
67
+ "bos_token": "<s>",
68
+ "clean_up_tokenization_spaces": false,
69
+ "eos_token": "</s>",
70
+ "legacy": false,
71
+ "model_max_length": 1000000000000000019884624838656,
72
+ "pad_token": "[PAD]",
73
+ "padding_side": "left",
74
+ "sp_model_kwargs": {},
75
+ "spaces_between_special_tokens": false,
76
+ "tokenizer_class": "LlamaTokenizer",
77
+ "unk_token": "<unk>",
78
+ "use_default_system_prompt": false
79
+ }
trainer_state.json ADDED
@@ -0,0 +1,1419 @@
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
1
+ {
2
+ "best_metric": null,
3
+ "best_model_checkpoint": null,
4
+ "epoch": 9.995713673381912,
5
+ "eval_steps": 500,
6
+ "global_step": 11660,
7
+ "is_hyper_param_search": false,
8
+ "is_local_process_zero": true,
9
+ "is_world_process_zero": true,
10
+ "log_history": [
11
+ {
12
+ "epoch": 0.04,
13
+ "learning_rate": 0.00019914236706689538,
14
+ "loss": 2.573,
15
+ "step": 50
16
+ },
17
+ {
18
+ "epoch": 0.09,
19
+ "learning_rate": 0.00019828473413379075,
20
+ "loss": 1.8291,
21
+ "step": 100
22
+ },
23
+ {
24
+ "epoch": 0.13,
25
+ "learning_rate": 0.00019742710120068612,
26
+ "loss": 1.7257,
27
+ "step": 150
28
+ },
29
+ {
30
+ "epoch": 0.17,
31
+ "learning_rate": 0.00019656946826758148,
32
+ "loss": 1.6338,
33
+ "step": 200
34
+ },
35
+ {
36
+ "epoch": 0.21,
37
+ "learning_rate": 0.00019571183533447685,
38
+ "loss": 1.6359,
39
+ "step": 250
40
+ },
41
+ {
42
+ "epoch": 0.26,
43
+ "learning_rate": 0.00019485420240137222,
44
+ "loss": 1.5935,
45
+ "step": 300
46
+ },
47
+ {
48
+ "epoch": 0.3,
49
+ "learning_rate": 0.0001939965694682676,
50
+ "loss": 1.5494,
51
+ "step": 350
52
+ },
53
+ {
54
+ "epoch": 0.34,
55
+ "learning_rate": 0.00019313893653516296,
56
+ "loss": 1.559,
57
+ "step": 400
58
+ },
59
+ {
60
+ "epoch": 0.39,
61
+ "learning_rate": 0.00019228130360205833,
62
+ "loss": 1.5403,
63
+ "step": 450
64
+ },
65
+ {
66
+ "epoch": 0.43,
67
+ "learning_rate": 0.0001914236706689537,
68
+ "loss": 1.528,
69
+ "step": 500
70
+ },
71
+ {
72
+ "epoch": 0.47,
73
+ "learning_rate": 0.00019056603773584906,
74
+ "loss": 1.4899,
75
+ "step": 550
76
+ },
77
+ {
78
+ "epoch": 0.51,
79
+ "learning_rate": 0.00018970840480274443,
80
+ "loss": 1.4545,
81
+ "step": 600
82
+ },
83
+ {
84
+ "epoch": 0.56,
85
+ "learning_rate": 0.0001888507718696398,
86
+ "loss": 1.4658,
87
+ "step": 650
88
+ },
89
+ {
90
+ "epoch": 0.6,
91
+ "learning_rate": 0.00018799313893653517,
92
+ "loss": 1.4666,
93
+ "step": 700
94
+ },
95
+ {
96
+ "epoch": 0.64,
97
+ "learning_rate": 0.00018713550600343054,
98
+ "loss": 1.4603,
99
+ "step": 750
100
+ },
101
+ {
102
+ "epoch": 0.69,
103
+ "learning_rate": 0.0001862778730703259,
104
+ "loss": 1.4256,
105
+ "step": 800
106
+ },
107
+ {
108
+ "epoch": 0.73,
109
+ "learning_rate": 0.00018542024013722128,
110
+ "loss": 1.4422,
111
+ "step": 850
112
+ },
113
+ {
114
+ "epoch": 0.77,
115
+ "learning_rate": 0.00018456260720411664,
116
+ "loss": 1.4255,
117
+ "step": 900
118
+ },
119
+ {
120
+ "epoch": 0.81,
121
+ "learning_rate": 0.000183704974271012,
122
+ "loss": 1.3872,
123
+ "step": 950
124
+ },
125
+ {
126
+ "epoch": 0.86,
127
+ "learning_rate": 0.00018284734133790738,
128
+ "loss": 1.3886,
129
+ "step": 1000
130
+ },
131
+ {
132
+ "epoch": 0.9,
133
+ "learning_rate": 0.00018198970840480275,
134
+ "loss": 1.4147,
135
+ "step": 1050
136
+ },
137
+ {
138
+ "epoch": 0.94,
139
+ "learning_rate": 0.00018113207547169812,
140
+ "loss": 1.405,
141
+ "step": 1100
142
+ },
143
+ {
144
+ "epoch": 0.99,
145
+ "learning_rate": 0.0001802744425385935,
146
+ "loss": 1.4027,
147
+ "step": 1150
148
+ },
149
+ {
150
+ "epoch": 1.03,
151
+ "learning_rate": 0.00017941680960548886,
152
+ "loss": 1.1787,
153
+ "step": 1200
154
+ },
155
+ {
156
+ "epoch": 1.07,
157
+ "learning_rate": 0.00017855917667238422,
158
+ "loss": 1.058,
159
+ "step": 1250
160
+ },
161
+ {
162
+ "epoch": 1.11,
163
+ "learning_rate": 0.00017770154373927962,
164
+ "loss": 1.0577,
165
+ "step": 1300
166
+ },
167
+ {
168
+ "epoch": 1.16,
169
+ "learning_rate": 0.00017684391080617496,
170
+ "loss": 1.0675,
171
+ "step": 1350
172
+ },
173
+ {
174
+ "epoch": 1.2,
175
+ "learning_rate": 0.00017598627787307033,
176
+ "loss": 1.0459,
177
+ "step": 1400
178
+ },
179
+ {
180
+ "epoch": 1.24,
181
+ "learning_rate": 0.0001751286449399657,
182
+ "loss": 1.072,
183
+ "step": 1450
184
+ },
185
+ {
186
+ "epoch": 1.29,
187
+ "learning_rate": 0.00017427101200686107,
188
+ "loss": 1.0654,
189
+ "step": 1500
190
+ },
191
+ {
192
+ "epoch": 1.33,
193
+ "learning_rate": 0.00017341337907375644,
194
+ "loss": 1.0594,
195
+ "step": 1550
196
+ },
197
+ {
198
+ "epoch": 1.37,
199
+ "learning_rate": 0.00017255574614065183,
200
+ "loss": 1.0594,
201
+ "step": 1600
202
+ },
203
+ {
204
+ "epoch": 1.41,
205
+ "learning_rate": 0.00017169811320754717,
206
+ "loss": 1.0851,
207
+ "step": 1650
208
+ },
209
+ {
210
+ "epoch": 1.46,
211
+ "learning_rate": 0.00017084048027444254,
212
+ "loss": 1.066,
213
+ "step": 1700
214
+ },
215
+ {
216
+ "epoch": 1.5,
217
+ "learning_rate": 0.0001699828473413379,
218
+ "loss": 1.0591,
219
+ "step": 1750
220
+ },
221
+ {
222
+ "epoch": 1.54,
223
+ "learning_rate": 0.00016912521440823328,
224
+ "loss": 1.0692,
225
+ "step": 1800
226
+ },
227
+ {
228
+ "epoch": 1.59,
229
+ "learning_rate": 0.00016826758147512865,
230
+ "loss": 1.0772,
231
+ "step": 1850
232
+ },
233
+ {
234
+ "epoch": 1.63,
235
+ "learning_rate": 0.00016740994854202404,
236
+ "loss": 1.075,
237
+ "step": 1900
238
+ },
239
+ {
240
+ "epoch": 1.67,
241
+ "learning_rate": 0.00016655231560891938,
242
+ "loss": 1.0607,
243
+ "step": 1950
244
+ },
245
+ {
246
+ "epoch": 1.71,
247
+ "learning_rate": 0.00016569468267581475,
248
+ "loss": 1.0682,
249
+ "step": 2000
250
+ },
251
+ {
252
+ "epoch": 1.76,
253
+ "learning_rate": 0.00016483704974271012,
254
+ "loss": 1.0616,
255
+ "step": 2050
256
+ },
257
+ {
258
+ "epoch": 1.8,
259
+ "learning_rate": 0.0001639794168096055,
260
+ "loss": 1.0655,
261
+ "step": 2100
262
+ },
263
+ {
264
+ "epoch": 1.84,
265
+ "learning_rate": 0.00016312178387650086,
266
+ "loss": 1.0555,
267
+ "step": 2150
268
+ },
269
+ {
270
+ "epoch": 1.89,
271
+ "learning_rate": 0.00016226415094339625,
272
+ "loss": 1.0761,
273
+ "step": 2200
274
+ },
275
+ {
276
+ "epoch": 1.93,
277
+ "learning_rate": 0.0001614065180102916,
278
+ "loss": 1.0776,
279
+ "step": 2250
280
+ },
281
+ {
282
+ "epoch": 1.97,
283
+ "learning_rate": 0.00016054888507718696,
284
+ "loss": 1.0574,
285
+ "step": 2300
286
+ },
287
+ {
288
+ "epoch": 2.01,
289
+ "learning_rate": 0.00015969125214408233,
290
+ "loss": 0.9443,
291
+ "step": 2350
292
+ },
293
+ {
294
+ "epoch": 2.06,
295
+ "learning_rate": 0.0001588336192109777,
296
+ "loss": 0.7435,
297
+ "step": 2400
298
+ },
299
+ {
300
+ "epoch": 2.1,
301
+ "learning_rate": 0.00015797598627787307,
302
+ "loss": 0.7457,
303
+ "step": 2450
304
+ },
305
+ {
306
+ "epoch": 2.14,
307
+ "learning_rate": 0.00015711835334476847,
308
+ "loss": 0.7553,
309
+ "step": 2500
310
+ },
311
+ {
312
+ "epoch": 2.19,
313
+ "learning_rate": 0.0001562607204116638,
314
+ "loss": 0.7452,
315
+ "step": 2550
316
+ },
317
+ {
318
+ "epoch": 2.23,
319
+ "learning_rate": 0.00015540308747855918,
320
+ "loss": 0.7736,
321
+ "step": 2600
322
+ },
323
+ {
324
+ "epoch": 2.27,
325
+ "learning_rate": 0.00015454545454545454,
326
+ "loss": 0.7649,
327
+ "step": 2650
328
+ },
329
+ {
330
+ "epoch": 2.31,
331
+ "learning_rate": 0.0001536878216123499,
332
+ "loss": 0.7557,
333
+ "step": 2700
334
+ },
335
+ {
336
+ "epoch": 2.36,
337
+ "learning_rate": 0.0001528301886792453,
338
+ "loss": 0.7556,
339
+ "step": 2750
340
+ },
341
+ {
342
+ "epoch": 2.4,
343
+ "learning_rate": 0.00015197255574614068,
344
+ "loss": 0.7751,
345
+ "step": 2800
346
+ },
347
+ {
348
+ "epoch": 2.44,
349
+ "learning_rate": 0.00015111492281303602,
350
+ "loss": 0.7613,
351
+ "step": 2850
352
+ },
353
+ {
354
+ "epoch": 2.49,
355
+ "learning_rate": 0.0001502572898799314,
356
+ "loss": 0.7751,
357
+ "step": 2900
358
+ },
359
+ {
360
+ "epoch": 2.53,
361
+ "learning_rate": 0.00014939965694682676,
362
+ "loss": 0.774,
363
+ "step": 2950
364
+ },
365
+ {
366
+ "epoch": 2.57,
367
+ "learning_rate": 0.00014854202401372212,
368
+ "loss": 0.7803,
369
+ "step": 3000
370
+ },
371
+ {
372
+ "epoch": 2.61,
373
+ "learning_rate": 0.00014768439108061752,
374
+ "loss": 0.7691,
375
+ "step": 3050
376
+ },
377
+ {
378
+ "epoch": 2.66,
379
+ "learning_rate": 0.0001468267581475129,
380
+ "loss": 0.7684,
381
+ "step": 3100
382
+ },
383
+ {
384
+ "epoch": 2.7,
385
+ "learning_rate": 0.00014596912521440823,
386
+ "loss": 0.7811,
387
+ "step": 3150
388
+ },
389
+ {
390
+ "epoch": 2.74,
391
+ "learning_rate": 0.0001451114922813036,
392
+ "loss": 0.7631,
393
+ "step": 3200
394
+ },
395
+ {
396
+ "epoch": 2.79,
397
+ "learning_rate": 0.00014425385934819897,
398
+ "loss": 0.7782,
399
+ "step": 3250
400
+ },
401
+ {
402
+ "epoch": 2.83,
403
+ "learning_rate": 0.00014339622641509434,
404
+ "loss": 0.7686,
405
+ "step": 3300
406
+ },
407
+ {
408
+ "epoch": 2.87,
409
+ "learning_rate": 0.00014253859348198973,
410
+ "loss": 0.7919,
411
+ "step": 3350
412
+ },
413
+ {
414
+ "epoch": 2.91,
415
+ "learning_rate": 0.0001416809605488851,
416
+ "loss": 0.7768,
417
+ "step": 3400
418
+ },
419
+ {
420
+ "epoch": 2.96,
421
+ "learning_rate": 0.00014082332761578044,
422
+ "loss": 0.7837,
423
+ "step": 3450
424
+ },
425
+ {
426
+ "epoch": 3.0,
427
+ "learning_rate": 0.0001399656946826758,
428
+ "loss": 0.7739,
429
+ "step": 3500
430
+ },
431
+ {
432
+ "epoch": 3.04,
433
+ "learning_rate": 0.00013910806174957118,
434
+ "loss": 0.4864,
435
+ "step": 3550
436
+ },
437
+ {
438
+ "epoch": 3.09,
439
+ "learning_rate": 0.00013825042881646655,
440
+ "loss": 0.4876,
441
+ "step": 3600
442
+ },
443
+ {
444
+ "epoch": 3.13,
445
+ "learning_rate": 0.00013739279588336194,
446
+ "loss": 0.4899,
447
+ "step": 3650
448
+ },
449
+ {
450
+ "epoch": 3.17,
451
+ "learning_rate": 0.0001365351629502573,
452
+ "loss": 0.4833,
453
+ "step": 3700
454
+ },
455
+ {
456
+ "epoch": 3.21,
457
+ "learning_rate": 0.00013567753001715265,
458
+ "loss": 0.5126,
459
+ "step": 3750
460
+ },
461
+ {
462
+ "epoch": 3.26,
463
+ "learning_rate": 0.00013481989708404802,
464
+ "loss": 0.4891,
465
+ "step": 3800
466
+ },
467
+ {
468
+ "epoch": 3.3,
469
+ "learning_rate": 0.0001339622641509434,
470
+ "loss": 0.5,
471
+ "step": 3850
472
+ },
473
+ {
474
+ "epoch": 3.34,
475
+ "learning_rate": 0.00013310463121783879,
476
+ "loss": 0.5078,
477
+ "step": 3900
478
+ },
479
+ {
480
+ "epoch": 3.39,
481
+ "learning_rate": 0.00013224699828473415,
482
+ "loss": 0.4931,
483
+ "step": 3950
484
+ },
485
+ {
486
+ "epoch": 3.43,
487
+ "learning_rate": 0.00013138936535162952,
488
+ "loss": 0.5062,
489
+ "step": 4000
490
+ },
491
+ {
492
+ "epoch": 3.47,
493
+ "learning_rate": 0.00013053173241852486,
494
+ "loss": 0.509,
495
+ "step": 4050
496
+ },
497
+ {
498
+ "epoch": 3.51,
499
+ "learning_rate": 0.00012967409948542023,
500
+ "loss": 0.5097,
501
+ "step": 4100
502
+ },
503
+ {
504
+ "epoch": 3.56,
505
+ "learning_rate": 0.0001288164665523156,
506
+ "loss": 0.5225,
507
+ "step": 4150
508
+ },
509
+ {
510
+ "epoch": 3.6,
511
+ "learning_rate": 0.000127958833619211,
512
+ "loss": 0.5133,
513
+ "step": 4200
514
+ },
515
+ {
516
+ "epoch": 3.64,
517
+ "learning_rate": 0.00012710120068610637,
518
+ "loss": 0.5183,
519
+ "step": 4250
520
+ },
521
+ {
522
+ "epoch": 3.69,
523
+ "learning_rate": 0.00012624356775300173,
524
+ "loss": 0.5259,
525
+ "step": 4300
526
+ },
527
+ {
528
+ "epoch": 3.73,
529
+ "learning_rate": 0.00012538593481989708,
530
+ "loss": 0.5172,
531
+ "step": 4350
532
+ },
533
+ {
534
+ "epoch": 3.77,
535
+ "learning_rate": 0.00012452830188679244,
536
+ "loss": 0.5135,
537
+ "step": 4400
538
+ },
539
+ {
540
+ "epoch": 3.81,
541
+ "learning_rate": 0.0001236706689536878,
542
+ "loss": 0.5218,
543
+ "step": 4450
544
+ },
545
+ {
546
+ "epoch": 3.86,
547
+ "learning_rate": 0.0001228130360205832,
548
+ "loss": 0.5252,
549
+ "step": 4500
550
+ },
551
+ {
552
+ "epoch": 3.9,
553
+ "learning_rate": 0.00012195540308747856,
554
+ "loss": 0.5139,
555
+ "step": 4550
556
+ },
557
+ {
558
+ "epoch": 3.94,
559
+ "learning_rate": 0.00012109777015437393,
560
+ "loss": 0.5267,
561
+ "step": 4600
562
+ },
563
+ {
564
+ "epoch": 3.99,
565
+ "learning_rate": 0.0001202401372212693,
566
+ "loss": 0.5193,
567
+ "step": 4650
568
+ },
569
+ {
570
+ "epoch": 4.03,
571
+ "learning_rate": 0.00011938250428816467,
572
+ "loss": 0.3797,
573
+ "step": 4700
574
+ },
575
+ {
576
+ "epoch": 4.07,
577
+ "learning_rate": 0.00011852487135506002,
578
+ "loss": 0.3031,
579
+ "step": 4750
580
+ },
581
+ {
582
+ "epoch": 4.11,
583
+ "learning_rate": 0.00011766723842195542,
584
+ "loss": 0.2867,
585
+ "step": 4800
586
+ },
587
+ {
588
+ "epoch": 4.16,
589
+ "learning_rate": 0.00011680960548885078,
590
+ "loss": 0.2986,
591
+ "step": 4850
592
+ },
593
+ {
594
+ "epoch": 4.2,
595
+ "learning_rate": 0.00011595197255574614,
596
+ "loss": 0.3021,
597
+ "step": 4900
598
+ },
599
+ {
600
+ "epoch": 4.24,
601
+ "learning_rate": 0.00011509433962264151,
602
+ "loss": 0.2983,
603
+ "step": 4950
604
+ },
605
+ {
606
+ "epoch": 4.29,
607
+ "learning_rate": 0.00011423670668953688,
608
+ "loss": 0.3014,
609
+ "step": 5000
610
+ },
611
+ {
612
+ "epoch": 4.33,
613
+ "learning_rate": 0.00011337907375643224,
614
+ "loss": 0.2966,
615
+ "step": 5050
616
+ },
617
+ {
618
+ "epoch": 4.37,
619
+ "learning_rate": 0.00011252144082332763,
620
+ "loss": 0.3014,
621
+ "step": 5100
622
+ },
623
+ {
624
+ "epoch": 4.41,
625
+ "learning_rate": 0.00011166380789022299,
626
+ "loss": 0.3105,
627
+ "step": 5150
628
+ },
629
+ {
630
+ "epoch": 4.46,
631
+ "learning_rate": 0.00011080617495711836,
632
+ "loss": 0.3115,
633
+ "step": 5200
634
+ },
635
+ {
636
+ "epoch": 4.5,
637
+ "learning_rate": 0.00010994854202401372,
638
+ "loss": 0.3139,
639
+ "step": 5250
640
+ },
641
+ {
642
+ "epoch": 4.54,
643
+ "learning_rate": 0.00010909090909090909,
644
+ "loss": 0.3089,
645
+ "step": 5300
646
+ },
647
+ {
648
+ "epoch": 4.59,
649
+ "learning_rate": 0.00010823327615780447,
650
+ "loss": 0.3069,
651
+ "step": 5350
652
+ },
653
+ {
654
+ "epoch": 4.63,
655
+ "learning_rate": 0.00010737564322469984,
656
+ "loss": 0.3181,
657
+ "step": 5400
658
+ },
659
+ {
660
+ "epoch": 4.67,
661
+ "learning_rate": 0.0001065180102915952,
662
+ "loss": 0.3129,
663
+ "step": 5450
664
+ },
665
+ {
666
+ "epoch": 4.71,
667
+ "learning_rate": 0.00010566037735849057,
668
+ "loss": 0.3199,
669
+ "step": 5500
670
+ },
671
+ {
672
+ "epoch": 4.76,
673
+ "learning_rate": 0.00010480274442538594,
674
+ "loss": 0.3079,
675
+ "step": 5550
676
+ },
677
+ {
678
+ "epoch": 4.8,
679
+ "learning_rate": 0.0001039451114922813,
680
+ "loss": 0.32,
681
+ "step": 5600
682
+ },
683
+ {
684
+ "epoch": 4.84,
685
+ "learning_rate": 0.00010308747855917669,
686
+ "loss": 0.3157,
687
+ "step": 5650
688
+ },
689
+ {
690
+ "epoch": 4.89,
691
+ "learning_rate": 0.00010222984562607205,
692
+ "loss": 0.3181,
693
+ "step": 5700
694
+ },
695
+ {
696
+ "epoch": 4.93,
697
+ "learning_rate": 0.00010137221269296742,
698
+ "loss": 0.3217,
699
+ "step": 5750
700
+ },
701
+ {
702
+ "epoch": 4.97,
703
+ "learning_rate": 0.00010051457975986278,
704
+ "loss": 0.3132,
705
+ "step": 5800
706
+ },
707
+ {
708
+ "epoch": 5.02,
709
+ "learning_rate": 9.965694682675816e-05,
710
+ "loss": 0.2703,
711
+ "step": 5850
712
+ },
713
+ {
714
+ "epoch": 5.06,
715
+ "learning_rate": 9.879931389365353e-05,
716
+ "loss": 0.1618,
717
+ "step": 5900
718
+ },
719
+ {
720
+ "epoch": 5.1,
721
+ "learning_rate": 9.794168096054888e-05,
722
+ "loss": 0.1767,
723
+ "step": 5950
724
+ },
725
+ {
726
+ "epoch": 5.14,
727
+ "learning_rate": 9.708404802744427e-05,
728
+ "loss": 0.1647,
729
+ "step": 6000
730
+ },
731
+ {
732
+ "epoch": 5.19,
733
+ "learning_rate": 9.622641509433963e-05,
734
+ "loss": 0.1725,
735
+ "step": 6050
736
+ },
737
+ {
738
+ "epoch": 5.23,
739
+ "learning_rate": 9.536878216123499e-05,
740
+ "loss": 0.1732,
741
+ "step": 6100
742
+ },
743
+ {
744
+ "epoch": 5.27,
745
+ "learning_rate": 9.451114922813037e-05,
746
+ "loss": 0.168,
747
+ "step": 6150
748
+ },
749
+ {
750
+ "epoch": 5.32,
751
+ "learning_rate": 9.365351629502574e-05,
752
+ "loss": 0.1769,
753
+ "step": 6200
754
+ },
755
+ {
756
+ "epoch": 5.36,
757
+ "learning_rate": 9.27958833619211e-05,
758
+ "loss": 0.1718,
759
+ "step": 6250
760
+ },
761
+ {
762
+ "epoch": 5.4,
763
+ "learning_rate": 9.193825042881648e-05,
764
+ "loss": 0.1734,
765
+ "step": 6300
766
+ },
767
+ {
768
+ "epoch": 5.44,
769
+ "learning_rate": 9.108061749571185e-05,
770
+ "loss": 0.1815,
771
+ "step": 6350
772
+ },
773
+ {
774
+ "epoch": 5.49,
775
+ "learning_rate": 9.02229845626072e-05,
776
+ "loss": 0.1761,
777
+ "step": 6400
778
+ },
779
+ {
780
+ "epoch": 5.53,
781
+ "learning_rate": 8.936535162950258e-05,
782
+ "loss": 0.1783,
783
+ "step": 6450
784
+ },
785
+ {
786
+ "epoch": 5.57,
787
+ "learning_rate": 8.850771869639795e-05,
788
+ "loss": 0.1764,
789
+ "step": 6500
790
+ },
791
+ {
792
+ "epoch": 5.62,
793
+ "learning_rate": 8.765008576329331e-05,
794
+ "loss": 0.1763,
795
+ "step": 6550
796
+ },
797
+ {
798
+ "epoch": 5.66,
799
+ "learning_rate": 8.679245283018869e-05,
800
+ "loss": 0.1684,
801
+ "step": 6600
802
+ },
803
+ {
804
+ "epoch": 5.7,
805
+ "learning_rate": 8.593481989708406e-05,
806
+ "loss": 0.174,
807
+ "step": 6650
808
+ },
809
+ {
810
+ "epoch": 5.74,
811
+ "learning_rate": 8.507718696397941e-05,
812
+ "loss": 0.1724,
813
+ "step": 6700
814
+ },
815
+ {
816
+ "epoch": 5.79,
817
+ "learning_rate": 8.42195540308748e-05,
818
+ "loss": 0.177,
819
+ "step": 6750
820
+ },
821
+ {
822
+ "epoch": 5.83,
823
+ "learning_rate": 8.336192109777016e-05,
824
+ "loss": 0.1753,
825
+ "step": 6800
826
+ },
827
+ {
828
+ "epoch": 5.87,
829
+ "learning_rate": 8.250428816466552e-05,
830
+ "loss": 0.1765,
831
+ "step": 6850
832
+ },
833
+ {
834
+ "epoch": 5.92,
835
+ "learning_rate": 8.16466552315609e-05,
836
+ "loss": 0.1748,
837
+ "step": 6900
838
+ },
839
+ {
840
+ "epoch": 5.96,
841
+ "learning_rate": 8.078902229845627e-05,
842
+ "loss": 0.1781,
843
+ "step": 6950
844
+ },
845
+ {
846
+ "epoch": 6.0,
847
+ "learning_rate": 7.993138936535164e-05,
848
+ "loss": 0.1838,
849
+ "step": 7000
850
+ },
851
+ {
852
+ "epoch": 6.04,
853
+ "learning_rate": 7.9073756432247e-05,
854
+ "loss": 0.0912,
855
+ "step": 7050
856
+ },
857
+ {
858
+ "epoch": 6.09,
859
+ "learning_rate": 7.821612349914237e-05,
860
+ "loss": 0.0874,
861
+ "step": 7100
862
+ },
863
+ {
864
+ "epoch": 6.13,
865
+ "learning_rate": 7.735849056603774e-05,
866
+ "loss": 0.0879,
867
+ "step": 7150
868
+ },
869
+ {
870
+ "epoch": 6.17,
871
+ "learning_rate": 7.650085763293311e-05,
872
+ "loss": 0.0922,
873
+ "step": 7200
874
+ },
875
+ {
876
+ "epoch": 6.22,
877
+ "learning_rate": 7.564322469982848e-05,
878
+ "loss": 0.0979,
879
+ "step": 7250
880
+ },
881
+ {
882
+ "epoch": 6.26,
883
+ "learning_rate": 7.478559176672385e-05,
884
+ "loss": 0.0853,
885
+ "step": 7300
886
+ },
887
+ {
888
+ "epoch": 6.3,
889
+ "learning_rate": 7.392795883361922e-05,
890
+ "loss": 0.0866,
891
+ "step": 7350
892
+ },
893
+ {
894
+ "epoch": 6.34,
895
+ "learning_rate": 7.307032590051459e-05,
896
+ "loss": 0.0876,
897
+ "step": 7400
898
+ },
899
+ {
900
+ "epoch": 6.39,
901
+ "learning_rate": 7.221269296740995e-05,
902
+ "loss": 0.0895,
903
+ "step": 7450
904
+ },
905
+ {
906
+ "epoch": 6.43,
907
+ "learning_rate": 7.135506003430532e-05,
908
+ "loss": 0.0925,
909
+ "step": 7500
910
+ },
911
+ {
912
+ "epoch": 6.47,
913
+ "learning_rate": 7.049742710120069e-05,
914
+ "loss": 0.0915,
915
+ "step": 7550
916
+ },
917
+ {
918
+ "epoch": 6.52,
919
+ "learning_rate": 6.963979416809606e-05,
920
+ "loss": 0.0844,
921
+ "step": 7600
922
+ },
923
+ {
924
+ "epoch": 6.56,
925
+ "learning_rate": 6.878216123499143e-05,
926
+ "loss": 0.0949,
927
+ "step": 7650
928
+ },
929
+ {
930
+ "epoch": 6.6,
931
+ "learning_rate": 6.79245283018868e-05,
932
+ "loss": 0.0893,
933
+ "step": 7700
934
+ },
935
+ {
936
+ "epoch": 6.64,
937
+ "learning_rate": 6.706689536878217e-05,
938
+ "loss": 0.0921,
939
+ "step": 7750
940
+ },
941
+ {
942
+ "epoch": 6.69,
943
+ "learning_rate": 6.620926243567753e-05,
944
+ "loss": 0.0879,
945
+ "step": 7800
946
+ },
947
+ {
948
+ "epoch": 6.73,
949
+ "learning_rate": 6.53516295025729e-05,
950
+ "loss": 0.0886,
951
+ "step": 7850
952
+ },
953
+ {
954
+ "epoch": 6.77,
955
+ "learning_rate": 6.449399656946827e-05,
956
+ "loss": 0.0912,
957
+ "step": 7900
958
+ },
959
+ {
960
+ "epoch": 6.82,
961
+ "learning_rate": 6.363636363636364e-05,
962
+ "loss": 0.0898,
963
+ "step": 7950
964
+ },
965
+ {
966
+ "epoch": 6.86,
967
+ "learning_rate": 6.277873070325901e-05,
968
+ "loss": 0.0936,
969
+ "step": 8000
970
+ },
971
+ {
972
+ "epoch": 6.9,
973
+ "learning_rate": 6.192109777015438e-05,
974
+ "loss": 0.0903,
975
+ "step": 8050
976
+ },
977
+ {
978
+ "epoch": 6.94,
979
+ "learning_rate": 6.106346483704975e-05,
980
+ "loss": 0.0876,
981
+ "step": 8100
982
+ },
983
+ {
984
+ "epoch": 6.99,
985
+ "learning_rate": 6.020583190394511e-05,
986
+ "loss": 0.0887,
987
+ "step": 8150
988
+ },
989
+ {
990
+ "epoch": 7.03,
991
+ "learning_rate": 5.9348198970840484e-05,
992
+ "loss": 0.0559,
993
+ "step": 8200
994
+ },
995
+ {
996
+ "epoch": 7.07,
997
+ "learning_rate": 5.849056603773585e-05,
998
+ "loss": 0.042,
999
+ "step": 8250
1000
+ },
1001
+ {
1002
+ "epoch": 7.12,
1003
+ "learning_rate": 5.763293310463123e-05,
1004
+ "loss": 0.0443,
1005
+ "step": 8300
1006
+ },
1007
+ {
1008
+ "epoch": 7.16,
1009
+ "learning_rate": 5.677530017152659e-05,
1010
+ "loss": 0.0419,
1011
+ "step": 8350
1012
+ },
1013
+ {
1014
+ "epoch": 7.2,
1015
+ "learning_rate": 5.591766723842196e-05,
1016
+ "loss": 0.0406,
1017
+ "step": 8400
1018
+ },
1019
+ {
1020
+ "epoch": 7.24,
1021
+ "learning_rate": 5.506003430531733e-05,
1022
+ "loss": 0.043,
1023
+ "step": 8450
1024
+ },
1025
+ {
1026
+ "epoch": 7.29,
1027
+ "learning_rate": 5.4202401372212695e-05,
1028
+ "loss": 0.042,
1029
+ "step": 8500
1030
+ },
1031
+ {
1032
+ "epoch": 7.33,
1033
+ "learning_rate": 5.3344768439108064e-05,
1034
+ "loss": 0.0404,
1035
+ "step": 8550
1036
+ },
1037
+ {
1038
+ "epoch": 7.37,
1039
+ "learning_rate": 5.248713550600344e-05,
1040
+ "loss": 0.0415,
1041
+ "step": 8600
1042
+ },
1043
+ {
1044
+ "epoch": 7.42,
1045
+ "learning_rate": 5.16295025728988e-05,
1046
+ "loss": 0.0443,
1047
+ "step": 8650
1048
+ },
1049
+ {
1050
+ "epoch": 7.46,
1051
+ "learning_rate": 5.077186963979417e-05,
1052
+ "loss": 0.0435,
1053
+ "step": 8700
1054
+ },
1055
+ {
1056
+ "epoch": 7.5,
1057
+ "learning_rate": 4.991423670668954e-05,
1058
+ "loss": 0.0393,
1059
+ "step": 8750
1060
+ },
1061
+ {
1062
+ "epoch": 7.54,
1063
+ "learning_rate": 4.9056603773584906e-05,
1064
+ "loss": 0.0424,
1065
+ "step": 8800
1066
+ },
1067
+ {
1068
+ "epoch": 7.59,
1069
+ "learning_rate": 4.819897084048028e-05,
1070
+ "loss": 0.0411,
1071
+ "step": 8850
1072
+ },
1073
+ {
1074
+ "epoch": 7.63,
1075
+ "learning_rate": 4.7341337907375644e-05,
1076
+ "loss": 0.0409,
1077
+ "step": 8900
1078
+ },
1079
+ {
1080
+ "epoch": 7.67,
1081
+ "learning_rate": 4.648370497427101e-05,
1082
+ "loss": 0.0431,
1083
+ "step": 8950
1084
+ },
1085
+ {
1086
+ "epoch": 7.72,
1087
+ "learning_rate": 4.562607204116639e-05,
1088
+ "loss": 0.0411,
1089
+ "step": 9000
1090
+ },
1091
+ {
1092
+ "epoch": 7.76,
1093
+ "learning_rate": 4.476843910806175e-05,
1094
+ "loss": 0.0378,
1095
+ "step": 9050
1096
+ },
1097
+ {
1098
+ "epoch": 7.8,
1099
+ "learning_rate": 4.391080617495712e-05,
1100
+ "loss": 0.0402,
1101
+ "step": 9100
1102
+ },
1103
+ {
1104
+ "epoch": 7.84,
1105
+ "learning_rate": 4.305317324185249e-05,
1106
+ "loss": 0.0396,
1107
+ "step": 9150
1108
+ },
1109
+ {
1110
+ "epoch": 7.89,
1111
+ "learning_rate": 4.219554030874786e-05,
1112
+ "loss": 0.0402,
1113
+ "step": 9200
1114
+ },
1115
+ {
1116
+ "epoch": 7.93,
1117
+ "learning_rate": 4.1337907375643224e-05,
1118
+ "loss": 0.0419,
1119
+ "step": 9250
1120
+ },
1121
+ {
1122
+ "epoch": 7.97,
1123
+ "learning_rate": 4.04802744425386e-05,
1124
+ "loss": 0.0403,
1125
+ "step": 9300
1126
+ },
1127
+ {
1128
+ "epoch": 8.02,
1129
+ "learning_rate": 3.962264150943397e-05,
1130
+ "loss": 0.0313,
1131
+ "step": 9350
1132
+ },
1133
+ {
1134
+ "epoch": 8.06,
1135
+ "learning_rate": 3.876500857632933e-05,
1136
+ "loss": 0.0147,
1137
+ "step": 9400
1138
+ },
1139
+ {
1140
+ "epoch": 8.1,
1141
+ "learning_rate": 3.7907375643224705e-05,
1142
+ "loss": 0.0174,
1143
+ "step": 9450
1144
+ },
1145
+ {
1146
+ "epoch": 8.14,
1147
+ "learning_rate": 3.704974271012007e-05,
1148
+ "loss": 0.0159,
1149
+ "step": 9500
1150
+ },
1151
+ {
1152
+ "epoch": 8.19,
1153
+ "learning_rate": 3.6192109777015435e-05,
1154
+ "loss": 0.0151,
1155
+ "step": 9550
1156
+ },
1157
+ {
1158
+ "epoch": 8.23,
1159
+ "learning_rate": 3.533447684391081e-05,
1160
+ "loss": 0.0165,
1161
+ "step": 9600
1162
+ },
1163
+ {
1164
+ "epoch": 8.27,
1165
+ "learning_rate": 3.447684391080618e-05,
1166
+ "loss": 0.0165,
1167
+ "step": 9650
1168
+ },
1169
+ {
1170
+ "epoch": 8.32,
1171
+ "learning_rate": 3.361921097770154e-05,
1172
+ "loss": 0.0154,
1173
+ "step": 9700
1174
+ },
1175
+ {
1176
+ "epoch": 8.36,
1177
+ "learning_rate": 3.2761578044596916e-05,
1178
+ "loss": 0.0158,
1179
+ "step": 9750
1180
+ },
1181
+ {
1182
+ "epoch": 8.4,
1183
+ "learning_rate": 3.1903945111492285e-05,
1184
+ "loss": 0.0162,
1185
+ "step": 9800
1186
+ },
1187
+ {
1188
+ "epoch": 8.44,
1189
+ "learning_rate": 3.1046312178387653e-05,
1190
+ "loss": 0.0142,
1191
+ "step": 9850
1192
+ },
1193
+ {
1194
+ "epoch": 8.49,
1195
+ "learning_rate": 3.018867924528302e-05,
1196
+ "loss": 0.0172,
1197
+ "step": 9900
1198
+ },
1199
+ {
1200
+ "epoch": 8.53,
1201
+ "learning_rate": 2.933104631217839e-05,
1202
+ "loss": 0.0178,
1203
+ "step": 9950
1204
+ },
1205
+ {
1206
+ "epoch": 8.57,
1207
+ "learning_rate": 2.847341337907376e-05,
1208
+ "loss": 0.0146,
1209
+ "step": 10000
1210
+ },
1211
+ {
1212
+ "epoch": 8.62,
1213
+ "learning_rate": 2.7615780445969124e-05,
1214
+ "loss": 0.0144,
1215
+ "step": 10050
1216
+ },
1217
+ {
1218
+ "epoch": 8.66,
1219
+ "learning_rate": 2.6758147512864496e-05,
1220
+ "loss": 0.0141,
1221
+ "step": 10100
1222
+ },
1223
+ {
1224
+ "epoch": 8.7,
1225
+ "learning_rate": 2.5900514579759865e-05,
1226
+ "loss": 0.0143,
1227
+ "step": 10150
1228
+ },
1229
+ {
1230
+ "epoch": 8.74,
1231
+ "learning_rate": 2.504288164665523e-05,
1232
+ "loss": 0.0164,
1233
+ "step": 10200
1234
+ },
1235
+ {
1236
+ "epoch": 8.79,
1237
+ "learning_rate": 2.4185248713550602e-05,
1238
+ "loss": 0.017,
1239
+ "step": 10250
1240
+ },
1241
+ {
1242
+ "epoch": 8.83,
1243
+ "learning_rate": 2.332761578044597e-05,
1244
+ "loss": 0.0156,
1245
+ "step": 10300
1246
+ },
1247
+ {
1248
+ "epoch": 8.87,
1249
+ "learning_rate": 2.246998284734134e-05,
1250
+ "loss": 0.0139,
1251
+ "step": 10350
1252
+ },
1253
+ {
1254
+ "epoch": 8.92,
1255
+ "learning_rate": 2.1612349914236708e-05,
1256
+ "loss": 0.0147,
1257
+ "step": 10400
1258
+ },
1259
+ {
1260
+ "epoch": 8.96,
1261
+ "learning_rate": 2.0754716981132076e-05,
1262
+ "loss": 0.0151,
1263
+ "step": 10450
1264
+ },
1265
+ {
1266
+ "epoch": 9.0,
1267
+ "learning_rate": 1.9897084048027445e-05,
1268
+ "loss": 0.015,
1269
+ "step": 10500
1270
+ },
1271
+ {
1272
+ "epoch": 9.04,
1273
+ "learning_rate": 1.9039451114922813e-05,
1274
+ "loss": 0.006,
1275
+ "step": 10550
1276
+ },
1277
+ {
1278
+ "epoch": 9.09,
1279
+ "learning_rate": 1.8181818181818182e-05,
1280
+ "loss": 0.0064,
1281
+ "step": 10600
1282
+ },
1283
+ {
1284
+ "epoch": 9.13,
1285
+ "learning_rate": 1.732418524871355e-05,
1286
+ "loss": 0.007,
1287
+ "step": 10650
1288
+ },
1289
+ {
1290
+ "epoch": 9.17,
1291
+ "learning_rate": 1.6466552315608923e-05,
1292
+ "loss": 0.0062,
1293
+ "step": 10700
1294
+ },
1295
+ {
1296
+ "epoch": 9.22,
1297
+ "learning_rate": 1.5608919382504288e-05,
1298
+ "loss": 0.0055,
1299
+ "step": 10750
1300
+ },
1301
+ {
1302
+ "epoch": 9.26,
1303
+ "learning_rate": 1.4751286449399656e-05,
1304
+ "loss": 0.0058,
1305
+ "step": 10800
1306
+ },
1307
+ {
1308
+ "epoch": 9.3,
1309
+ "learning_rate": 1.3893653516295027e-05,
1310
+ "loss": 0.0062,
1311
+ "step": 10850
1312
+ },
1313
+ {
1314
+ "epoch": 9.34,
1315
+ "learning_rate": 1.3036020583190395e-05,
1316
+ "loss": 0.0057,
1317
+ "step": 10900
1318
+ },
1319
+ {
1320
+ "epoch": 9.39,
1321
+ "learning_rate": 1.2178387650085764e-05,
1322
+ "loss": 0.0064,
1323
+ "step": 10950
1324
+ },
1325
+ {
1326
+ "epoch": 9.43,
1327
+ "learning_rate": 1.1320754716981132e-05,
1328
+ "loss": 0.0062,
1329
+ "step": 11000
1330
+ },
1331
+ {
1332
+ "epoch": 9.47,
1333
+ "learning_rate": 1.0463121783876503e-05,
1334
+ "loss": 0.006,
1335
+ "step": 11050
1336
+ },
1337
+ {
1338
+ "epoch": 9.52,
1339
+ "learning_rate": 9.60548885077187e-06,
1340
+ "loss": 0.0064,
1341
+ "step": 11100
1342
+ },
1343
+ {
1344
+ "epoch": 9.56,
1345
+ "learning_rate": 8.747855917667238e-06,
1346
+ "loss": 0.0061,
1347
+ "step": 11150
1348
+ },
1349
+ {
1350
+ "epoch": 9.6,
1351
+ "learning_rate": 7.890222984562608e-06,
1352
+ "loss": 0.0056,
1353
+ "step": 11200
1354
+ },
1355
+ {
1356
+ "epoch": 9.64,
1357
+ "learning_rate": 7.032590051457977e-06,
1358
+ "loss": 0.0056,
1359
+ "step": 11250
1360
+ },
1361
+ {
1362
+ "epoch": 9.69,
1363
+ "learning_rate": 6.1749571183533456e-06,
1364
+ "loss": 0.0059,
1365
+ "step": 11300
1366
+ },
1367
+ {
1368
+ "epoch": 9.73,
1369
+ "learning_rate": 5.317324185248713e-06,
1370
+ "loss": 0.0062,
1371
+ "step": 11350
1372
+ },
1373
+ {
1374
+ "epoch": 9.77,
1375
+ "learning_rate": 4.459691252144083e-06,
1376
+ "loss": 0.0056,
1377
+ "step": 11400
1378
+ },
1379
+ {
1380
+ "epoch": 9.82,
1381
+ "learning_rate": 3.6020583190394517e-06,
1382
+ "loss": 0.0061,
1383
+ "step": 11450
1384
+ },
1385
+ {
1386
+ "epoch": 9.86,
1387
+ "learning_rate": 2.74442538593482e-06,
1388
+ "loss": 0.0061,
1389
+ "step": 11500
1390
+ },
1391
+ {
1392
+ "epoch": 9.9,
1393
+ "learning_rate": 1.8867924528301887e-06,
1394
+ "loss": 0.0056,
1395
+ "step": 11550
1396
+ },
1397
+ {
1398
+ "epoch": 9.94,
1399
+ "learning_rate": 1.0291595197255577e-06,
1400
+ "loss": 0.0058,
1401
+ "step": 11600
1402
+ },
1403
+ {
1404
+ "epoch": 9.99,
1405
+ "learning_rate": 1.7152658662092624e-07,
1406
+ "loss": 0.0064,
1407
+ "step": 11650
1408
+ }
1409
+ ],
1410
+ "logging_steps": 50,
1411
+ "max_steps": 11660,
1412
+ "num_input_tokens_seen": 0,
1413
+ "num_train_epochs": 10,
1414
+ "save_steps": 500,
1415
+ "total_flos": 2.652879536528646e+18,
1416
+ "train_batch_size": 16,
1417
+ "trial_name": null,
1418
+ "trial_params": null
1419
+ }
training_args.bin ADDED
@@ -0,0 +1,3 @@
 
 
 
 
1
+ version https://git-lfs.github.com/spec/v1
2
+ oid sha256:a26eb80cfd8463bda8ddb4cba9a8643c6c7a0aa7e47374a5256a809e42505d9b
3
+ size 4664