joshuaclymer commited on
Commit
dc369b8
1 Parent(s): 529c1e8

Upload folder using huggingface_hub

Browse files
README.md ADDED
@@ -0,0 +1,21 @@
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
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: True
17
+ - bnb_4bit_compute_dtype: float16
18
+ ### Framework versions
19
+
20
+
21
+ - PEFT 0.5.0
adapter_config.json ADDED
@@ -0,0 +1,21 @@
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
1
+ {
2
+ "auto_mapping": null,
3
+ "base_model_name_or_path": "models/llama-30b",
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": 0.0,
12
+ "modules_to_save": null,
13
+ "peft_type": "LORA",
14
+ "r": 64,
15
+ "revision": null,
16
+ "target_modules": [
17
+ "q_proj",
18
+ "v_proj"
19
+ ],
20
+ "task_type": "SEQ_CLS"
21
+ }
adapter_model.bin ADDED
@@ -0,0 +1,3 @@
 
 
 
 
1
+ version https://git-lfs.github.com/spec/v1
2
+ oid sha256:98185ce4b789e09230b6b5eca0b7966e6d874f2f7d579efda4a91eeed5377d74
3
+ size 409081617
special_tokens_map.json ADDED
@@ -0,0 +1,24 @@
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
1
+ {
2
+ "bos_token": {
3
+ "content": "<s>",
4
+ "lstrip": false,
5
+ "normalized": true,
6
+ "rstrip": false,
7
+ "single_word": false
8
+ },
9
+ "eos_token": {
10
+ "content": "</s>",
11
+ "lstrip": false,
12
+ "normalized": true,
13
+ "rstrip": false,
14
+ "single_word": false
15
+ },
16
+ "pad_token": "</s>",
17
+ "unk_token": {
18
+ "content": "<unk>",
19
+ "lstrip": false,
20
+ "normalized": true,
21
+ "rstrip": false,
22
+ "single_word": false
23
+ }
24
+ }
tokenizer.json ADDED
The diff for this file is too large to render. See raw diff
 
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 +1,32 @@
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
1
+ {
2
+ "bos_token": {
3
+ "__type": "AddedToken",
4
+ "content": "<s>",
5
+ "lstrip": false,
6
+ "normalized": true,
7
+ "rstrip": false,
8
+ "single_word": false
9
+ },
10
+ "clean_up_tokenization_spaces": false,
11
+ "eos_token": {
12
+ "__type": "AddedToken",
13
+ "content": "</s>",
14
+ "lstrip": false,
15
+ "normalized": true,
16
+ "rstrip": false,
17
+ "single_word": false
18
+ },
19
+ "model_max_length": 2048,
20
+ "pad_token": null,
21
+ "sp_model_kwargs": {},
22
+ "tokenizer_class": "LlamaTokenizer",
23
+ "unk_token": {
24
+ "__type": "AddedToken",
25
+ "content": "<unk>",
26
+ "lstrip": false,
27
+ "normalized": true,
28
+ "rstrip": false,
29
+ "single_word": false
30
+ },
31
+ "use_default_system_prompt": true
32
+ }
train_args.json ADDED
@@ -0,0 +1,123 @@
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
1
+ {
2
+ "initial_model_dir": "models/llama-30b",
3
+ "distribution_id": "math_exam_questions",
4
+ "date_trained": "11/10/2023 07:25:10",
5
+ "output_dir": "models/classify_lora/llama-30b-math_exam_questions",
6
+ "overwrite_output_dir": false,
7
+ "do_train": false,
8
+ "do_eval": true,
9
+ "do_predict": false,
10
+ "evaluation_strategy": "steps",
11
+ "prediction_loss_only": false,
12
+ "per_device_train_batch_size": 8,
13
+ "per_device_eval_batch_size": 8,
14
+ "per_gpu_train_batch_size": null,
15
+ "per_gpu_eval_batch_size": null,
16
+ "gradient_accumulation_steps": 1,
17
+ "eval_accumulation_steps": null,
18
+ "eval_delay": 0,
19
+ "learning_rate": 0.0002,
20
+ "weight_decay": 0.0,
21
+ "adam_beta1": 0.9,
22
+ "adam_beta2": 0.999,
23
+ "adam_epsilon": 1e-08,
24
+ "max_grad_norm": 0.3,
25
+ "num_train_epochs": 2,
26
+ "max_steps": 100,
27
+ "lr_scheduler_type": "constant",
28
+ "warmup_ratio": 0.03,
29
+ "warmup_steps": 0,
30
+ "log_level": "passive",
31
+ "log_level_replica": "warning",
32
+ "log_on_each_node": true,
33
+ "logging_dir": "models/classify_lora/llama-30b-math_exam_questions/runs/Oct11_06-16-21_compute-permanent-node-990",
34
+ "logging_strategy": "steps",
35
+ "logging_first_step": false,
36
+ "logging_steps": 1,
37
+ "logging_nan_inf_filter": true,
38
+ "save_strategy": "steps",
39
+ "save_steps": 25,
40
+ "save_total_limit": 0,
41
+ "save_safetensors": false,
42
+ "save_on_each_node": false,
43
+ "no_cuda": false,
44
+ "use_cpu": false,
45
+ "use_mps_device": false,
46
+ "seed": 42,
47
+ "data_seed": null,
48
+ "jit_mode_eval": false,
49
+ "use_ipex": false,
50
+ "bf16": false,
51
+ "fp16": false,
52
+ "fp16_opt_level": "O1",
53
+ "half_precision_backend": "auto",
54
+ "bf16_full_eval": false,
55
+ "fp16_full_eval": false,
56
+ "tf32": null,
57
+ "local_rank": 2,
58
+ "ddp_backend": null,
59
+ "tpu_num_cores": null,
60
+ "tpu_metrics_debug": false,
61
+ "debug": [],
62
+ "dataloader_drop_last": false,
63
+ "eval_steps": 25,
64
+ "dataloader_num_workers": 0,
65
+ "past_index": -1,
66
+ "run_name": "train|models-classify_lora-llama-30b-math_exam_questions",
67
+ "disable_tqdm": false,
68
+ "remove_unused_columns": false,
69
+ "label_names": null,
70
+ "load_best_model_at_end": false,
71
+ "metric_for_best_model": "eval_math_exam_questions_score",
72
+ "greater_is_better": true,
73
+ "ignore_data_skip": false,
74
+ "sharded_ddp": [],
75
+ "fsdp": [],
76
+ "fsdp_min_num_params": 0,
77
+ "fsdp_config": {
78
+ "min_num_params": 0,
79
+ "xla": false,
80
+ "xla_fsdp_grad_ckpt": false
81
+ },
82
+ "fsdp_transformer_layer_cls_to_wrap": null,
83
+ "deepspeed": "configs/ds_zero_1.json",
84
+ "label_smoothing_factor": 0.0,
85
+ "optim": "paged_adamw_32bit",
86
+ "optim_args": null,
87
+ "adafactor": false,
88
+ "group_by_length": false,
89
+ "length_column_name": "length",
90
+ "report_to": [
91
+ "wandb"
92
+ ],
93
+ "ddp_find_unused_parameters": false,
94
+ "ddp_bucket_cap_mb": null,
95
+ "ddp_broadcast_buffers": null,
96
+ "dataloader_pin_memory": true,
97
+ "skip_memory_metrics": true,
98
+ "use_legacy_prediction_loop": false,
99
+ "push_to_hub": false,
100
+ "resume_from_checkpoint": null,
101
+ "hub_model_id": null,
102
+ "hub_strategy": "every_save",
103
+ "hub_token": null,
104
+ "hub_private_repo": false,
105
+ "hub_always_push": false,
106
+ "gradient_checkpointing": false,
107
+ "include_inputs_for_metrics": false,
108
+ "fp16_backend": "auto",
109
+ "push_to_hub_model_id": null,
110
+ "push_to_hub_organization": null,
111
+ "push_to_hub_token": null,
112
+ "_n_gpu": 1,
113
+ "mp_parameters": "",
114
+ "auto_find_batch_size": false,
115
+ "full_determinism": false,
116
+ "torchdynamo": null,
117
+ "ray_scope": "last",
118
+ "ddp_timeout": 1800,
119
+ "torch_compile": false,
120
+ "torch_compile_backend": null,
121
+ "torch_compile_mode": null,
122
+ "dispatch_batches": null
123
+ }
training_args.bin ADDED
@@ -0,0 +1,3 @@
 
 
 
 
1
+ version https://git-lfs.github.com/spec/v1
2
+ oid sha256:e5128efa9cac339774d815dfb2be56bc8b5eb7b3e15bafb31fea9459929434ce
3
+ size 5115
training_logs.json ADDED
@@ -0,0 +1,1067 @@
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
1
+ [
2
+ {
3
+ "loss": 0.6476,
4
+ "learning_rate": 0.0002,
5
+ "epoch": 0.02,
6
+ "step": 1
7
+ },
8
+ {
9
+ "loss": 0.6668,
10
+ "learning_rate": 0.0002,
11
+ "epoch": 0.04,
12
+ "step": 2
13
+ },
14
+ {
15
+ "loss": 0.6615,
16
+ "learning_rate": 0.0002,
17
+ "epoch": 0.05,
18
+ "step": 3
19
+ },
20
+ {
21
+ "loss": 0.5895,
22
+ "learning_rate": 0.0002,
23
+ "epoch": 0.07,
24
+ "step": 4
25
+ },
26
+ {
27
+ "loss": 0.6124,
28
+ "learning_rate": 0.0002,
29
+ "epoch": 0.09,
30
+ "step": 5
31
+ },
32
+ {
33
+ "loss": 0.5953,
34
+ "learning_rate": 0.0002,
35
+ "epoch": 0.11,
36
+ "step": 6
37
+ },
38
+ {
39
+ "loss": 0.5356,
40
+ "learning_rate": 0.0002,
41
+ "epoch": 0.12,
42
+ "step": 7
43
+ },
44
+ {
45
+ "loss": 0.4604,
46
+ "learning_rate": 0.0002,
47
+ "epoch": 0.14,
48
+ "step": 8
49
+ },
50
+ {
51
+ "loss": 0.4143,
52
+ "learning_rate": 0.0002,
53
+ "epoch": 0.16,
54
+ "step": 9
55
+ },
56
+ {
57
+ "loss": 0.4553,
58
+ "learning_rate": 0.0002,
59
+ "epoch": 0.18,
60
+ "step": 10
61
+ },
62
+ {
63
+ "loss": 0.3538,
64
+ "learning_rate": 0.0002,
65
+ "epoch": 0.2,
66
+ "step": 11
67
+ },
68
+ {
69
+ "loss": 0.3389,
70
+ "learning_rate": 0.0002,
71
+ "epoch": 0.21,
72
+ "step": 12
73
+ },
74
+ {
75
+ "loss": 0.2639,
76
+ "learning_rate": 0.0002,
77
+ "epoch": 0.23,
78
+ "step": 13
79
+ },
80
+ {
81
+ "loss": 0.2096,
82
+ "learning_rate": 0.0002,
83
+ "epoch": 0.25,
84
+ "step": 14
85
+ },
86
+ {
87
+ "loss": 0.2002,
88
+ "learning_rate": 0.0002,
89
+ "epoch": 0.27,
90
+ "step": 15
91
+ },
92
+ {
93
+ "loss": 0.2374,
94
+ "learning_rate": 0.0002,
95
+ "epoch": 0.29,
96
+ "step": 16
97
+ },
98
+ {
99
+ "loss": 0.2475,
100
+ "learning_rate": 0.0002,
101
+ "epoch": 0.3,
102
+ "step": 17
103
+ },
104
+ {
105
+ "loss": 0.1994,
106
+ "learning_rate": 0.0002,
107
+ "epoch": 0.32,
108
+ "step": 18
109
+ },
110
+ {
111
+ "loss": 0.1406,
112
+ "learning_rate": 0.0002,
113
+ "epoch": 0.34,
114
+ "step": 19
115
+ },
116
+ {
117
+ "loss": 0.055,
118
+ "learning_rate": 0.0002,
119
+ "epoch": 0.36,
120
+ "step": 20
121
+ },
122
+ {
123
+ "loss": 0.1168,
124
+ "learning_rate": 0.0002,
125
+ "epoch": 0.38,
126
+ "step": 21
127
+ },
128
+ {
129
+ "loss": 0.0916,
130
+ "learning_rate": 0.0002,
131
+ "epoch": 0.39,
132
+ "step": 22
133
+ },
134
+ {
135
+ "loss": 0.1596,
136
+ "learning_rate": 0.0002,
137
+ "epoch": 0.41,
138
+ "step": 23
139
+ },
140
+ {
141
+ "loss": 0.1585,
142
+ "learning_rate": 0.0002,
143
+ "epoch": 0.43,
144
+ "step": 24
145
+ },
146
+ {
147
+ "loss": 0.184,
148
+ "learning_rate": 0.0002,
149
+ "epoch": 0.45,
150
+ "step": 25
151
+ },
152
+ {
153
+ "eval_math_exam_questions_loss": 0.09979354590177536,
154
+ "eval_math_exam_questions_score": -0.029867494478821754,
155
+ "eval_math_exam_questions_brier_score": 0.029867494478821754,
156
+ "eval_math_exam_questions_average_probability": 0.9305303692817688,
157
+ "eval_math_exam_questions_accuracy": 0.95,
158
+ "eval_math_exam_questions_probabilities": [
159
+ 0.8511927723884583,
160
+ 0.8560279011726379,
161
+ 0.9999574422836304,
162
+ 0.9483713507652283,
163
+ 0.2606664001941681,
164
+ 0.9442870616912842,
165
+ 0.9947615265846252,
166
+ 0.9007341265678406,
167
+ 0.9449180364608765,
168
+ 0.41139501333236694,
169
+ 0.9922882318496704,
170
+ 0.9992316961288452,
171
+ 0.9993873834609985,
172
+ 0.9968752861022949,
173
+ 0.9976814985275269,
174
+ 0.9999994039535522,
175
+ 0.9999996423721313,
176
+ 0.9999994039535522,
177
+ 0.9571589231491089,
178
+ 0.995111882686615,
179
+ 0.979494571685791,
180
+ 0.574057400226593,
181
+ 0.7285555601119995,
182
+ 0.3839860260486603,
183
+ 0.9998674392700195,
184
+ 0.9982432126998901,
185
+ 0.9999611377716064,
186
+ 0.9999970197677612,
187
+ 0.9999997615814209,
188
+ 1.0,
189
+ 0.9638392329216003,
190
+ 0.968817949295044,
191
+ 0.2049836367368698,
192
+ 0.9999809265136719,
193
+ 0.9998273253440857,
194
+ 0.9999629259109497,
195
+ 0.9595451951026917,
196
+ 0.9992743134498596,
197
+ 0.9850507974624634,
198
+ 0.9250211715698242,
199
+ 0.840378999710083,
200
+ 0.8428780436515808,
201
+ 0.9994524121284485,
202
+ 0.9991476535797119,
203
+ 0.9997357726097107,
204
+ 0.9995108842849731,
205
+ 0.9998007416725159,
206
+ 0.9996703863143921,
207
+ 0.9999984502792358,
208
+ 0.9999967813491821,
209
+ 0.999998927116394,
210
+ 0.9999979734420776,
211
+ 0.9999997615814209,
212
+ 0.9999996423721313,
213
+ 0.9957982897758484,
214
+ 0.9765607714653015,
215
+ 0.996933102607727,
216
+ 0.8950393795967102,
217
+ 0.9991758465766907,
218
+ 0.9990474581718445,
219
+ 0.7260539531707764,
220
+ 0.8293086290359497,
221
+ 0.381984144449234,
222
+ 0.9996906518936157,
223
+ 0.9999486207962036,
224
+ 0.9999747276306152,
225
+ 0.994877815246582,
226
+ 0.991081714630127,
227
+ 0.9968804121017456,
228
+ 0.9998242259025574,
229
+ 0.9999198913574219,
230
+ 0.999997615814209,
231
+ 0.9805970788002014,
232
+ 0.9670814871788025,
233
+ 0.8692526817321777,
234
+ 0.9998607635498047,
235
+ 0.9999896287918091,
236
+ 0.999993085861206,
237
+ 0.9961829781532288,
238
+ 0.9215685129165649,
239
+ 0.998610258102417,
240
+ 0.9952474236488342,
241
+ 0.7989624738693237,
242
+ 0.9947852492332458,
243
+ 0.9998492002487183,
244
+ 0.9999179840087891,
245
+ 0.9999663829803467,
246
+ 0.9998860359191895,
247
+ 0.9998941421508789,
248
+ 0.9978526830673218,
249
+ 0.8546462059020996,
250
+ 0.8965560793876648,
251
+ 0.6786884069442749,
252
+ 0.9975112676620483,
253
+ 0.9996652603149414,
254
+ 0.9990561604499817,
255
+ 0.9986518025398254,
256
+ 0.9981924891471863,
257
+ 0.998735249042511,
258
+ 0.9286666512489319
259
+ ],
260
+ "eval_math_exam_questions_runtime": 58.7461,
261
+ "eval_math_exam_questions_samples_per_second": 1.702,
262
+ "eval_math_exam_questions_steps_per_second": 0.068,
263
+ "epoch": 0.45,
264
+ "step": 25
265
+ },
266
+ {
267
+ "loss": 0.0599,
268
+ "learning_rate": 0.0002,
269
+ "epoch": 0.46,
270
+ "step": 26
271
+ },
272
+ {
273
+ "loss": 0.0445,
274
+ "learning_rate": 0.0002,
275
+ "epoch": 0.48,
276
+ "step": 27
277
+ },
278
+ {
279
+ "loss": 0.0625,
280
+ "learning_rate": 0.0002,
281
+ "epoch": 0.5,
282
+ "step": 28
283
+ },
284
+ {
285
+ "loss": 0.0125,
286
+ "learning_rate": 0.0002,
287
+ "epoch": 0.52,
288
+ "step": 29
289
+ },
290
+ {
291
+ "loss": 0.2316,
292
+ "learning_rate": 0.0002,
293
+ "epoch": 0.54,
294
+ "step": 30
295
+ },
296
+ {
297
+ "loss": 0.0746,
298
+ "learning_rate": 0.0002,
299
+ "epoch": 0.55,
300
+ "step": 31
301
+ },
302
+ {
303
+ "loss": 0.1032,
304
+ "learning_rate": 0.0002,
305
+ "epoch": 0.57,
306
+ "step": 32
307
+ },
308
+ {
309
+ "loss": 0.1413,
310
+ "learning_rate": 0.0002,
311
+ "epoch": 0.59,
312
+ "step": 33
313
+ },
314
+ {
315
+ "loss": 0.1345,
316
+ "learning_rate": 0.0002,
317
+ "epoch": 0.61,
318
+ "step": 34
319
+ },
320
+ {
321
+ "loss": 0.0688,
322
+ "learning_rate": 0.0002,
323
+ "epoch": 0.62,
324
+ "step": 35
325
+ },
326
+ {
327
+ "loss": 0.0762,
328
+ "learning_rate": 0.0002,
329
+ "epoch": 0.64,
330
+ "step": 36
331
+ },
332
+ {
333
+ "loss": 0.0552,
334
+ "learning_rate": 0.0002,
335
+ "epoch": 0.66,
336
+ "step": 37
337
+ },
338
+ {
339
+ "loss": 0.0356,
340
+ "learning_rate": 0.0002,
341
+ "epoch": 0.68,
342
+ "step": 38
343
+ },
344
+ {
345
+ "loss": 0.0543,
346
+ "learning_rate": 0.0002,
347
+ "epoch": 0.7,
348
+ "step": 39
349
+ },
350
+ {
351
+ "loss": 0.0279,
352
+ "learning_rate": 0.0002,
353
+ "epoch": 0.71,
354
+ "step": 40
355
+ },
356
+ {
357
+ "loss": 0.0704,
358
+ "learning_rate": 0.0002,
359
+ "epoch": 0.73,
360
+ "step": 41
361
+ },
362
+ {
363
+ "loss": 0.039,
364
+ "learning_rate": 0.0002,
365
+ "epoch": 0.75,
366
+ "step": 42
367
+ },
368
+ {
369
+ "loss": 0.0114,
370
+ "learning_rate": 0.0002,
371
+ "epoch": 0.77,
372
+ "step": 43
373
+ },
374
+ {
375
+ "loss": 0.0043,
376
+ "learning_rate": 0.0002,
377
+ "epoch": 0.79,
378
+ "step": 44
379
+ },
380
+ {
381
+ "loss": 0.1562,
382
+ "learning_rate": 0.0002,
383
+ "epoch": 0.8,
384
+ "step": 45
385
+ },
386
+ {
387
+ "loss": 0.0495,
388
+ "learning_rate": 0.0002,
389
+ "epoch": 0.82,
390
+ "step": 46
391
+ },
392
+ {
393
+ "loss": 0.1683,
394
+ "learning_rate": 0.0002,
395
+ "epoch": 0.84,
396
+ "step": 47
397
+ },
398
+ {
399
+ "loss": 0.0118,
400
+ "learning_rate": 0.0002,
401
+ "epoch": 0.86,
402
+ "step": 48
403
+ },
404
+ {
405
+ "loss": 0.0708,
406
+ "learning_rate": 0.0002,
407
+ "epoch": 0.88,
408
+ "step": 49
409
+ },
410
+ {
411
+ "loss": 0.0445,
412
+ "learning_rate": 0.0002,
413
+ "epoch": 0.89,
414
+ "step": 50
415
+ },
416
+ {
417
+ "eval_math_exam_questions_loss": 0.11060654371976852,
418
+ "eval_math_exam_questions_score": -0.04164460673928261,
419
+ "eval_math_exam_questions_brier_score": 0.04164460673928261,
420
+ "eval_math_exam_questions_average_probability": 0.9377254247665405,
421
+ "eval_math_exam_questions_accuracy": 0.92,
422
+ "eval_math_exam_questions_probabilities": [
423
+ 0.9878444075584412,
424
+ 0.9649227261543274,
425
+ 0.9999626874923706,
426
+ 0.9997627139091492,
427
+ 0.41481563448905945,
428
+ 0.9997966885566711,
429
+ 0.9999992847442627,
430
+ 0.9999576807022095,
431
+ 0.9999995231628418,
432
+ 0.13421274721622467,
433
+ 0.9997405409812927,
434
+ 0.9999984502792358,
435
+ 0.9999998807907104,
436
+ 0.9999960660934448,
437
+ 0.9999959468841553,
438
+ 1.0,
439
+ 1.0,
440
+ 1.0,
441
+ 0.9999942779541016,
442
+ 0.9999998807907104,
443
+ 0.9999129772186279,
444
+ 0.4325127899646759,
445
+ 0.19135379791259766,
446
+ 0.2162252813577652,
447
+ 1.0,
448
+ 0.9999597072601318,
449
+ 1.0,
450
+ 0.9999998807907104,
451
+ 1.0,
452
+ 1.0,
453
+ 0.9999549388885498,
454
+ 0.9999803304672241,
455
+ 0.9946361184120178,
456
+ 0.9999997615814209,
457
+ 0.9998887777328491,
458
+ 0.9999957084655762,
459
+ 0.9995691180229187,
460
+ 1.0,
461
+ 0.99983811378479,
462
+ 0.9979630708694458,
463
+ 0.9983990788459778,
464
+ 0.9632781147956848,
465
+ 0.9999992847442627,
466
+ 0.9999983310699463,
467
+ 0.9999996423721313,
468
+ 0.9999991655349731,
469
+ 0.999947190284729,
470
+ 0.9998934268951416,
471
+ 1.0,
472
+ 1.0,
473
+ 1.0,
474
+ 1.0,
475
+ 1.0,
476
+ 1.0,
477
+ 0.9999369382858276,
478
+ 0.9989650249481201,
479
+ 0.9998283386230469,
480
+ 0.9974669218063354,
481
+ 1.0,
482
+ 1.0,
483
+ 0.7936062216758728,
484
+ 0.9307928085327148,
485
+ 0.27603790163993835,
486
+ 0.9999816417694092,
487
+ 0.9999940395355225,
488
+ 0.9999995231628418,
489
+ 0.9996922016143799,
490
+ 0.988324761390686,
491
+ 0.9975664615631104,
492
+ 0.9999992847442627,
493
+ 0.9999995231628418,
494
+ 1.0,
495
+ 0.9978724718093872,
496
+ 0.3203738033771515,
497
+ 0.3396102786064148,
498
+ 0.9999998807907104,
499
+ 1.0,
500
+ 1.0,
501
+ 0.9998663663864136,
502
+ 0.988074004650116,
503
+ 0.9999923706054688,
504
+ 0.9999964237213135,
505
+ 0.9248444437980652,
506
+ 0.9999958276748657,
507
+ 1.0,
508
+ 1.0,
509
+ 1.0,
510
+ 1.0,
511
+ 1.0,
512
+ 0.9999954700469971,
513
+ 0.9937870502471924,
514
+ 0.9950788021087646,
515
+ 0.9376111030578613,
516
+ 1.0,
517
+ 1.0,
518
+ 1.0,
519
+ 0.9999998807907104,
520
+ 0.9999986886978149,
521
+ 0.9999997615814209,
522
+ 0.9989570379257202
523
+ ],
524
+ "eval_math_exam_questions_runtime": 58.7331,
525
+ "eval_math_exam_questions_samples_per_second": 1.703,
526
+ "eval_math_exam_questions_steps_per_second": 0.068,
527
+ "epoch": 0.89,
528
+ "step": 50
529
+ },
530
+ {
531
+ "loss": 0.1862,
532
+ "learning_rate": 0.0002,
533
+ "epoch": 0.91,
534
+ "step": 51
535
+ },
536
+ {
537
+ "loss": 0.1376,
538
+ "learning_rate": 0.0002,
539
+ "epoch": 0.93,
540
+ "step": 52
541
+ },
542
+ {
543
+ "loss": 0.1167,
544
+ "learning_rate": 0.0002,
545
+ "epoch": 0.95,
546
+ "step": 53
547
+ },
548
+ {
549
+ "loss": 0.1163,
550
+ "learning_rate": 0.0002,
551
+ "epoch": 0.96,
552
+ "step": 54
553
+ },
554
+ {
555
+ "loss": 0.0308,
556
+ "learning_rate": 0.0002,
557
+ "epoch": 0.98,
558
+ "step": 55
559
+ },
560
+ {
561
+ "loss": 0.0515,
562
+ "learning_rate": 0.0002,
563
+ "epoch": 1.0,
564
+ "step": 56
565
+ },
566
+ {
567
+ "loss": 0.0876,
568
+ "learning_rate": 0.0002,
569
+ "epoch": 1.02,
570
+ "step": 57
571
+ },
572
+ {
573
+ "loss": 0.0215,
574
+ "learning_rate": 0.0002,
575
+ "epoch": 1.04,
576
+ "step": 58
577
+ },
578
+ {
579
+ "loss": 0.0454,
580
+ "learning_rate": 0.0002,
581
+ "epoch": 1.05,
582
+ "step": 59
583
+ },
584
+ {
585
+ "loss": 0.0335,
586
+ "learning_rate": 0.0002,
587
+ "epoch": 1.07,
588
+ "step": 60
589
+ },
590
+ {
591
+ "loss": 0.0248,
592
+ "learning_rate": 0.0002,
593
+ "epoch": 1.09,
594
+ "step": 61
595
+ },
596
+ {
597
+ "loss": 0.0028,
598
+ "learning_rate": 0.0002,
599
+ "epoch": 1.11,
600
+ "step": 62
601
+ },
602
+ {
603
+ "loss": 0.0175,
604
+ "learning_rate": 0.0002,
605
+ "epoch": 1.12,
606
+ "step": 63
607
+ },
608
+ {
609
+ "loss": 0.0078,
610
+ "learning_rate": 0.0002,
611
+ "epoch": 1.14,
612
+ "step": 64
613
+ },
614
+ {
615
+ "loss": 0.0113,
616
+ "learning_rate": 0.0002,
617
+ "epoch": 1.16,
618
+ "step": 65
619
+ },
620
+ {
621
+ "loss": 0.0119,
622
+ "learning_rate": 0.0002,
623
+ "epoch": 1.18,
624
+ "step": 66
625
+ },
626
+ {
627
+ "loss": 0.0235,
628
+ "learning_rate": 0.0002,
629
+ "epoch": 1.2,
630
+ "step": 67
631
+ },
632
+ {
633
+ "loss": 0.0116,
634
+ "learning_rate": 0.0002,
635
+ "epoch": 1.21,
636
+ "step": 68
637
+ },
638
+ {
639
+ "loss": 0.0018,
640
+ "learning_rate": 0.0002,
641
+ "epoch": 1.23,
642
+ "step": 69
643
+ },
644
+ {
645
+ "loss": 0.0024,
646
+ "learning_rate": 0.0002,
647
+ "epoch": 1.25,
648
+ "step": 70
649
+ },
650
+ {
651
+ "loss": 0.003,
652
+ "learning_rate": 0.0002,
653
+ "epoch": 1.27,
654
+ "step": 71
655
+ },
656
+ {
657
+ "loss": 0.0044,
658
+ "learning_rate": 0.0002,
659
+ "epoch": 1.29,
660
+ "step": 72
661
+ },
662
+ {
663
+ "loss": 0.0151,
664
+ "learning_rate": 0.0002,
665
+ "epoch": 1.3,
666
+ "step": 73
667
+ },
668
+ {
669
+ "loss": 0.0019,
670
+ "learning_rate": 0.0002,
671
+ "epoch": 1.32,
672
+ "step": 74
673
+ },
674
+ {
675
+ "loss": 0.0003,
676
+ "learning_rate": 0.0002,
677
+ "epoch": 1.34,
678
+ "step": 75
679
+ },
680
+ {
681
+ "eval_math_exam_questions_loss": 0.05752657726407051,
682
+ "eval_math_exam_questions_score": -0.019282517954707146,
683
+ "eval_math_exam_questions_brier_score": 0.019282517954707146,
684
+ "eval_math_exam_questions_average_probability": 0.9707407355308533,
685
+ "eval_math_exam_questions_accuracy": 0.97,
686
+ "eval_math_exam_questions_probabilities": [
687
+ 0.9951311349868774,
688
+ 0.8724949359893799,
689
+ 0.9999651908874512,
690
+ 1.0,
691
+ 0.8811224102973938,
692
+ 1.0,
693
+ 1.0,
694
+ 1.0,
695
+ 1.0,
696
+ 0.103812575340271,
697
+ 0.9999701976776123,
698
+ 0.9999998807907104,
699
+ 1.0,
700
+ 0.9999998807907104,
701
+ 1.0,
702
+ 1.0,
703
+ 1.0,
704
+ 1.0,
705
+ 1.0,
706
+ 1.0,
707
+ 0.9999974966049194,
708
+ 0.9944435954093933,
709
+ 0.9979708790779114,
710
+ 0.4370083212852478,
711
+ 1.0,
712
+ 0.9999988079071045,
713
+ 1.0,
714
+ 1.0,
715
+ 1.0,
716
+ 1.0,
717
+ 1.0,
718
+ 1.0,
719
+ 1.0,
720
+ 1.0,
721
+ 0.9999998807907104,
722
+ 1.0,
723
+ 0.14551357924938202,
724
+ 1.0,
725
+ 0.999981164932251,
726
+ 0.9999678134918213,
727
+ 0.9999961853027344,
728
+ 0.9987561702728271,
729
+ 1.0,
730
+ 1.0,
731
+ 1.0,
732
+ 1.0,
733
+ 1.0,
734
+ 0.9999998807907104,
735
+ 1.0,
736
+ 1.0,
737
+ 1.0,
738
+ 1.0,
739
+ 1.0,
740
+ 1.0,
741
+ 1.0,
742
+ 0.9999978542327881,
743
+ 1.0,
744
+ 0.999945878982544,
745
+ 1.0,
746
+ 1.0,
747
+ 0.8369179964065552,
748
+ 0.9991326928138733,
749
+ 0.8580266237258911,
750
+ 1.0,
751
+ 1.0,
752
+ 1.0,
753
+ 0.9999998807907104,
754
+ 0.9998944997787476,
755
+ 0.9997386336326599,
756
+ 1.0,
757
+ 1.0,
758
+ 1.0,
759
+ 1.0,
760
+ 0.9999858140945435,
761
+ 0.9989979863166809,
762
+ 0.9989114999771118,
763
+ 1.0,
764
+ 1.0,
765
+ 0.9999665021896362,
766
+ 0.9914140701293945,
767
+ 0.9999996423721313,
768
+ 1.0,
769
+ 0.9750034809112549,
770
+ 1.0,
771
+ 1.0,
772
+ 1.0,
773
+ 1.0,
774
+ 1.0,
775
+ 1.0,
776
+ 1.0,
777
+ 0.9999885559082031,
778
+ 0.999998927116394,
779
+ 0.9905345439910889,
780
+ 1.0,
781
+ 1.0,
782
+ 1.0,
783
+ 1.0,
784
+ 1.0,
785
+ 1.0,
786
+ 0.9994863271713257
787
+ ],
788
+ "eval_math_exam_questions_runtime": 58.6753,
789
+ "eval_math_exam_questions_samples_per_second": 1.704,
790
+ "eval_math_exam_questions_steps_per_second": 0.068,
791
+ "epoch": 1.34,
792
+ "step": 75
793
+ },
794
+ {
795
+ "loss": 0.0139,
796
+ "learning_rate": 0.0002,
797
+ "epoch": 1.36,
798
+ "step": 76
799
+ },
800
+ {
801
+ "loss": 0.0121,
802
+ "learning_rate": 0.0002,
803
+ "epoch": 1.38,
804
+ "step": 77
805
+ },
806
+ {
807
+ "loss": 0.0001,
808
+ "learning_rate": 0.0002,
809
+ "epoch": 1.39,
810
+ "step": 78
811
+ },
812
+ {
813
+ "loss": 0.0117,
814
+ "learning_rate": 0.0002,
815
+ "epoch": 1.41,
816
+ "step": 79
817
+ },
818
+ {
819
+ "loss": 0.0046,
820
+ "learning_rate": 0.0002,
821
+ "epoch": 1.43,
822
+ "step": 80
823
+ },
824
+ {
825
+ "loss": 0.3396,
826
+ "learning_rate": 0.0002,
827
+ "epoch": 1.45,
828
+ "step": 81
829
+ },
830
+ {
831
+ "loss": 0.0,
832
+ "learning_rate": 0.0002,
833
+ "epoch": 1.46,
834
+ "step": 82
835
+ },
836
+ {
837
+ "loss": 0.0014,
838
+ "learning_rate": 0.0002,
839
+ "epoch": 1.48,
840
+ "step": 83
841
+ },
842
+ {
843
+ "loss": 0.0034,
844
+ "learning_rate": 0.0002,
845
+ "epoch": 1.5,
846
+ "step": 84
847
+ },
848
+ {
849
+ "loss": 0.0026,
850
+ "learning_rate": 0.0002,
851
+ "epoch": 1.52,
852
+ "step": 85
853
+ },
854
+ {
855
+ "loss": 0.001,
856
+ "learning_rate": 0.0002,
857
+ "epoch": 1.54,
858
+ "step": 86
859
+ },
860
+ {
861
+ "loss": 0.0026,
862
+ "learning_rate": 0.0002,
863
+ "epoch": 1.55,
864
+ "step": 87
865
+ },
866
+ {
867
+ "loss": 0.006,
868
+ "learning_rate": 0.0002,
869
+ "epoch": 1.57,
870
+ "step": 88
871
+ },
872
+ {
873
+ "loss": 0.0001,
874
+ "learning_rate": 0.0002,
875
+ "epoch": 1.59,
876
+ "step": 89
877
+ },
878
+ {
879
+ "loss": 0.0024,
880
+ "learning_rate": 0.0002,
881
+ "epoch": 1.61,
882
+ "step": 90
883
+ },
884
+ {
885
+ "loss": 0.0031,
886
+ "learning_rate": 0.0002,
887
+ "epoch": 1.62,
888
+ "step": 91
889
+ },
890
+ {
891
+ "loss": 0.0005,
892
+ "learning_rate": 0.0002,
893
+ "epoch": 1.64,
894
+ "step": 92
895
+ },
896
+ {
897
+ "loss": 0.006,
898
+ "learning_rate": 0.0002,
899
+ "epoch": 1.66,
900
+ "step": 93
901
+ },
902
+ {
903
+ "loss": 0.0056,
904
+ "learning_rate": 0.0002,
905
+ "epoch": 1.68,
906
+ "step": 94
907
+ },
908
+ {
909
+ "loss": 0.0003,
910
+ "learning_rate": 0.0002,
911
+ "epoch": 1.7,
912
+ "step": 95
913
+ },
914
+ {
915
+ "loss": 0.007,
916
+ "learning_rate": 0.0002,
917
+ "epoch": 1.71,
918
+ "step": 96
919
+ },
920
+ {
921
+ "loss": 0.1818,
922
+ "learning_rate": 0.0002,
923
+ "epoch": 1.73,
924
+ "step": 97
925
+ },
926
+ {
927
+ "loss": 0.0089,
928
+ "learning_rate": 0.0002,
929
+ "epoch": 1.75,
930
+ "step": 98
931
+ },
932
+ {
933
+ "loss": 0.0022,
934
+ "learning_rate": 0.0002,
935
+ "epoch": 1.77,
936
+ "step": 99
937
+ },
938
+ {
939
+ "loss": 0.0025,
940
+ "learning_rate": 0.0002,
941
+ "epoch": 1.79,
942
+ "step": 100
943
+ },
944
+ {
945
+ "eval_math_exam_questions_loss": 0.214411199092865,
946
+ "eval_math_exam_questions_score": -0.03251197189092636,
947
+ "eval_math_exam_questions_brier_score": 0.03251197189092636,
948
+ "eval_math_exam_questions_average_probability": 0.9538560509681702,
949
+ "eval_math_exam_questions_accuracy": 0.96,
950
+ "eval_math_exam_questions_probabilities": [
951
+ 0.9958956241607666,
952
+ 0.8318325877189636,
953
+ 0.9999992847442627,
954
+ 1.0,
955
+ 8.109304872050416e-06,
956
+ 1.0,
957
+ 1.0,
958
+ 1.0,
959
+ 1.0,
960
+ 0.29970213770866394,
961
+ 0.9999974966049194,
962
+ 1.0,
963
+ 1.0,
964
+ 0.9999998807907104,
965
+ 1.0,
966
+ 1.0,
967
+ 1.0,
968
+ 1.0,
969
+ 0.9999997615814209,
970
+ 1.0,
971
+ 0.999977707862854,
972
+ 0.999778687953949,
973
+ 0.9999997615814209,
974
+ 0.30764925479888916,
975
+ 1.0,
976
+ 0.9999996423721313,
977
+ 1.0,
978
+ 0.9999992847442627,
979
+ 1.0,
980
+ 1.0,
981
+ 1.0,
982
+ 1.0,
983
+ 1.0,
984
+ 1.0,
985
+ 1.0,
986
+ 1.0,
987
+ 0.6386927366256714,
988
+ 1.0,
989
+ 0.9961808919906616,
990
+ 0.9999927282333374,
991
+ 0.9999998807907104,
992
+ 0.8549012541770935,
993
+ 0.9999998807907104,
994
+ 1.0,
995
+ 1.0,
996
+ 1.0,
997
+ 1.0,
998
+ 1.0,
999
+ 1.0,
1000
+ 1.0,
1001
+ 1.0,
1002
+ 0.9999997615814209,
1003
+ 1.0,
1004
+ 1.0,
1005
+ 1.0,
1006
+ 0.9999995231628418,
1007
+ 1.0,
1008
+ 0.9937735199928284,
1009
+ 1.0,
1010
+ 1.0,
1011
+ 0.9984048008918762,
1012
+ 1.0,
1013
+ 0.9988483190536499,
1014
+ 0.9999972581863403,
1015
+ 1.0,
1016
+ 1.0,
1017
+ 1.0,
1018
+ 0.9999980926513672,
1019
+ 0.9998224377632141,
1020
+ 0.9999912977218628,
1021
+ 1.0,
1022
+ 1.0,
1023
+ 0.7528015971183777,
1024
+ 0.003164754481986165,
1025
+ 0.9934700727462769,
1026
+ 0.9953631162643433,
1027
+ 1.0,
1028
+ 1.0,
1029
+ 0.9982792139053345,
1030
+ 0.9575594067573547,
1031
+ 0.9999983310699463,
1032
+ 1.0,
1033
+ 0.7891086935997009,
1034
+ 1.0,
1035
+ 1.0,
1036
+ 1.0,
1037
+ 1.0,
1038
+ 1.0,
1039
+ 1.0,
1040
+ 0.9999996423721313,
1041
+ 0.9999997615814209,
1042
+ 0.9999990463256836,
1043
+ 0.9808980822563171,
1044
+ 1.0,
1045
+ 1.0,
1046
+ 1.0,
1047
+ 1.0,
1048
+ 1.0,
1049
+ 1.0,
1050
+ 0.9995265007019043
1051
+ ],
1052
+ "eval_math_exam_questions_runtime": 58.6781,
1053
+ "eval_math_exam_questions_samples_per_second": 1.704,
1054
+ "eval_math_exam_questions_steps_per_second": 0.068,
1055
+ "epoch": 1.79,
1056
+ "step": 100
1057
+ },
1058
+ {
1059
+ "train_runtime": 4114.901,
1060
+ "train_samples_per_second": 0.778,
1061
+ "train_steps_per_second": 0.024,
1062
+ "total_flos": 0.0,
1063
+ "train_loss": 0.11991490689004422,
1064
+ "epoch": 1.79,
1065
+ "step": 100
1066
+ }
1067
+ ]]