Hugging Face's logo

ugaoo
/
deepseek_70b

PEFT
Safetensors
llama
Generated from Trainer
4-bit precision
bitsandbytes
Model card Files
xet
Community
ugaoo commited on
Commit
4f49dd6
·
verified ·
1 Parent(s): 2dc901c

Upload folder using huggingface_hub

Browse files
This view is limited to 50 files because it contains too many changes.   See raw diff
Files changed (50) hide show
  1. .gitattributes +7 -0
  2. README.md +145 -0
  3. adapter_config.json +40 -0
  4. adapter_model.safetensors +3 -0
  5. checkpoint-112/README.md +202 -0
  6. checkpoint-112/adapter_config.json +40 -0
  7. checkpoint-112/adapter_model.safetensors +3 -0
  8. checkpoint-112/global_step111/bf16_zero_pp_rank_0_mp_rank_00_optim_states.pt +3 -0
  9. checkpoint-112/global_step111/bf16_zero_pp_rank_1_mp_rank_00_optim_states.pt +3 -0
  10. checkpoint-112/global_step111/bf16_zero_pp_rank_2_mp_rank_00_optim_states.pt +3 -0
  11. checkpoint-112/global_step111/bf16_zero_pp_rank_3_mp_rank_00_optim_states.pt +3 -0
  12. checkpoint-112/global_step111/mp_rank_00_model_states.pt +3 -0
  13. checkpoint-112/latest +1 -0
  14. checkpoint-112/rng_state_0.pth +3 -0
  15. checkpoint-112/rng_state_1.pth +3 -0
  16. checkpoint-112/rng_state_2.pth +3 -0
  17. checkpoint-112/rng_state_3.pth +3 -0
  18. checkpoint-112/scheduler.pt +3 -0
  19. checkpoint-112/special_tokens_map.json +23 -0
  20. checkpoint-112/tokenizer.json +3 -0
  21. checkpoint-112/tokenizer_config.json +2067 -0
  22. checkpoint-112/trainer_state.json +817 -0
  23. checkpoint-112/training_args.bin +3 -0
  24. checkpoint-112/zero_to_fp32.py +604 -0
  25. checkpoint-140/README.md +202 -0
  26. checkpoint-140/adapter_config.json +40 -0
  27. checkpoint-140/adapter_model.safetensors +3 -0
  28. checkpoint-140/global_step138/bf16_zero_pp_rank_0_mp_rank_00_optim_states.pt +3 -0
  29. checkpoint-140/global_step138/bf16_zero_pp_rank_1_mp_rank_00_optim_states.pt +3 -0
  30. checkpoint-140/global_step138/bf16_zero_pp_rank_2_mp_rank_00_optim_states.pt +3 -0
  31. checkpoint-140/global_step138/bf16_zero_pp_rank_3_mp_rank_00_optim_states.pt +3 -0
  32. checkpoint-140/global_step138/mp_rank_00_model_states.pt +3 -0
  33. checkpoint-140/latest +1 -0
  34. checkpoint-140/rng_state_0.pth +3 -0
  35. checkpoint-140/rng_state_1.pth +3 -0
  36. checkpoint-140/rng_state_2.pth +3 -0
  37. checkpoint-140/rng_state_3.pth +3 -0
  38. checkpoint-140/scheduler.pt +3 -0
  39. checkpoint-140/special_tokens_map.json +23 -0
  40. checkpoint-140/tokenizer.json +3 -0
  41. checkpoint-140/tokenizer_config.json +2067 -0
  42. checkpoint-140/trainer_state.json +1013 -0
  43. checkpoint-140/training_args.bin +3 -0
  44. checkpoint-140/zero_to_fp32.py +604 -0
  45. checkpoint-168/README.md +202 -0
  46. checkpoint-168/adapter_config.json +40 -0
  47. checkpoint-168/adapter_model.safetensors +3 -0
  48. checkpoint-168/global_step166/bf16_zero_pp_rank_0_mp_rank_00_optim_states.pt +3 -0
  49. checkpoint-168/global_step166/bf16_zero_pp_rank_1_mp_rank_00_optim_states.pt +3 -0
  50. checkpoint-168/global_step166/bf16_zero_pp_rank_2_mp_rank_00_optim_states.pt +3 -0
.gitattributes CHANGED
@@ -33,3 +33,10 @@ 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
+ checkpoint-112/tokenizer.json filter=lfs diff=lfs merge=lfs -text
37
+ checkpoint-140/tokenizer.json filter=lfs diff=lfs merge=lfs -text
38
+ checkpoint-168/tokenizer.json filter=lfs diff=lfs merge=lfs -text
39
+ checkpoint-28/tokenizer.json filter=lfs diff=lfs merge=lfs -text
40
+ checkpoint-56/tokenizer.json filter=lfs diff=lfs merge=lfs -text
41
+ checkpoint-84/tokenizer.json filter=lfs diff=lfs merge=lfs -text
42
+ tokenizer.json filter=lfs diff=lfs merge=lfs -text
README.md ADDED
@@ -0,0 +1,145 @@
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
1
+ ---
2
+ library_name: peft
3
+ license: mit
4
+ base_model: deepseek-ai/DeepSeek-R1-Distill-Llama-70B
5
+ tags:
6
+ - generated_from_trainer
7
+ datasets:
8
+ - ugaoo/transformed_short_answer_dataset_prompt
9
+ model-index:
10
+ - name: out/transformed_short_answer_dataset_prompt_deepseek70bllama
11
+ results: []
12
+ ---
13
+
14
+ <!-- This model card has been generated automatically according to the information the Trainer had access to. You
15
+ should probably proofread and complete it, then remove this comment. -->
16
+
17
+ [<img src="https://raw.githubusercontent.com/axolotl-ai-cloud/axolotl/main/image/axolotl-badge-web.png" alt="Built with Axolotl" width="200" height="32"/>](https://github.com/axolotl-ai-cloud/axolotl)
18
+ <details><summary>See axolotl config</summary>
19
+
20
+ axolotl version: `0.8.0.dev0`
21
+ ```yaml
22
+ base_model: deepseek-ai/DeepSeek-R1-Distill-Llama-70B
23
+ model_type: AutoModelForCausalLM
24
+ tokenizer_type: AutoTokenizer
25
+ trust_remote_code: true
26
+
27
+ load_in_8bit: false
28
+ load_in_4bit: true
29
+ strict: false
30
+
31
+ datasets:
32
+ - path: ugaoo/transformed_short_answer_dataset_prompt
33
+ type: alpaca
34
+ val_set_size: 0
35
+ output_dir: ./out/transformed_short_answer_dataset_prompt_deepseek70bllama
36
+
37
+ sequence_len: 4000
38
+ sample_packing: true
39
+ pad_to_sequence_len: true
40
+
41
+ adapter: qlora
42
+ lora_r: 256
43
+ lora_alpha: 512
44
+ lora_dropout: 0.05
45
+ lora_target_linear: true
46
+ lora_target_modules:
47
+ - q_proj
48
+ - k_proj
49
+ - v_proj
50
+ - o_proj
51
+ - up_proj
52
+ - down_proj
53
+ - gate_proj
54
+ lora_modules_to_save:
55
+ - embed_tokens
56
+ - lm_head
57
+
58
+ wandb_project: cosmosearch
59
+ wandb_entity:
60
+ wandb_watch:
61
+ wandb_name: transformed_short_answer_dataset_prompt_deepseek70bllama
62
+ wandb_log_model:
63
+
64
+ gradient_accumulation_steps: 3
65
+ micro_batch_size: 4
66
+ num_epochs: 6
67
+ optimizer: adamw_torch
68
+ lr_scheduler: cosine
69
+ learning_rate: 5e-6
70
+
71
+ train_on_inputs: false
72
+ group_by_length: false
73
+ bf16: auto
74
+ fp16: false
75
+ tf32: false
76
+
77
+ gradient_checkpointing: true
78
+ early_stopping_patience:
79
+ resume_from_checkpoint:
80
+ logging_steps: 1
81
+ xformers_attention:
82
+ flash_attention: true
83
+
84
+ warmup_steps: 100
85
+ evals_per_epoch: 6
86
+ eval_table_size:
87
+ saves_per_epoch: 1
88
+ debug:
89
+ deepspeed:
90
+ weight_decay: 0.0
91
+ fsdp:
92
+ fsdp_config:
93
+ save_total_limit: 6
94
+ special_tokens:
95
+ pad_token: <｜end▁of▁sentence｜>
96
+ ```
97
+
98
+ </details><br>
99
+
100
+ # out/transformed_short_answer_dataset_prompt_deepseek70bllama
101
+
102
+ This model is a fine-tuned version of [deepseek-ai/DeepSeek-R1-Distill-Llama-70B](https://huggingface.co/deepseek-ai/DeepSeek-R1-Distill-Llama-70B) on the ugaoo/transformed_short_answer_dataset_prompt dataset.
103
+
104
+ ## Model description
105
+
106
+ More information needed
107
+
108
+ ## Intended uses & limitations
109
+
110
+ More information needed
111
+
112
+ ## Training and evaluation data
113
+
114
+ More information needed
115
+
116
+ ## Training procedure
117
+
118
+ ### Training hyperparameters
119
+
120
+ The following hyperparameters were used during training:
121
+ - learning_rate: 5e-06
122
+ - train_batch_size: 4
123
+ - eval_batch_size: 4
124
+ - seed: 42
125
+ - distributed_type: multi-GPU
126
+ - num_devices: 4
127
+ - gradient_accumulation_steps: 3
128
+ - total_train_batch_size: 48
129
+ - total_eval_batch_size: 16
130
+ - optimizer: Use OptimizerNames.ADAMW_TORCH with betas=(0.9,0.999) and epsilon=1e-08 and optimizer_args=No additional optimizer arguments
131
+ - lr_scheduler_type: cosine
132
+ - lr_scheduler_warmup_steps: 100
133
+ - num_epochs: 6.0
134
+
135
+ ### Training results
136
+
137
+
138
+
139
+ ### Framework versions
140
+
141
+ - PEFT 0.14.0
142
+ - Transformers 4.49.0
143
+ - Pytorch 2.5.1+cu124
144
+ - Datasets 3.2.0
145
+ - Tokenizers 0.21.1
adapter_config.json ADDED
@@ -0,0 +1,40 @@
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
1
+ {
2
+ "alpha_pattern": {},
3
+ "auto_mapping": null,
4
+ "base_model_name_or_path": "deepseek-ai/DeepSeek-R1-Distill-Llama-70B",
5
+ "bias": "none",
6
+ "eva_config": null,
7
+ "exclude_modules": null,
8
+ "fan_in_fan_out": null,
9
+ "inference_mode": true,
10
+ "init_lora_weights": true,
11
+ "layer_replication": null,
12
+ "layers_pattern": null,
13
+ "layers_to_transform": null,
14
+ "loftq_config": {},
15
+ "lora_alpha": 512,
16
+ "lora_bias": false,
17
+ "lora_dropout": 0.05,
18
+ "megatron_config": null,
19
+ "megatron_core": "megatron.core",
20
+ "modules_to_save": [
21
+ "embed_tokens",
22
+ "lm_head"
23
+ ],
24
+ "peft_type": "LORA",
25
+ "r": 256,
26
+ "rank_pattern": {},
27
+ "revision": null,
28
+ "target_modules": [
29
+ "q_proj",
30
+ "o_proj",
31
+ "down_proj",
32
+ "gate_proj",
33
+ "v_proj",
34
+ "k_proj",
35
+ "up_proj"
36
+ ],
37
+ "task_type": "CAUSAL_LM",
38
+ "use_dora": false,
39
+ "use_rslora": false
40
+ }
adapter_model.safetensors ADDED
@@ -0,0 +1,3 @@
 
 
 
 
1
+ version https://git-lfs.github.com/spec/v1
2
+ oid sha256:d2b845ca0fbcec9230e875bb8f2fe063ff8b87bbe037863318ec4c7f8ffc431c
3
+ size 10829849744
checkpoint-112/README.md ADDED
@@ -0,0 +1,202 @@
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
1
+ ---
2
+ base_model: deepseek-ai/DeepSeek-R1-Distill-Llama-70B
3
+ library_name: peft
4
+ ---
5
+
6
+ # Model Card for Model ID
7
+
8
+ <!-- Provide a quick summary of what the model is/does. -->
9
+
10
+
11
+
12
+ ## Model Details
13
+
14
+ ### Model Description
15
+
16
+ <!-- Provide a longer summary of what this model is. -->
17
+
18
+
19
+
20
+ - **Developed by:** [More Information Needed]
21
+ - **Funded by [optional]:** [More Information Needed]
22
+ - **Shared by [optional]:** [More Information Needed]
23
+ - **Model type:** [More Information Needed]
24
+ - **Language(s) (NLP):** [More Information Needed]
25
+ - **License:** [More Information Needed]
26
+ - **Finetuned from model [optional]:** [More Information Needed]
27
+
28
+ ### Model Sources [optional]
29
+
30
+ <!-- Provide the basic links for the model. -->
31
+
32
+ - **Repository:** [More Information Needed]
33
+ - **Paper [optional]:** [More Information Needed]
34
+ - **Demo [optional]:** [More Information Needed]
35
+
36
+ ## Uses
37
+
38
+ <!-- Address questions around how the model is intended to be used, including the foreseeable users of the model and those affected by the model. -->
39
+
40
+ ### Direct Use
41
+
42
+ <!-- This section is for the model use without fine-tuning or plugging into a larger ecosystem/app. -->
43
+
44
+ [More Information Needed]
45
+
46
+ ### Downstream Use [optional]
47
+
48
+ <!-- This section is for the model use when fine-tuned for a task, or when plugged into a larger ecosystem/app -->
49
+
50
+ [More Information Needed]
51
+
52
+ ### Out-of-Scope Use
53
+
54
+ <!-- This section addresses misuse, malicious use, and uses that the model will not work well for. -->
55
+
56
+ [More Information Needed]
57
+
58
+ ## Bias, Risks, and Limitations
59
+
60
+ <!-- This section is meant to convey both technical and sociotechnical limitations. -->
61
+
62
+ [More Information Needed]
63
+
64
+ ### Recommendations
65
+
66
+ <!-- This section is meant to convey recommendations with respect to the bias, risk, and technical limitations. -->
67
+
68
+ Users (both direct and downstream) should be made aware of the risks, biases and limitations of the model. More information needed for further recommendations.
69
+
70
+ ## How to Get Started with the Model
71
+
72
+ Use the code below to get started with the model.
73
+
74
+ [More Information Needed]
75
+
76
+ ## Training Details
77
+
78
+ ### Training Data
79
+
80
+ <!-- This should link to a Dataset Card, perhaps with a short stub of information on what the training data is all about as well as documentation related to data pre-processing or additional filtering. -->
81
+
82
+ [More Information Needed]
83
+
84
+ ### Training Procedure
85
+
86
+ <!-- This relates heavily to the Technical Specifications. Content here should link to that section when it is relevant to the training procedure. -->
87
+
88
+ #### Preprocessing [optional]
89
+
90
+ [More Information Needed]
91
+
92
+
93
+ #### Training Hyperparameters
94
+
95
+ - **Training regime:** [More Information Needed] <!--fp32, fp16 mixed precision, bf16 mixed precision, bf16 non-mixed precision, fp16 non-mixed precision, fp8 mixed precision -->
96
+
97
+ #### Speeds, Sizes, Times [optional]
98
+
99
+ <!-- This section provides information about throughput, start/end time, checkpoint size if relevant, etc. -->
100
+
101
+ [More Information Needed]
102
+
103
+ ## Evaluation
104
+
105
+ <!-- This section describes the evaluation protocols and provides the results. -->
106
+
107
+ ### Testing Data, Factors & Metrics
108
+
109
+ #### Testing Data
110
+
111
+ <!-- This should link to a Dataset Card if possible. -->
112
+
113
+ [More Information Needed]
114
+
115
+ #### Factors
116
+
117
+ <!-- These are the things the evaluation is disaggregating by, e.g., subpopulations or domains. -->
118
+
119
+ [More Information Needed]
120
+
121
+ #### Metrics
122
+
123
+ <!-- These are the evaluation metrics being used, ideally with a description of why. -->
124
+
125
+ [More Information Needed]
126
+
127
+ ### Results
128
+
129
+ [More Information Needed]
130
+
131
+ #### Summary
132
+
133
+
134
+
135
+ ## Model Examination [optional]
136
+
137
+ <!-- Relevant interpretability work for the model goes here -->
138
+
139
+ [More Information Needed]
140
+
141
+ ## Environmental Impact
142
+
143
+ <!-- Total emissions (in grams of CO2eq) and additional considerations, such as electricity usage, go here. Edit the suggested text below accordingly -->
144
+
145
+ Carbon emissions can be estimated using the [Machine Learning Impact calculator](https://mlco2.github.io/impact#compute) presented in [Lacoste et al. (2019)](https://arxiv.org/abs/1910.09700).
146
+
147
+ - **Hardware Type:** [More Information Needed]
148
+ - **Hours used:** [More Information Needed]
149
+ - **Cloud Provider:** [More Information Needed]
150
+ - **Compute Region:** [More Information Needed]
151
+ - **Carbon Emitted:** [More Information Needed]
152
+
153
+ ## Technical Specifications [optional]
154
+
155
+ ### Model Architecture and Objective
156
+
157
+ [More Information Needed]
158
+
159
+ ### Compute Infrastructure
160
+
161
+ [More Information Needed]
162
+
163
+ #### Hardware
164
+
165
+ [More Information Needed]
166
+
167
+ #### Software
168
+
169
+ [More Information Needed]
170
+
171
+ ## Citation [optional]
172
+
173
+ <!-- If there is a paper or blog post introducing the model, the APA and Bibtex information for that should go in this section. -->
174
+
175
+ **BibTeX:**
176
+
177
+ [More Information Needed]
178
+
179
+ **APA:**
180
+
181
+ [More Information Needed]
182
+
183
+ ## Glossary [optional]
184
+
185
+ <!-- If relevant, include terms and calculations in this section that can help readers understand the model or model card. -->
186
+
187
+ [More Information Needed]
188
+
189
+ ## More Information [optional]
190
+
191
+ [More Information Needed]
192
+
193
+ ## Model Card Authors [optional]
194
+
195
+ [More Information Needed]
196
+
197
+ ## Model Card Contact
198
+
199
+ [More Information Needed]
200
+ ### Framework versions
201
+
202
+ - PEFT 0.14.0
checkpoint-112/adapter_config.json ADDED
@@ -0,0 +1,40 @@
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
1
+ {
2
+ "alpha_pattern": {},
3
+ "auto_mapping": null,
4
+ "base_model_name_or_path": "deepseek-ai/DeepSeek-R1-Distill-Llama-70B",
5
+ "bias": "none",
6
+ "eva_config": null,
7
+ "exclude_modules": null,
8
+ "fan_in_fan_out": null,
9
+ "inference_mode": true,
10
+ "init_lora_weights": true,
11
+ "layer_replication": null,
12
+ "layers_pattern": null,
13
+ "layers_to_transform": null,
14
+ "loftq_config": {},
15
+ "lora_alpha": 512,
16
+ "lora_bias": false,
17
+ "lora_dropout": 0.05,
18
+ "megatron_config": null,
19
+ "megatron_core": "megatron.core",
20
+ "modules_to_save": [
21
+ "embed_tokens",
22
+ "lm_head"
23
+ ],
24
+ "peft_type": "LORA",
25
+ "r": 256,
26
+ "rank_pattern": {},
27
+ "revision": null,
28
+ "target_modules": [
29
+ "q_proj",
30
+ "o_proj",
31
+ "down_proj",
32
+ "gate_proj",
33
+ "v_proj",
34
+ "k_proj",
35
+ "up_proj"
36
+ ],
37
+ "task_type": "CAUSAL_LM",
38
+ "use_dora": false,
39
+ "use_rslora": false
40
+ }
checkpoint-112/adapter_model.safetensors ADDED
@@ -0,0 +1,3 @@
 
 
 
 
1
+ version https://git-lfs.github.com/spec/v1
2
+ oid sha256:6fd58fb1e4353caf6551f6af926b3b635bbbd2ff8e0dba9dea29f7649c033312
3
+ size 10829849744
checkpoint-112/global_step111/bf16_zero_pp_rank_0_mp_rank_00_optim_states.pt ADDED
@@ -0,0 +1,3 @@
 
 
 
 
1
+ version https://git-lfs.github.com/spec/v1
2
+ oid sha256:7d7208506736a5f8d58ea54bfc29d8842104ce07d1aba571a5d17b12571c1a5c
3
+ size 16244554780
checkpoint-112/global_step111/bf16_zero_pp_rank_1_mp_rank_00_optim_states.pt ADDED
@@ -0,0 +1,3 @@
 
 
 
 
1
+ version https://git-lfs.github.com/spec/v1
2
+ oid sha256:1948c9af7a3141297998a55909b14a59f8362ded552a4d78c209a1a6ff4cf1ad
3
+ size 16244593500
checkpoint-112/global_step111/bf16_zero_pp_rank_2_mp_rank_00_optim_states.pt ADDED
@@ -0,0 +1,3 @@
 
 
 
 
1
+ version https://git-lfs.github.com/spec/v1
2
+ oid sha256:9e2334651fd2114670bb18d81f390612577dc6c6f1272655d2858102aaa2f978
3
+ size 16244594204
checkpoint-112/global_step111/bf16_zero_pp_rank_3_mp_rank_00_optim_states.pt ADDED
@@ -0,0 +1,3 @@
 
 
 
 
1
+ version https://git-lfs.github.com/spec/v1
2
+ oid sha256:071c079f277f47734770a29c89c7d7f99e46af3479913ddda07b4fb43317de75
3
+ size 16244554268
checkpoint-112/global_step111/mp_rank_00_model_states.pt ADDED
@@ -0,0 +1,3 @@
 
 
 
 
1
+ version https://git-lfs.github.com/spec/v1
2
+ oid sha256:b7f6bee5ff4e7f952c760e7ec6429311517592c8d43648d370619d9b7494028d
3
+ size 11918643933
checkpoint-112/latest ADDED
@@ -0,0 +1 @@
 
 
1
+ global_step111
checkpoint-112/rng_state_0.pth ADDED
@@ -0,0 +1,3 @@
 
 
 
 
1
+ version https://git-lfs.github.com/spec/v1
2
+ oid sha256:8c4a83536d833fa6829308cb2a5e932f2be1fc396261874a5a721fc16cbe75db
3
+ size 15024
checkpoint-112/rng_state_1.pth ADDED
@@ -0,0 +1,3 @@
 
 
 
 
1
+ version https://git-lfs.github.com/spec/v1
2
+ oid sha256:5d9fd6e5a37ed2d05ed69dd54e4cfbffe064f77bfb918562d470d0dbfa0e948f
3
+ size 15024
checkpoint-112/rng_state_2.pth ADDED
@@ -0,0 +1,3 @@
 
 
 
 
1
+ version https://git-lfs.github.com/spec/v1
2
+ oid sha256:798098cfd68f8db180daf7e2f77e995e711e8e60fdf4ab15c3ff55b543b668bd
3
+ size 15024
checkpoint-112/rng_state_3.pth ADDED
@@ -0,0 +1,3 @@
 
 
 
 
1
+ version https://git-lfs.github.com/spec/v1
2
+ oid sha256:0493e8749745659c1452d4d3bb363fa5aec6ecf3fbfa603780fc025ff158da3e
3
+ size 15024
checkpoint-112/scheduler.pt ADDED
@@ -0,0 +1,3 @@
 
 
 
 
1
+ version https://git-lfs.github.com/spec/v1
2
+ oid sha256:64d234916b239591a29032d33744a24bcff84c476af45ee59319641c78a7e735
3
+ size 1064
checkpoint-112/special_tokens_map.json ADDED
@@ -0,0 +1,23 @@
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
1
+ {
2
+ "bos_token": {
3
+ "content": "<｜begin▁of▁sentence｜>",
4
+ "lstrip": false,
5
+ "normalized": false,
6
+ "rstrip": false,
7
+ "single_word": false
8
+ },
9
+ "eos_token": {
10
+ "content": "<｜end▁of▁sentence｜>",
11
+ "lstrip": false,
12
+ "normalized": false,
13
+ "rstrip": false,
14
+ "single_word": false
15
+ },
16
+ "pad_token": {
17
+ "content": "<｜end▁of▁sentence｜>",
18
+ "lstrip": false,
19
+ "normalized": false,
20
+ "rstrip": false,
21
+ "single_word": false
22
+ }
23
+ }
checkpoint-112/tokenizer.json ADDED
@@ -0,0 +1,3 @@
 
 
 
 
1
+ version https://git-lfs.github.com/spec/v1
2
+ oid sha256:d91915040cfac999d8c55f4b5bc6e67367c065e3a7a4e4b9438ce1f256addd86
3
+ size 17209530
checkpoint-112/tokenizer_config.json ADDED
@@ -0,0 +1,2067 @@
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
1
+ {
2
+ "add_bos_token": true,
3
+ "add_eos_token": false,
4
+ "add_prefix_space": null,
5
+ "added_tokens_decoder": {
6
+ "128000": {
7
+ "content": "<｜begin▁of▁sentence｜>",
8
+ "lstrip": false,
9
+ "normalized": false,
10
+ "rstrip": false,
11
+ "single_word": false,
12
+ "special": true
13
+ },
14
+ "128001": {
15
+ "content": "<｜end▁of▁sentence｜>",
16
+ "lstrip": false,
17
+ "normalized": false,
18
+ "rstrip": false,
19
+ "single_word": false,
20
+ "special": true
21
+ },
22
+ "128002": {
23
+ "content": "<|reserved_special_token_0|>",
24
+ "lstrip": false,
25
+ "normalized": false,
26
+ "rstrip": false,
27
+ "single_word": false,
28
+ "special": true
29
+ },
30
+ "128003": {
31
+ "content": "<|reserved_special_token_1|>",
32
+ "lstrip": false,
33
+ "normalized": false,
34
+ "rstrip": false,
35
+ "single_word": false,
36
+ "special": true
37
+ },
38
+ "128004": {
39
+ "content": "<|finetune_right_pad_id|>",
40
+ "lstrip": false,
41
+ "normalized": false,
42
+ "rstrip": false,
43
+ "single_word": false,
44
+ "special": true
45
+ },
46
+ "128005": {
47
+ "content": "<|reserved_special_token_2|>",
48
+ "lstrip": false,
49
+ "normalized": false,
50
+ "rstrip": false,
51
+ "single_word": false,
52
+ "special": true
53
+ },
54
+ "128006": {
55
+ "content": "<|start_header_id|>",
56
+ "lstrip": false,
57
+ "normalized": false,
58
+ "rstrip": false,
59
+ "single_word": false,
60
+ "special": true
61
+ },
62
+ "128007": {
63
+ "content": "<|end_header_id|>",
64
+ "lstrip": false,
65
+ "normalized": false,
66
+ "rstrip": false,
67
+ "single_word": false,
68
+ "special": true
69
+ },
70
+ "128008": {
71
+ "content": "<|eom_id|>",
72
+ "lstrip": false,
73
+ "normalized": false,
74
+ "rstrip": false,
75
+ "single_word": false,
76
+ "special": true
77
+ },
78
+ "128009": {
79
+ "content": "<|eot_id|>",
80
+ "lstrip": false,
81
+ "normalized": false,
82
+ "rstrip": false,
83
+ "single_word": false,
84
+ "special": true
85
+ },
86
+ "128010": {
87
+ "content": "<|python_tag|>",
88
+ "lstrip": false,
89
+ "normalized": false,
90
+ "rstrip": false,
91
+ "single_word": false,
92
+ "special": true
93
+ },
94
+ "128011": {
95
+ "content": "<｜User｜>",
96
+ "lstrip": false,
97
+ "normalized": false,
98
+ "rstrip": false,
99
+ "single_word": false,
100
+ "special": false
101
+ },
102
+ "128012": {
103
+ "content": "<｜Assistant｜>",
104
+ "lstrip": false,
105
+ "normalized": false,
106
+ "rstrip": false,
107
+ "single_word": false,
108
+ "special": false
109
+ },
110
+ "128013": {
111
+ "content": "<think>",
112
+ "lstrip": false,
113
+ "normalized": false,
114
+ "rstrip": false,
115
+ "single_word": false,
116
+ "special": false
117
+ },
118
+ "128014": {
119
+ "content": "</think>",
120
+ "lstrip": false,
121
+ "normalized": false,
122
+ "rstrip": false,
123
+ "single_word": false,
124
+ "special": false
125
+ },
126
+ "128015": {
127
+ "content": "<｜▁pad▁｜>",
128
+ "lstrip": false,
129
+ "normalized": false,
130
+ "rstrip": false,
131
+ "single_word": false,
132
+ "special": true
133
+ },
134
+ "128016": {
135
+ "content": "<|reserved_special_token_8|>",
136
+ "lstrip": false,
137
+ "normalized": false,
138
+ "rstrip": false,
139
+ "single_word": false,
140
+ "special": true
141
+ },
142
+ "128017": {
143
+ "content": "<|reserved_special_token_9|>",
144
+ "lstrip": false,
145
+ "normalized": false,
146
+ "rstrip": false,
147
+ "single_word": false,
148
+ "special": true
149
+ },
150
+ "128018": {
151
+ "content": "<|reserved_special_token_10|>",
152
+ "lstrip": false,
153
+ "normalized": false,
154
+ "rstrip": false,
155
+ "single_word": false,
156
+ "special": true
157
+ },
158
+ "128019": {
159
+ "content": "<|reserved_special_token_11|>",
160
+ "lstrip": false,
161
+ "normalized": false,
162
+ "rstrip": false,
163
+ "single_word": false,
164
+ "special": true
165
+ },
166
+ "128020": {
167
+ "content": "<|reserved_special_token_12|>",
168
+ "lstrip": false,
169
+ "normalized": false,
170
+ "rstrip": false,
171
+ "single_word": false,
172
+ "special": true
173
+ },
174
+ "128021": {
175
+ "content": "<|reserved_special_token_13|>",
176
+ "lstrip": false,
177
+ "normalized": false,
178
+ "rstrip": false,
179
+ "single_word": false,
180
+ "special": true
181
+ },
182
+ "128022": {
183
+ "content": "<|reserved_special_token_14|>",
184
+ "lstrip": false,
185
+ "normalized": false,
186
+ "rstrip": false,
187
+ "single_word": false,
188
+ "special": true
189
+ },
190
+ "128023": {
191
+ "content": "<|reserved_special_token_15|>",
192
+ "lstrip": false,
193
+ "normalized": false,
194
+ "rstrip": false,
195
+ "single_word": false,
196
+ "special": true
197
+ },
198
+ "128024": {
199
+ "content": "<|reserved_special_token_16|>",
200
+ "lstrip": false,
201
+ "normalized": false,
202
+ "rstrip": false,
203
+ "single_word": false,
204
+ "special": true
205
+ },
206
+ "128025": {
207
+ "content": "<|reserved_special_token_17|>",
208
+ "lstrip": false,
209
+ "normalized": false,
210
+ "rstrip": false,
211
+ "single_word": false,
212
+ "special": true
213
+ },
214
+ "128026": {
215
+ "content": "<|reserved_special_token_18|>",
216
+ "lstrip": false,
217
+ "normalized": false,
218
+ "rstrip": false,
219
+ "single_word": false,
220
+ "special": true
221
+ },
222
+ "128027": {
223
+ "content": "<|reserved_special_token_19|>",
224
+ "lstrip": false,
225
+ "normalized": false,
226
+ "rstrip": false,
227
+ "single_word": false,
228
+ "special": true
229
+ },
230
+ "128028": {
231
+ "content": "<|reserved_special_token_20|>",
232
+ "lstrip": false,
233
+ "normalized": false,
234
+ "rstrip": false,
235
+ "single_word": false,
236
+ "special": true
237
+ },
238
+ "128029": {
239
+ "content": "<|reserved_special_token_21|>",
240
+ "lstrip": false,
241
+ "normalized": false,
242
+ "rstrip": false,
243
+ "single_word": false,
244
+ "special": true
245
+ },
246
+ "128030": {
247
+ "content": "<|reserved_special_token_22|>",
248
+ "lstrip": false,
249
+ "normalized": false,
250
+ "rstrip": false,
251
+ "single_word": false,
252
+ "special": true
253
+ },
254
+ "128031": {
255
+ "content": "<|reserved_special_token_23|>",
256
+ "lstrip": false,
257
+ "normalized": false,
258
+ "rstrip": false,
259
+ "single_word": false,
260
+ "special": true
261
+ },
262
+ "128032": {
263
+ "content": "<|reserved_special_token_24|>",
264
+ "lstrip": false,
265
+ "normalized": false,
266
+ "rstrip": false,
267
+ "single_word": false,
268
+ "special": true
269
+ },
270
+ "128033": {
271
+ "content": "<|reserved_special_token_25|>",
272
+ "lstrip": false,
273
+ "normalized": false,
274
+ "rstrip": false,
275
+ "single_word": false,
276
+ "special": true
277
+ },
278
+ "128034": {
279
+ "content": "<|reserved_special_token_26|>",
280
+ "lstrip": false,
281
+ "normalized": false,
282
+ "rstrip": false,
283
+ "single_word": false,
284
+ "special": true
285
+ },
286
+ "128035": {
287
+ "content": "<|reserved_special_token_27|>",
288
+ "lstrip": false,
289
+ "normalized": false,
290
+ "rstrip": false,
291
+ "single_word": false,
292
+ "special": true
293
+ },
294
+ "128036": {
295
+ "content": "<|reserved_special_token_28|>",
296
+ "lstrip": false,
297
+ "normalized": false,
298
+ "rstrip": false,
299
+ "single_word": false,
300
+ "special": true
301
+ },
302
+ "128037": {
303
+ "content": "<|reserved_special_token_29|>",
304
+ "lstrip": false,
305
+ "normalized": false,
306
+ "rstrip": false,
307
+ "single_word": false,
308
+ "special": true
309
+ },
310
+ "128038": {
311
+ "content": "<|reserved_special_token_30|>",
312
+ "lstrip": false,
313
+ "normalized": false,
314
+ "rstrip": false,
315
+ "single_word": false,
316
+ "special": true
317
+ },
318
+ "128039": {
319
+ "content": "<|reserved_special_token_31|>",
320
+ "lstrip": false,
321
+ "normalized": false,
322
+ "rstrip": false,
323
+ "single_word": false,
324
+ "special": true
325
+ },
326
+ "128040": {
327
+ "content": "<|reserved_special_token_32|>",
328
+ "lstrip": false,
329
+ "normalized": false,
330
+ "rstrip": false,
331
+ "single_word": false,
332
+ "special": true
333
+ },
334
+ "128041": {
335
+ "content": "<|reserved_special_token_33|>",
336
+ "lstrip": false,
337
+ "normalized": false,
338
+ "rstrip": false,
339
+ "single_word": false,
340
+ "special": true
341
+ },
342
+ "128042": {
343
+ "content": "<|reserved_special_token_34|>",
344
+ "lstrip": false,
345
+ "normalized": false,
346
+ "rstrip": false,
347
+ "single_word": false,
348
+ "special": true
349
+ },
350
+ "128043": {
351
+ "content": "<|reserved_special_token_35|>",
352
+ "lstrip": false,
353
+ "normalized": false,
354
+ "rstrip": false,
355
+ "single_word": false,
356
+ "special": true
357
+ },
358
+ "128044": {
359
+ "content": "<|reserved_special_token_36|>",
360
+ "lstrip": false,
361
+ "normalized": false,
362
+ "rstrip": false,
363
+ "single_word": false,
364
+ "special": true
365
+ },
366
+ "128045": {
367
+ "content": "<|reserved_special_token_37|>",
368
+ "lstrip": false,
369
+ "normalized": false,
370
+ "rstrip": false,
371
+ "single_word": false,
372
+ "special": true
373
+ },
374
+ "128046": {
375
+ "content": "<|reserved_special_token_38|>",
376
+ "lstrip": false,
377
+ "normalized": false,
378
+ "rstrip": false,
379
+ "single_word": false,
380
+ "special": true
381
+ },
382
+ "128047": {
383
+ "content": "<|reserved_special_token_39|>",
384
+ "lstrip": false,
385
+ "normalized": false,
386
+ "rstrip": false,
387
+ "single_word": false,
388
+ "special": true
389
+ },
390
+ "128048": {
391
+ "content": "<|reserved_special_token_40|>",
392
+ "lstrip": false,
393
+ "normalized": false,
394
+ "rstrip": false,
395
+ "single_word": false,
396
+ "special": true
397
+ },
398
+ "128049": {
399
+ "content": "<|reserved_special_token_41|>",
400
+ "lstrip": false,
401
+ "normalized": false,
402
+ "rstrip": false,
403
+ "single_word": false,
404
+ "special": true
405
+ },
406
+ "128050": {
407
+ "content": "<|reserved_special_token_42|>",
408
+ "lstrip": false,
409
+ "normalized": false,
410
+ "rstrip": false,
411
+ "single_word": false,
412
+ "special": true
413
+ },
414
+ "128051": {
415
+ "content": "<|reserved_special_token_43|>",
416
+ "lstrip": false,
417
+ "normalized": false,
418
+ "rstrip": false,
419
+ "single_word": false,
420
+ "special": true
421
+ },
422
+ "128052": {
423
+ "content": "<|reserved_special_token_44|>",
424
+ "lstrip": false,
425
+ "normalized": false,
426
+ "rstrip": false,
427
+ "single_word": false,
428
+ "special": true
429
+ },
430
+ "128053": {
431
+ "content": "<|reserved_special_token_45|>",
432
+ "lstrip": false,
433
+ "normalized": false,
434
+ "rstrip": false,
435
+ "single_word": false,
436
+ "special": true
437
+ },
438
+ "128054": {
439
+ "content": "<|reserved_special_token_46|>",
440
+ "lstrip": false,
441
+ "normalized": false,
442
+ "rstrip": false,
443
+ "single_word": false,
444
+ "special": true
445
+ },
446
+ "128055": {
447
+ "content": "<|reserved_special_token_47|>",
448
+ "lstrip": false,
449
+ "normalized": false,
450
+ "rstrip": false,
451
+ "single_word": false,
452
+ "special": true
453
+ },
454
+ "128056": {
455
+ "content": "<|reserved_special_token_48|>",
456
+ "lstrip": false,
457
+ "normalized": false,
458
+ "rstrip": false,
459
+ "single_word": false,
460
+ "special": true
461
+ },
462
+ "128057": {
463
+ "content": "<|reserved_special_token_49|>",
464
+ "lstrip": false,
465
+ "normalized": false,
466
+ "rstrip": false,
467
+ "single_word": false,
468
+ "special": true
469
+ },
470
+ "128058": {
471
+ "content": "<|reserved_special_token_50|>",
472
+ "lstrip": false,
473
+ "normalized": false,
474
+ "rstrip": false,
475
+ "single_word": false,
476
+ "special": true
477
+ },
478
+ "128059": {
479
+ "content": "<|reserved_special_token_51|>",
480
+ "lstrip": false,
481
+ "normalized": false,
482
+ "rstrip": false,
483
+ "single_word": false,
484
+ "special": true
485
+ },
486
+ "128060": {
487
+ "content": "<|reserved_special_token_52|>",
488
+ "lstrip": false,
489
+ "normalized": false,
490
+ "rstrip": false,
491
+ "single_word": false,
492
+ "special": true
493
+ },
494
+ "128061": {
495
+ "content": "<|reserved_special_token_53|>",
496
+ "lstrip": false,
497
+ "normalized": false,
498
+ "rstrip": false,
499
+ "single_word": false,
500
+ "special": true
501
+ },
502
+ "128062": {
503
+ "content": "<|reserved_special_token_54|>",
504
+ "lstrip": false,
505
+ "normalized": false,
506
+ "rstrip": false,
507
+ "single_word": false,
508
+ "special": true
509
+ },
510
+ "128063": {
511
+ "content": "<|reserved_special_token_55|>",
512
+ "lstrip": false,
513
+ "normalized": false,
514
+ "rstrip": false,
515
+ "single_word": false,
516
+ "special": true
517
+ },
518
+ "128064": {
519
+ "content": "<|reserved_special_token_56|>",
520
+ "lstrip": false,
521
+ "normalized": false,
522
+ "rstrip": false,
523
+ "single_word": false,
524
+ "special": true
525
+ },
526
+ "128065": {
527
+ "content": "<|reserved_special_token_57|>",
528
+ "lstrip": false,
529
+ "normalized": false,
530
+ "rstrip": false,
531
+ "single_word": false,
532
+ "special": true
533
+ },
534
+ "128066": {
535
+ "content": "<|reserved_special_token_58|>",
536
+ "lstrip": false,
537
+ "normalized": false,
538
+ "rstrip": false,
539
+ "single_word": false,
540
+ "special": true
541
+ },
542
+ "128067": {
543
+ "content": "<|reserved_special_token_59|>",
544
+ "lstrip": false,
545
+ "normalized": false,
546
+ "rstrip": false,
547
+ "single_word": false,
548
+ "special": true
549
+ },
550
+ "128068": {
551
+ "content": "<|reserved_special_token_60|>",
552
+ "lstrip": false,
553
+ "normalized": false,
554
+ "rstrip": false,
555
+ "single_word": false,
556
+ "special": true
557
+ },
558
+ "128069": {
559
+ "content": "<|reserved_special_token_61|>",
560
+ "lstrip": false,
561
+ "normalized": false,
562
+ "rstrip": false,
563
+ "single_word": false,
564
+ "special": true
565
+ },
566
+ "128070": {
567
+ "content": "<|reserved_special_token_62|>",
568
+ "lstrip": false,
569
+ "normalized": false,
570
+ "rstrip": false,
571
+ "single_word": false,
572
+ "special": true
573
+ },
574
+ "128071": {
575
+ "content": "<|reserved_special_token_63|>",
576
+ "lstrip": false,
577
+ "normalized": false,
578
+ "rstrip": false,
579
+ "single_word": false,
580
+ "special": true
581
+ },
582
+ "128072": {
583
+ "content": "<|reserved_special_token_64|>",
584
+ "lstrip": false,
585
+ "normalized": false,
586
+ "rstrip": false,
587
+ "single_word": false,
588
+ "special": true
589
+ },
590
+ "128073": {
591
+ "content": "<|reserved_special_token_65|>",
592
+ "lstrip": false,
593
+ "normalized": false,
594
+ "rstrip": false,
595
+ "single_word": false,
596
+ "special": true
597
+ },
598
+ "128074": {
599
+ "content": "<|reserved_special_token_66|>",
600
+ "lstrip": false,
601
+ "normalized": false,
602
+ "rstrip": false,
603
+ "single_word": false,
604
+ "special": true
605
+ },
606
+ "128075": {
607
+ "content": "<|reserved_special_token_67|>",
608
+ "lstrip": false,
609
+ "normalized": false,
610
+ "rstrip": false,
611
+ "single_word": false,
612
+ "special": true
613
+ },
614
+ "128076": {
615
+ "content": "<|reserved_special_token_68|>",
616
+ "lstrip": false,
617
+ "normalized": false,
618
+ "rstrip": false,
619
+ "single_word": false,
620
+ "special": true
621
+ },
622
+ "128077": {
623
+ "content": "<|reserved_special_token_69|>",
624
+ "lstrip": false,
625
+ "normalized": false,
626
+ "rstrip": false,
627
+ "single_word": false,
628
+ "special": true
629
+ },
630
+ "128078": {
631
+ "content": "<|reserved_special_token_70|>",
632
+ "lstrip": false,
633
+ "normalized": false,
634
+ "rstrip": false,
635
+ "single_word": false,
636
+ "special": true
637
+ },
638
+ "128079": {
639
+ "content": "<|reserved_special_token_71|>",
640
+ "lstrip": false,
641
+ "normalized": false,
642
+ "rstrip": false,
643
+ "single_word": false,
644
+ "special": true
645
+ },
646
+ "128080": {
647
+ "content": "<|reserved_special_token_72|>",
648
+ "lstrip": false,
649
+ "normalized": false,
650
+ "rstrip": false,
651
+ "single_word": false,
652
+ "special": true
653
+ },
654
+ "128081": {
655
+ "content": "<|reserved_special_token_73|>",
656
+ "lstrip": false,
657
+ "normalized": false,
658
+ "rstrip": false,
659
+ "single_word": false,
660
+ "special": true
661
+ },
662
+ "128082": {
663
+ "content": "<|reserved_special_token_74|>",
664
+ "lstrip": false,
665
+ "normalized": false,
666
+ "rstrip": false,
667
+ "single_word": false,
668
+ "special": true
669
+ },
670
+ "128083": {
671
+ "content": "<|reserved_special_token_75|>",
672
+ "lstrip": false,
673
+ "normalized": false,
674
+ "rstrip": false,
675
+ "single_word": false,
676
+ "special": true
677
+ },
678
+ "128084": {
679
+ "content": "<|reserved_special_token_76|>",
680
+ "lstrip": false,
681
+ "normalized": false,
682
+ "rstrip": false,
683
+ "single_word": false,
684
+ "special": true
685
+ },
686
+ "128085": {
687
+ "content": "<|reserved_special_token_77|>",
688
+ "lstrip": false,
689
+ "normalized": false,
690
+ "rstrip": false,
691
+ "single_word": false,
692
+ "special": true
693
+ },
694
+ "128086": {
695
+ "content": "<|reserved_special_token_78|>",
696
+ "lstrip": false,
697
+ "normalized": false,
698
+ "rstrip": false,
699
+ "single_word": false,
700
+ "special": true
701
+ },
702
+ "128087": {
703
+ "content": "<|reserved_special_token_79|>",
704
+ "lstrip": false,
705
+ "normalized": false,
706
+ "rstrip": false,
707
+ "single_word": false,
708
+ "special": true
709
+ },
710
+ "128088": {
711
+ "content": "<|reserved_special_token_80|>",
712
+ "lstrip": false,
713
+ "normalized": false,
714
+ "rstrip": false,
715
+ "single_word": false,
716
+ "special": true
717
+ },
718
+ "128089": {
719
+ "content": "<|reserved_special_token_81|>",
720
+ "lstrip": false,
721
+ "normalized": false,
722
+ "rstrip": false,
723
+ "single_word": false,
724
+ "special": true
725
+ },
726
+ "128090": {
727
+ "content": "<|reserved_special_token_82|>",
728
+ "lstrip": false,
729
+ "normalized": false,
730
+ "rstrip": false,
731
+ "single_word": false,
732
+ "special": true
733
+ },
734
+ "128091": {
735
+ "content": "<|reserved_special_token_83|>",
736
+ "lstrip": false,
737
+ "normalized": false,
738
+ "rstrip": false,
739
+ "single_word": false,
740
+ "special": true
741
+ },
742
+ "128092": {
743
+ "content": "<|reserved_special_token_84|>",
744
+ "lstrip": false,
745
+ "normalized": false,
746
+ "rstrip": false,
747
+ "single_word": false,
748
+ "special": true
749
+ },
750
+ "128093": {
751
+ "content": "<|reserved_special_token_85|>",
752
+ "lstrip": false,
753
+ "normalized": false,
754
+ "rstrip": false,
755
+ "single_word": false,
756
+ "special": true
757
+ },
758
+ "128094": {
759
+ "content": "<|reserved_special_token_86|>",
760
+ "lstrip": false,
761
+ "normalized": false,
762
+ "rstrip": false,
763
+ "single_word": false,
764
+ "special": true
765
+ },
766
+ "128095": {
767
+ "content": "<|reserved_special_token_87|>",
768
+ "lstrip": false,
769
+ "normalized": false,
770
+ "rstrip": false,
771
+ "single_word": false,
772
+ "special": true
773
+ },
774
+ "128096": {
775
+ "content": "<|reserved_special_token_88|>",
776
+ "lstrip": false,
777
+ "normalized": false,
778
+ "rstrip": false,
779
+ "single_word": false,
780
+ "special": true
781
+ },
782
+ "128097": {
783
+ "content": "<|reserved_special_token_89|>",
784
+ "lstrip": false,
785
+ "normalized": false,
786
+ "rstrip": false,
787
+ "single_word": false,
788
+ "special": true
789
+ },
790
+ "128098": {
791
+ "content": "<|reserved_special_token_90|>",
792
+ "lstrip": false,
793
+ "normalized": false,
794
+ "rstrip": false,
795
+ "single_word": false,
796
+ "special": true
797
+ },
798
+ "128099": {
799
+ "content": "<|reserved_special_token_91|>",
800
+ "lstrip": false,
801
+ "normalized": false,
802
+ "rstrip": false,
803
+ "single_word": false,
804
+ "special": true
805
+ },
806
+ "128100": {
807
+ "content": "<|reserved_special_token_92|>",
808
+ "lstrip": false,
809
+ "normalized": false,
810
+ "rstrip": false,
811
+ "single_word": false,
812
+ "special": true
813
+ },
814
+ "128101": {
815
+ "content": "<|reserved_special_token_93|>",
816
+ "lstrip": false,
817
+ "normalized": false,
818
+ "rstrip": false,
819
+ "single_word": false,
820
+ "special": true
821
+ },
822
+ "128102": {
823
+ "content": "<|reserved_special_token_94|>",
824
+ "lstrip": false,
825
+ "normalized": false,
826
+ "rstrip": false,
827
+ "single_word": false,
828
+ "special": true
829
+ },
830
+ "128103": {
831
+ "content": "<|reserved_special_token_95|>",
832
+ "lstrip": false,
833
+ "normalized": false,
834
+ "rstrip": false,
835
+ "single_word": false,
836
+ "special": true
837
+ },
838
+ "128104": {
839
+ "content": "<|reserved_special_token_96|>",
840
+ "lstrip": false,
841
+ "normalized": false,
842
+ "rstrip": false,
843
+ "single_word": false,
844
+ "special": true
845
+ },
846
+ "128105": {
847
+ "content": "<|reserved_special_token_97|>",
848
+ "lstrip": false,
849
+ "normalized": false,
850
+ "rstrip": false,
851
+ "single_word": false,
852
+ "special": true
853
+ },
854
+ "128106": {
855
+ "content": "<|reserved_special_token_98|>",
856
+ "lstrip": false,
857
+ "normalized": false,
858
+ "rstrip": false,
859
+ "single_word": false,
860
+ "special": true
861
+ },
862
+ "128107": {
863
+ "content": "<|reserved_special_token_99|>",
864
+ "lstrip": false,
865
+ "normalized": false,
866
+ "rstrip": false,
867
+ "single_word": false,
868
+ "special": true
869
+ },
870
+ "128108": {
871
+ "content": "<|reserved_special_token_100|>",
872
+ "lstrip": false,
873
+ "normalized": false,
874
+ "rstrip": false,
875
+ "single_word": false,
876
+ "special": true
877
+ },
878
+ "128109": {
879
+ "content": "<|reserved_special_token_101|>",
880
+ "lstrip": false,
881
+ "normalized": false,
882
+ "rstrip": false,
883
+ "single_word": false,
884
+ "special": true
885
+ },
886
+ "128110": {
887
+ "content": "<|reserved_special_token_102|>",
888
+ "lstrip": false,
889
+ "normalized": false,
890
+ "rstrip": false,
891
+ "single_word": false,
892
+ "special": true
893
+ },
894
+ "128111": {
895
+ "content": "<|reserved_special_token_103|>",
896
+ "lstrip": false,
897
+ "normalized": false,
898
+ "rstrip": false,
899
+ "single_word": false,
900
+ "special": true
901
+ },
902
+ "128112": {
903
+ "content": "<|reserved_special_token_104|>",
904
+ "lstrip": false,
905
+ "normalized": false,
906
+ "rstrip": false,
907
+ "single_word": false,
908
+ "special": true
909
+ },
910
+ "128113": {
911
+ "content": "<|reserved_special_token_105|>",
912
+ "lstrip": false,
913
+ "normalized": false,
914
+ "rstrip": false,
915
+ "single_word": false,
916
+ "special": true
917
+ },
918
+ "128114": {
919
+ "content": "<|reserved_special_token_106|>",
920
+ "lstrip": false,
921
+ "normalized": false,
922
+ "rstrip": false,
923
+ "single_word": false,
924
+ "special": true
925
+ },
926
+ "128115": {
927
+ "content": "<|reserved_special_token_107|>",
928
+ "lstrip": false,
929
+ "normalized": false,
930
+ "rstrip": false,
931
+ "single_word": false,
932
+ "special": true
933
+ },
934
+ "128116": {
935
+ "content": "<|reserved_special_token_108|>",
936
+ "lstrip": false,
937
+ "normalized": false,
938
+ "rstrip": false,
939
+ "single_word": false,
940
+ "special": true
941
+ },
942
+ "128117": {
943
+ "content": "<|reserved_special_token_109|>",
944
+ "lstrip": false,
945
+ "normalized": false,
946
+ "rstrip": false,
947
+ "single_word": false,
948
+ "special": true
949
+ },
950
+ "128118": {
951
+ "content": "<|reserved_special_token_110|>",
952
+ "lstrip": false,
953
+ "normalized": false,
954
+ "rstrip": false,
955
+ "single_word": false,
956
+ "special": true
957
+ },
958
+ "128119": {
959
+ "content": "<|reserved_special_token_111|>",
960
+ "lstrip": false,
961
+ "normalized": false,
962
+ "rstrip": false,
963
+ "single_word": false,
964
+ "special": true
965
+ },
966
+ "128120": {
967
+ "content": "<|reserved_special_token_112|>",
968
+ "lstrip": false,
969
+ "normalized": false,
970
+ "rstrip": false,
971
+ "single_word": false,
972
+ "special": true
973
+ },
974
+ "128121": {
975
+ "content": "<|reserved_special_token_113|>",
976
+ "lstrip": false,
977
+ "normalized": false,
978
+ "rstrip": false,
979
+ "single_word": false,
980
+ "special": true
981
+ },
982
+ "128122": {
983
+ "content": "<|reserved_special_token_114|>",
984
+ "lstrip": false,
985
+ "normalized": false,
986
+ "rstrip": false,
987
+ "single_word": false,
988
+ "special": true
989
+ },
990
+ "128123": {
991
+ "content": "<|reserved_special_token_115|>",
992
+ "lstrip": false,
993
+ "normalized": false,
994
+ "rstrip": false,
995
+ "single_word": false,
996
+ "special": true
997
+ },
998
+ "128124": {
999
+ "content": "<|reserved_special_token_116|>",
1000
+ "lstrip": false,
1001
+ "normalized": false,
1002
+ "rstrip": false,
1003
+ "single_word": false,
1004
+ "special": true
1005
+ },
1006
+ "128125": {
1007
+ "content": "<|reserved_special_token_117|>",
1008
+ "lstrip": false,
1009
+ "normalized": false,
1010
+ "rstrip": false,
1011
+ "single_word": false,
1012
+ "special": true
1013
+ },
1014
+ "128126": {
1015
+ "content": "<|reserved_special_token_118|>",
1016
+ "lstrip": false,
1017
+ "normalized": false,
1018
+ "rstrip": false,
1019
+ "single_word": false,
1020
+ "special": true
1021
+ },
1022
+ "128127": {
1023
+ "content": "<|reserved_special_token_119|>",
1024
+ "lstrip": false,
1025
+ "normalized": false,
1026
+ "rstrip": false,
1027
+ "single_word": false,
1028
+ "special": true
1029
+ },
1030
+ "128128": {
1031
+ "content": "<|reserved_special_token_120|>",
1032
+ "lstrip": false,
1033
+ "normalized": false,
1034
+ "rstrip": false,
1035
+ "single_word": false,
1036
+ "special": true
1037
+ },
1038
+ "128129": {
1039
+ "content": "<|reserved_special_token_121|>",
1040
+ "lstrip": false,
1041
+ "normalized": false,
1042
+ "rstrip": false,
1043
+ "single_word": false,
1044
+ "special": true
1045
+ },
1046
+ "128130": {
1047
+ "content": "<|reserved_special_token_122|>",
1048
+ "lstrip": false,
1049
+ "normalized": false,
1050
+ "rstrip": false,
1051
+ "single_word": false,
1052
+ "special": true
1053
+ },
1054
+ "128131": {
1055
+ "content": "<|reserved_special_token_123|>",
1056
+ "lstrip": false,
1057
+ "normalized": false,
1058
+ "rstrip": false,
1059
+ "single_word": false,
1060
+ "special": true
1061
+ },
1062
+ "128132": {
1063
+ "content": "<|reserved_special_token_124|>",
1064
+ "lstrip": false,
1065
+ "normalized": false,
1066
+ "rstrip": false,
1067
+ "single_word": false,
1068
+ "special": true
1069
+ },
1070
+ "128133": {
1071
+ "content": "<|reserved_special_token_125|>",
1072
+ "lstrip": false,
1073
+ "normalized": false,
1074
+ "rstrip": false,
1075
+ "single_word": false,
1076
+ "special": true
1077
+ },
1078
+ "128134": {
1079
+ "content": "<|reserved_special_token_126|>",
1080
+ "lstrip": false,
1081
+ "normalized": false,
1082
+ "rstrip": false,
1083
+ "single_word": false,
1084
+ "special": true
1085
+ },
1086
+ "128135": {
1087
+ "content": "<|reserved_special_token_127|>",
1088
+ "lstrip": false,
1089
+ "normalized": false,
1090
+ "rstrip": false,
1091
+ "single_word": false,
1092
+ "special": true
1093
+ },
1094
+ "128136": {
1095
+ "content": "<|reserved_special_token_128|>",
1096
+ "lstrip": false,
1097
+ "normalized": false,
1098
+ "rstrip": false,
1099
+ "single_word": false,
1100
+ "special": true
1101
+ },
1102
+ "128137": {
1103
+ "content": "<|reserved_special_token_129|>",
1104
+ "lstrip": false,
1105
+ "normalized": false,
1106
+ "rstrip": false,
1107
+ "single_word": false,
1108
+ "special": true
1109
+ },
1110
+ "128138": {
1111
+ "content": "<|reserved_special_token_130|>",
1112
+ "lstrip": false,
1113
+ "normalized": false,
1114
+ "rstrip": false,
1115
+ "single_word": false,
1116
+ "special": true
1117
+ },
1118
+ "128139": {
1119
+ "content": "<|reserved_special_token_131|>",
1120
+ "lstrip": false,
1121
+ "normalized": false,
1122
+ "rstrip": false,
1123
+ "single_word": false,
1124
+ "special": true
1125
+ },
1126
+ "128140": {
1127
+ "content": "<|reserved_special_token_132|>",
1128
+ "lstrip": false,
1129
+ "normalized": false,
1130
+ "rstrip": false,
1131
+ "single_word": false,
1132
+ "special": true
1133
+ },
1134
+ "128141": {
1135
+ "content": "<|reserved_special_token_133|>",
1136
+ "lstrip": false,
1137
+ "normalized": false,
1138
+ "rstrip": false,
1139
+ "single_word": false,
1140
+ "special": true
1141
+ },
1142
+ "128142": {
1143
+ "content": "<|reserved_special_token_134|>",
1144
+ "lstrip": false,
1145
+ "normalized": false,
1146
+ "rstrip": false,
1147
+ "single_word": false,
1148
+ "special": true
1149
+ },
1150
+ "128143": {
1151
+ "content": "<|reserved_special_token_135|>",
1152
+ "lstrip": false,
1153
+ "normalized": false,
1154
+ "rstrip": false,
1155
+ "single_word": false,
1156
+ "special": true
1157
+ },
1158
+ "128144": {
1159
+ "content": "<|reserved_special_token_136|>",
1160
+ "lstrip": false,
1161
+ "normalized": false,
1162
+ "rstrip": false,
1163
+ "single_word": false,
1164
+ "special": true
1165
+ },
1166
+ "128145": {
1167
+ "content": "<|reserved_special_token_137|>",
1168
+ "lstrip": false,
1169
+ "normalized": false,
1170
+ "rstrip": false,
1171
+ "single_word": false,
1172
+ "special": true
1173
+ },
1174
+ "128146": {
1175
+ "content": "<|reserved_special_token_138|>",
1176
+ "lstrip": false,
1177
+ "normalized": false,
1178
+ "rstrip": false,
1179
+ "single_word": false,
1180
+ "special": true
1181
+ },
1182
+ "128147": {
1183
+ "content": "<|reserved_special_token_139|>",
1184
+ "lstrip": false,
1185
+ "normalized": false,
1186
+ "rstrip": false,
1187
+ "single_word": false,
1188
+ "special": true
1189
+ },
1190
+ "128148": {
1191
+ "content": "<|reserved_special_token_140|>",
1192
+ "lstrip": false,
1193
+ "normalized": false,
1194
+ "rstrip": false,
1195
+ "single_word": false,
1196
+ "special": true
1197
+ },
1198
+ "128149": {
1199
+ "content": "<|reserved_special_token_141|>",
1200
+ "lstrip": false,
1201
+ "normalized": false,
1202
+ "rstrip": false,
1203
+ "single_word": false,
1204
+ "special": true
1205
+ },
1206
+ "128150": {
1207
+ "content": "<|reserved_special_token_142|>",
1208
+ "lstrip": false,
1209
+ "normalized": false,
1210
+ "rstrip": false,
1211
+ "single_word": false,
1212
+ "special": true
1213
+ },
1214
+ "128151": {
1215
+ "content": "<|reserved_special_token_143|>",
1216
+ "lstrip": false,
1217
+ "normalized": false,
1218
+ "rstrip": false,
1219
+ "single_word": false,
1220
+ "special": true
1221
+ },
1222
+ "128152": {
1223
+ "content": "<|reserved_special_token_144|>",
1224
+ "lstrip": false,
1225
+ "normalized": false,
1226
+ "rstrip": false,
1227
+ "single_word": false,
1228
+ "special": true
1229
+ },
1230
+ "128153": {
1231
+ "content": "<|reserved_special_token_145|>",
1232
+ "lstrip": false,
1233
+ "normalized": false,
1234
+ "rstrip": false,
1235
+ "single_word": false,
1236
+ "special": true
1237
+ },
1238
+ "128154": {
1239
+ "content": "<|reserved_special_token_146|>",
1240
+ "lstrip": false,
1241
+ "normalized": false,
1242
+ "rstrip": false,
1243
+ "single_word": false,
1244
+ "special": true
1245
+ },
1246
+ "128155": {
1247
+ "content": "<|reserved_special_token_147|>",
1248
+ "lstrip": false,
1249
+ "normalized": false,
1250
+ "rstrip": false,
1251
+ "single_word": false,
1252
+ "special": true
1253
+ },
1254
+ "128156": {
1255
+ "content": "<|reserved_special_token_148|>",
1256
+ "lstrip": false,
1257
+ "normalized": false,
1258
+ "rstrip": false,
1259
+ "single_word": false,
1260
+ "special": true
1261
+ },
1262
+ "128157": {
1263
+ "content": "<|reserved_special_token_149|>",
1264
+ "lstrip": false,
1265
+ "normalized": false,
1266
+ "rstrip": false,
1267
+ "single_word": false,
1268
+ "special": true
1269
+ },
1270
+ "128158": {
1271
+ "content": "<|reserved_special_token_150|>",
1272
+ "lstrip": false,
1273
+ "normalized": false,
1274
+ "rstrip": false,
1275
+ "single_word": false,
1276
+ "special": true
1277
+ },
1278
+ "128159": {
1279
+ "content": "<|reserved_special_token_151|>",
1280
+ "lstrip": false,
1281
+ "normalized": false,
1282
+ "rstrip": false,
1283
+ "single_word": false,
1284
+ "special": true
1285
+ },
1286
+ "128160": {
1287
+ "content": "<|reserved_special_token_152|>",
1288
+ "lstrip": false,
1289
+ "normalized": false,
1290
+ "rstrip": false,
1291
+ "single_word": false,
1292
+ "special": true
1293
+ },
1294
+ "128161": {
1295
+ "content": "<|reserved_special_token_153|>",
1296
+ "lstrip": false,
1297
+ "normalized": false,
1298
+ "rstrip": false,
1299
+ "single_word": false,
1300
+ "special": true
1301
+ },
1302
+ "128162": {
1303
+ "content": "<|reserved_special_token_154|>",
1304
+ "lstrip": false,
1305
+ "normalized": false,
1306
+ "rstrip": false,
1307
+ "single_word": false,
1308
+ "special": true
1309
+ },
1310
+ "128163": {
1311
+ "content": "<|reserved_special_token_155|>",
1312
+ "lstrip": false,
1313
+ "normalized": false,
1314
+ "rstrip": false,
1315
+ "single_word": false,
1316
+ "special": true
1317
+ },
1318
+ "128164": {
1319
+ "content": "<|reserved_special_token_156|>",
1320
+ "lstrip": false,
1321
+ "normalized": false,
1322
+ "rstrip": false,
1323
+ "single_word": false,
1324
+ "special": true
1325
+ },
1326
+ "128165": {
1327
+ "content": "<|reserved_special_token_157|>",
1328
+ "lstrip": false,
1329
+ "normalized": false,
1330
+ "rstrip": false,
1331
+ "single_word": false,
1332
+ "special": true
1333
+ },
1334
+ "128166": {
1335
+ "content": "<|reserved_special_token_158|>",
1336
+ "lstrip": false,
1337
+ "normalized": false,
1338
+ "rstrip": false,
1339
+ "single_word": false,
1340
+ "special": true
1341
+ },
1342
+ "128167": {
1343
+ "content": "<|reserved_special_token_159|>",
1344
+ "lstrip": false,
1345
+ "normalized": false,
1346
+ "rstrip": false,
1347
+ "single_word": false,
1348
+ "special": true
1349
+ },
1350
+ "128168": {
1351
+ "content": "<|reserved_special_token_160|>",
1352
+ "lstrip": false,
1353
+ "normalized": false,
1354
+ "rstrip": false,
1355
+ "single_word": false,
1356
+ "special": true
1357
+ },
1358
+ "128169": {
1359
+ "content": "<|reserved_special_token_161|>",
1360
+ "lstrip": false,
1361
+ "normalized": false,
1362
+ "rstrip": false,
1363
+ "single_word": false,
1364
+ "special": true
1365
+ },
1366
+ "128170": {
1367
+ "content": "<|reserved_special_token_162|>",
1368
+ "lstrip": false,
1369
+ "normalized": false,
1370
+ "rstrip": false,
1371
+ "single_word": false,
1372
+ "special": true
1373
+ },
1374
+ "128171": {
1375
+ "content": "<|reserved_special_token_163|>",
1376
+ "lstrip": false,
1377
+ "normalized": false,
1378
+ "rstrip": false,
1379
+ "single_word": false,
1380
+ "special": true
1381
+ },
1382
+ "128172": {
1383
+ "content": "<|reserved_special_token_164|>",
1384
+ "lstrip": false,
1385
+ "normalized": false,
1386
+ "rstrip": false,
1387
+ "single_word": false,
1388
+ "special": true
1389
+ },
1390
+ "128173": {
1391
+ "content": "<|reserved_special_token_165|>",
1392
+ "lstrip": false,
1393
+ "normalized": false,
1394
+ "rstrip": false,
1395
+ "single_word": false,
1396
+ "special": true
1397
+ },
1398
+ "128174": {
1399
+ "content": "<|reserved_special_token_166|>",
1400
+ "lstrip": false,
1401
+ "normalized": false,
1402
+ "rstrip": false,
1403
+ "single_word": false,
1404
+ "special": true
1405
+ },
1406
+ "128175": {
1407
+ "content": "<|reserved_special_token_167|>",
1408
+ "lstrip": false,
1409
+ "normalized": false,
1410
+ "rstrip": false,
1411
+ "single_word": false,
1412
+ "special": true
1413
+ },
1414
+ "128176": {
1415
+ "content": "<|reserved_special_token_168|>",
1416
+ "lstrip": false,
1417
+ "normalized": false,
1418
+ "rstrip": false,
1419
+ "single_word": false,
1420
+ "special": true
1421
+ },
1422
+ "128177": {
1423
+ "content": "<|reserved_special_token_169|>",
1424
+ "lstrip": false,
1425
+ "normalized": false,
1426
+ "rstrip": false,
1427
+ "single_word": false,
1428
+ "special": true
1429
+ },
1430
+ "128178": {
1431
+ "content": "<|reserved_special_token_170|>",
1432
+ "lstrip": false,
1433
+ "normalized": false,
1434
+ "rstrip": false,
1435
+ "single_word": false,
1436
+ "special": true
1437
+ },
1438
+ "128179": {
1439
+ "content": "<|reserved_special_token_171|>",
1440
+ "lstrip": false,
1441
+ "normalized": false,
1442
+ "rstrip": false,
1443
+ "single_word": false,
1444
+ "special": true
1445
+ },
1446
+ "128180": {
1447
+ "content": "<|reserved_special_token_172|>",
1448
+ "lstrip": false,
1449
+ "normalized": false,
1450
+ "rstrip": false,
1451
+ "single_word": false,
1452
+ "special": true
1453
+ },
1454
+ "128181": {
1455
+ "content": "<|reserved_special_token_173|>",
1456
+ "lstrip": false,
1457
+ "normalized": false,
1458
+ "rstrip": false,
1459
+ "single_word": false,
1460
+ "special": true
1461
+ },
1462
+ "128182": {
1463
+ "content": "<|reserved_special_token_174|>",
1464
+ "lstrip": false,
1465
+ "normalized": false,
1466
+ "rstrip": false,
1467
+ "single_word": false,
1468
+ "special": true
1469
+ },
1470
+ "128183": {
1471
+ "content": "<|reserved_special_token_175|>",
1472
+ "lstrip": false,
1473
+ "normalized": false,
1474
+ "rstrip": false,
1475
+ "single_word": false,
1476
+ "special": true
1477
+ },
1478
+ "128184": {
1479
+ "content": "<|reserved_special_token_176|>",
1480
+ "lstrip": false,
1481
+ "normalized": false,
1482
+ "rstrip": false,
1483
+ "single_word": false,
1484
+ "special": true
1485
+ },
1486
+ "128185": {
1487
+ "content": "<|reserved_special_token_177|>",
1488
+ "lstrip": false,
1489
+ "normalized": false,
1490
+ "rstrip": false,
1491
+ "single_word": false,
1492
+ "special": true
1493
+ },
1494
+ "128186": {
1495
+ "content": "<|reserved_special_token_178|>",
1496
+ "lstrip": false,
1497
+ "normalized": false,
1498
+ "rstrip": false,
1499
+ "single_word": false,
1500
+ "special": true
1501
+ },
1502
+ "128187": {
1503
+ "content": "<|reserved_special_token_179|>",
1504
+ "lstrip": false,
1505
+ "normalized": false,
1506
+ "rstrip": false,
1507
+ "single_word": false,
1508
+ "special": true
1509
+ },
1510
+ "128188": {
1511
+ "content": "<|reserved_special_token_180|>",
1512
+ "lstrip": false,
1513
+ "normalized": false,
1514
+ "rstrip": false,
1515
+ "single_word": false,
1516
+ "special": true
1517
+ },
1518
+ "128189": {
1519
+ "content": "<|reserved_special_token_181|>",
1520
+ "lstrip": false,
1521
+ "normalized": false,
1522
+ "rstrip": false,
1523
+ "single_word": false,
1524
+ "special": true
1525
+ },
1526
+ "128190": {
1527
+ "content": "<|reserved_special_token_182|>",
1528
+ "lstrip": false,
1529
+ "normalized": false,
1530
+ "rstrip": false,
1531
+ "single_word": false,
1532
+ "special": true
1533
+ },
1534
+ "128191": {
1535
+ "content": "<|reserved_special_token_183|>",
1536
+ "lstrip": false,
1537
+ "normalized": false,
1538
+ "rstrip": false,
1539
+ "single_word": false,
1540
+ "special": true
1541
+ },
1542
+ "128192": {
1543
+ "content": "<|reserved_special_token_184|>",
1544
+ "lstrip": false,
1545
+ "normalized": false,
1546
+ "rstrip": false,
1547
+ "single_word": false,
1548
+ "special": true
1549
+ },
1550
+ "128193": {
1551
+ "content": "<|reserved_special_token_185|>",
1552
+ "lstrip": false,
1553
+ "normalized": false,
1554
+ "rstrip": false,
1555
+ "single_word": false,
1556
+ "special": true
1557
+ },
1558
+ "128194": {
1559
+ "content": "<|reserved_special_token_186|>",
1560
+ "lstrip": false,
1561
+ "normalized": false,
1562
+ "rstrip": false,
1563
+ "single_word": false,
1564
+ "special": true
1565
+ },
1566
+ "128195": {
1567
+ "content": "<|reserved_special_token_187|>",
1568
+ "lstrip": false,
1569
+ "normalized": false,
1570
+ "rstrip": false,
1571
+ "single_word": false,
1572
+ "special": true
1573
+ },
1574
+ "128196": {
1575
+ "content": "<|reserved_special_token_188|>",
1576
+ "lstrip": false,
1577
+ "normalized": false,
1578
+ "rstrip": false,
1579
+ "single_word": false,
1580
+ "special": true
1581
+ },
1582
+ "128197": {
1583
+ "content": "<|reserved_special_token_189|>",
1584
+ "lstrip": false,
1585
+ "normalized": false,
1586
+ "rstrip": false,
1587
+ "single_word": false,
1588
+ "special": true
1589
+ },
1590
+ "128198": {
1591
+ "content": "<|reserved_special_token_190|>",
1592
+ "lstrip": false,
1593
+ "normalized": false,
1594
+ "rstrip": false,
1595
+ "single_word": false,
1596
+ "special": true
1597
+ },
1598
+ "128199": {
1599
+ "content": "<|reserved_special_token_191|>",
1600
+ "lstrip": false,
1601
+ "normalized": false,
1602
+ "rstrip": false,
1603
+ "single_word": false,
1604
+ "special": true
1605
+ },
1606
+ "128200": {
1607
+ "content": "<|reserved_special_token_192|>",
1608
+ "lstrip": false,
1609
+ "normalized": false,
1610
+ "rstrip": false,
1611
+ "single_word": false,
1612
+ "special": true
1613
+ },
1614
+ "128201": {
1615
+ "content": "<|reserved_special_token_193|>",
1616
+ "lstrip": false,
1617
+ "normalized": false,
1618
+ "rstrip": false,
1619
+ "single_word": false,
1620
+ "special": true
1621
+ },
1622
+ "128202": {
1623
+ "content": "<|reserved_special_token_194|>",
1624
+ "lstrip": false,
1625
+ "normalized": false,
1626
+ "rstrip": false,
1627
+ "single_word": false,
1628
+ "special": true
1629
+ },
1630
+ "128203": {
1631
+ "content": "<|reserved_special_token_195|>",
1632
+ "lstrip": false,
1633
+ "normalized": false,
1634
+ "rstrip": false,
1635
+ "single_word": false,
1636
+ "special": true
1637
+ },
1638
+ "128204": {
1639
+ "content": "<|reserved_special_token_196|>",
1640
+ "lstrip": false,
1641
+ "normalized": false,
1642
+ "rstrip": false,
1643
+ "single_word": false,
1644
+ "special": true
1645
+ },
1646
+ "128205": {
1647
+ "content": "<|reserved_special_token_197|>",
1648
+ "lstrip": false,
1649
+ "normalized": false,
1650
+ "rstrip": false,
1651
+ "single_word": false,
1652
+ "special": true
1653
+ },
1654
+ "128206": {
1655
+ "content": "<|reserved_special_token_198|>",
1656
+ "lstrip": false,
1657
+ "normalized": false,
1658
+ "rstrip": false,
1659
+ "single_word": false,
1660
+ "special": true
1661
+ },
1662
+ "128207": {
1663
+ "content": "<|reserved_special_token_199|>",
1664
+ "lstrip": false,
1665
+ "normalized": false,
1666
+ "rstrip": false,
1667
+ "single_word": false,
1668
+ "special": true
1669
+ },
1670
+ "128208": {
1671
+ "content": "<|reserved_special_token_200|>",
1672
+ "lstrip": false,
1673
+ "normalized": false,
1674
+ "rstrip": false,
1675
+ "single_word": false,
1676
+ "special": true
1677
+ },
1678
+ "128209": {
1679
+ "content": "<|reserved_special_token_201|>",
1680
+ "lstrip": false,
1681
+ "normalized": false,
1682
+ "rstrip": false,
1683
+ "single_word": false,
1684
+ "special": true
1685
+ },
1686
+ "128210": {
1687
+ "content": "<|reserved_special_token_202|>",
1688
+ "lstrip": false,
1689
+ "normalized": false,
1690
+ "rstrip": false,
1691
+ "single_word": false,
1692
+ "special": true
1693
+ },
1694
+ "128211": {
1695
+ "content": "<|reserved_special_token_203|>",
1696
+ "lstrip": false,
1697
+ "normalized": false,
1698
+ "rstrip": false,
1699
+ "single_word": false,
1700
+ "special": true
1701
+ },
1702
+ "128212": {
1703
+ "content": "<|reserved_special_token_204|>",
1704
+ "lstrip": false,
1705
+ "normalized": false,
1706
+ "rstrip": false,
1707
+ "single_word": false,
1708
+ "special": true
1709
+ },
1710
+ "128213": {
1711
+ "content": "<|reserved_special_token_205|>",
1712
+ "lstrip": false,
1713
+ "normalized": false,
1714
+ "rstrip": false,
1715
+ "single_word": false,
1716
+ "special": true
1717
+ },
1718
+ "128214": {
1719
+ "content": "<|reserved_special_token_206|>",
1720
+ "lstrip": false,
1721
+ "normalized": false,
1722
+ "rstrip": false,
1723
+ "single_word": false,
1724
+ "special": true
1725
+ },
1726
+ "128215": {
1727
+ "content": "<|reserved_special_token_207|>",
1728
+ "lstrip": false,
1729
+ "normalized": false,
1730
+ "rstrip": false,
1731
+ "single_word": false,
1732
+ "special": true
1733
+ },
1734
+ "128216": {
1735
+ "content": "<|reserved_special_token_208|>",
1736
+ "lstrip": false,
1737
+ "normalized": false,
1738
+ "rstrip": false,
1739
+ "single_word": false,
1740
+ "special": true
1741
+ },
1742
+ "128217": {
1743
+ "content": "<|reserved_special_token_209|>",
1744
+ "lstrip": false,
1745
+ "normalized": false,
1746
+ "rstrip": false,
1747
+ "single_word": false,
1748
+ "special": true
1749
+ },
1750
+ "128218": {
1751
+ "content": "<|reserved_special_token_210|>",
1752
+ "lstrip": false,
1753
+ "normalized": false,
1754
+ "rstrip": false,
1755
+ "single_word": false,
1756
+ "special": true
1757
+ },
1758
+ "128219": {
1759
+ "content": "<|reserved_special_token_211|>",
1760
+ "lstrip": false,
1761
+ "normalized": false,
1762
+ "rstrip": false,
1763
+ "single_word": false,
1764
+ "special": true
1765
+ },
1766
+ "128220": {
1767
+ "content": "<|reserved_special_token_212|>",
1768
+ "lstrip": false,
1769
+ "normalized": false,
1770
+ "rstrip": false,
1771
+ "single_word": false,
1772
+ "special": true
1773
+ },
1774
+ "128221": {
1775
+ "content": "<|reserved_special_token_213|>",
1776
+ "lstrip": false,
1777
+ "normalized": false,
1778
+ "rstrip": false,
1779
+ "single_word": false,
1780
+ "special": true
1781
+ },
1782
+ "128222": {
1783
+ "content": "<|reserved_special_token_214|>",
1784
+ "lstrip": false,
1785
+ "normalized": false,
1786
+ "rstrip": false,
1787
+ "single_word": false,
1788
+ "special": true
1789
+ },
1790
+ "128223": {
1791
+ "content": "<|reserved_special_token_215|>",
1792
+ "lstrip": false,
1793
+ "normalized": false,
1794
+ "rstrip": false,
1795
+ "single_word": false,
1796
+ "special": true
1797
+ },
1798
+ "128224": {
1799
+ "content": "<|reserved_special_token_216|>",
1800
+ "lstrip": false,
1801
+ "normalized": false,
1802
+ "rstrip": false,
1803
+ "single_word": false,
1804
+ "special": true
1805
+ },
1806
+ "128225": {
1807
+ "content": "<|reserved_special_token_217|>",
1808
+ "lstrip": false,
1809
+ "normalized": false,
1810
+ "rstrip": false,
1811
+ "single_word": false,
1812
+ "special": true
1813
+ },
1814
+ "128226": {
1815
+ "content": "<|reserved_special_token_218|>",
1816
+ "lstrip": false,
1817
+ "normalized": false,
1818
+ "rstrip": false,
1819
+ "single_word": false,
1820
+ "special": true
1821
+ },
1822
+ "128227": {
1823
+ "content": "<|reserved_special_token_219|>",
1824
+ "lstrip": false,
1825
+ "normalized": false,
1826
+ "rstrip": false,
1827
+ "single_word": false,
1828
+ "special": true
1829
+ },
1830
+ "128228": {
1831
+ "content": "<|reserved_special_token_220|>",
1832
+ "lstrip": false,
1833
+ "normalized": false,
1834
+ "rstrip": false,
1835
+ "single_word": false,
1836
+ "special": true
1837
+ },
1838
+ "128229": {
1839
+ "content": "<|reserved_special_token_221|>",
1840
+ "lstrip": false,
1841
+ "normalized": false,
1842
+ "rstrip": false,
1843
+ "single_word": false,
1844
+ "special": true
1845
+ },
1846
+ "128230": {
1847
+ "content": "<|reserved_special_token_222|>",
1848
+ "lstrip": false,
1849
+ "normalized": false,
1850
+ "rstrip": false,
1851
+ "single_word": false,
1852
+ "special": true
1853
+ },
1854
+ "128231": {
1855
+ "content": "<|reserved_special_token_223|>",
1856
+ "lstrip": false,
1857
+ "normalized": false,
1858
+ "rstrip": false,
1859
+ "single_word": false,
1860
+ "special": true
1861
+ },
1862
+ "128232": {
1863
+ "content": "<|reserved_special_token_224|>",
1864
+ "lstrip": false,
1865
+ "normalized": false,
1866
+ "rstrip": false,
1867
+ "single_word": false,
1868
+ "special": true
1869
+ },
1870
+ "128233": {
1871
+ "content": "<|reserved_special_token_225|>",
1872
+ "lstrip": false,
1873
+ "normalized": false,
1874
+ "rstrip": false,
1875
+ "single_word": false,
1876
+ "special": true
1877
+ },
1878
+ "128234": {
1879
+ "content": "<|reserved_special_token_226|>",
1880
+ "lstrip": false,
1881
+ "normalized": false,
1882
+ "rstrip": false,
1883
+ "single_word": false,
1884
+ "special": true
1885
+ },
1886
+ "128235": {
1887
+ "content": "<|reserved_special_token_227|>",
1888
+ "lstrip": false,
1889
+ "normalized": false,
1890
+ "rstrip": false,
1891
+ "single_word": false,
1892
+ "special": true
1893
+ },
1894
+ "128236": {
1895
+ "content": "<|reserved_special_token_228|>",
1896
+ "lstrip": false,
1897
+ "normalized": false,
1898
+ "rstrip": false,
1899
+ "single_word": false,
1900
+ "special": true
1901
+ },
1902
+ "128237": {
1903
+ "content": "<|reserved_special_token_229|>",
1904
+ "lstrip": false,
1905
+ "normalized": false,
1906
+ "rstrip": false,
1907
+ "single_word": false,
1908
+ "special": true
1909
+ },
1910
+ "128238": {
1911
+ "content": "<|reserved_special_token_230|>",
1912
+ "lstrip": false,
1913
+ "normalized": false,
1914
+ "rstrip": false,
1915
+ "single_word": false,
1916
+ "special": true
1917
+ },
1918
+ "128239": {
1919
+ "content": "<|reserved_special_token_231|>",
1920
+ "lstrip": false,
1921
+ "normalized": false,
1922
+ "rstrip": false,
1923
+ "single_word": false,
1924
+ "special": true
1925
+ },
1926
+ "128240": {
1927
+ "content": "<|reserved_special_token_232|>",
1928
+ "lstrip": false,
1929
+ "normalized": false,
1930
+ "rstrip": false,
1931
+ "single_word": false,
1932
+ "special": true
1933
+ },
1934
+ "128241": {
1935
+ "content": "<|reserved_special_token_233|>",
1936
+ "lstrip": false,
1937
+ "normalized": false,
1938
+ "rstrip": false,
1939
+ "single_word": false,
1940
+ "special": true
1941
+ },
1942
+ "128242": {
1943
+ "content": "<|reserved_special_token_234|>",
1944
+ "lstrip": false,
1945
+ "normalized": false,
1946
+ "rstrip": false,
1947
+ "single_word": false,
1948
+ "special": true
1949
+ },
1950
+ "128243": {
1951
+ "content": "<|reserved_special_token_235|>",
1952
+ "lstrip": false,
1953
+ "normalized": false,
1954
+ "rstrip": false,
1955
+ "single_word": false,
1956
+ "special": true
1957
+ },
1958
+ "128244": {
1959
+ "content": "<|reserved_special_token_236|>",
1960
+ "lstrip": false,
1961
+ "normalized": false,
1962
+ "rstrip": false,
1963
+ "single_word": false,
1964
+ "special": true
1965
+ },
1966
+ "128245": {
1967
+ "content": "<|reserved_special_token_237|>",
1968
+ "lstrip": false,
1969
+ "normalized": false,
1970
+ "rstrip": false,
1971
+ "single_word": false,
1972
+ "special": true
1973
+ },
1974
+ "128246": {
1975
+ "content": "<|reserved_special_token_238|>",
1976
+ "lstrip": false,
1977
+ "normalized": false,
1978
+ "rstrip": false,
1979
+ "single_word": false,
1980
+ "special": true
1981
+ },
1982
+ "128247": {
1983
+ "content": "<|reserved_special_token_239|>",
1984
+ "lstrip": false,
1985
+ "normalized": false,
1986
+ "rstrip": false,
1987
+ "single_word": false,
1988
+ "special": true
1989
+ },
1990
+ "128248": {
1991
+ "content": "<|reserved_special_token_240|>",
1992
+ "lstrip": false,
1993
+ "normalized": false,
1994
+ "rstrip": false,
1995
+ "single_word": false,
1996
+ "special": true
1997
+ },
1998
+ "128249": {
1999
+ "content": "<|reserved_special_token_241|>",
2000
+ "lstrip": false,
2001
+ "normalized": false,
2002
+ "rstrip": false,
2003
+ "single_word": false,
2004
+ "special": true
2005
+ },
2006
+ "128250": {
2007
+ "content": "<|reserved_special_token_242|>",
2008
+ "lstrip": false,
2009
+ "normalized": false,
2010
+ "rstrip": false,
2011
+ "single_word": false,
2012
+ "special": true
2013
+ },
2014
+ "128251": {
2015
+ "content": "<|reserved_special_token_243|>",
2016
+ "lstrip": false,
2017
+ "normalized": false,
2018
+ "rstrip": false,
2019
+ "single_word": false,
2020
+ "special": true
2021
+ },
2022
+ "128252": {
2023
+ "content": "<|reserved_special_token_244|>",
2024
+ "lstrip": false,
2025
+ "normalized": false,
2026
+ "rstrip": false,
2027
+ "single_word": false,
2028
+ "special": true
2029
+ },
2030
+ "128253": {
2031
+ "content": "<|reserved_special_token_245|>",
2032
+ "lstrip": false,
2033
+ "normalized": false,
2034
+ "rstrip": false,
2035
+ "single_word": false,
2036
+ "special": true
2037
+ },
2038
+ "128254": {
2039
+ "content": "<|reserved_special_token_246|>",
2040
+ "lstrip": false,
2041
+ "normalized": false,
2042
+ "rstrip": false,
2043
+ "single_word": false,
2044
+ "special": true
2045
+ },
2046
+ "128255": {
2047
+ "content": "<|reserved_special_token_247|>",
2048
+ "lstrip": false,
2049
+ "normalized": false,
2050
+ "rstrip": false,
2051
+ "single_word": false,
2052
+ "special": true
2053
+ }
2054
+ },
2055
+ "bos_token": "<｜begin▁of▁sentence｜>",
2056
+ "chat_template": "{% if not add_generation_prompt is defined %}{% set add_generation_prompt = false %}{% endif %}{% set ns = namespace(is_first=false, is_tool=false, is_output_first=true, system_prompt='') %}{%- for message in messages %}{%- if message['role'] == 'system' %}{% set ns.system_prompt = message['content'] %}{%- endif %}{%- endfor %}{{bos_token}}{{ns.system_prompt}}{%- for message in messages %}{%- if message['role'] == 'user' %}{%- set ns.is_tool = false -%}{{'<｜User｜>' + message['content']}}{%- endif %}{%- if message['role'] == 'assistant' and message['content'] is none %}{%- set ns.is_tool = false -%}{%- for tool in message['tool_calls']%}{%- if not ns.is_first %}{{'<｜Assistant｜><｜tool▁calls▁begin｜><｜tool▁call▁begin｜>' + tool['type'] + '<｜tool▁sep｜>' + tool['function']['name'] + '\\n' + '```json' + '\\n' + tool['function']['arguments'] + '\\n' + '```' + '<｜tool▁call▁end｜>'}}{%- set ns.is_first = true -%}{%- else %}{{'\\n' + '<｜tool▁call▁begin｜>' + tool['type'] + '<｜tool▁sep｜>' + tool['function']['name'] + '\\n' + '```json' + '\\n' + tool['function']['arguments'] + '\\n' + '```' + '<｜tool▁call▁end｜>'}}{{'<｜tool▁calls▁end｜><｜end▁of▁sentence｜>'}}{%- endif %}{%- endfor %}{%- endif %}{%- if message['role'] == 'assistant' and message['content'] is not none %}{%- if ns.is_tool %}{{'<｜tool▁outputs▁end｜>' + message['content'] + '<｜end▁of▁sentence｜>'}}{%- set ns.is_tool = false -%}{%- else %}{% set content = message['content'] %}{% if '</think>' in content %}{% set content = content.split('</think>')[-1] %}{% endif %}{{'<｜Assistant｜>' + content + '<｜end▁of▁sentence｜>'}}{%- endif %}{%- endif %}{%- if message['role'] == 'tool' %}{%- set ns.is_tool = true -%}{%- if ns.is_output_first %}{{'<｜tool▁outputs▁begin｜><｜tool▁output▁begin｜>' + message['content'] + '<｜tool▁output▁end｜>'}}{%- set ns.is_output_first = false %}{%- else %}{{'\\n<｜tool▁output▁begin｜>' + message['content'] + '<｜tool▁output▁end｜>'}}{%- endif %}{%- endif %}{%- endfor -%}{% if ns.is_tool %}{{'<｜tool▁outputs▁end｜>'}}{% endif %}{% if add_generation_prompt and not ns.is_tool %}{{'<｜Assistant｜><think>\\n'}}{% endif %}",
2057
+ "clean_up_tokenization_spaces": false,
2058
+ "eos_token": "<｜end▁of▁sentence｜>",
2059
+ "extra_special_tokens": {},
2060
+ "legacy": true,
2061
+ "model_max_length": 16384,
2062
+ "pad_token": "<｜end▁of▁sentence｜>",
2063
+ "sp_model_kwargs": {},
2064
+ "tokenizer_class": "LlamaTokenizerFast",
2065
+ "unk_token": null,
2066
+ "use_default_system_prompt": false
2067
+ }
checkpoint-112/trainer_state.json ADDED
@@ -0,0 +1,817 @@
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
1
+ {
2
+ "best_metric": null,
3
+ "best_model_checkpoint": null,
4
+ "epoch": 3.8720930232558137,
5
+ "eval_steps": 500,
6
+ "global_step": 112,
7
+ "is_hyper_param_search": false,
8
+ "is_local_process_zero": true,
9
+ "is_world_process_zero": true,
10
+ "log_history": [
11
+ {
12
+ "epoch": 0.03488372093023256,
13
+ "grad_norm": 43.01280212402344,
14
+ "learning_rate": 5.0000000000000004e-08,
15
+ "loss": 2.7496,
16
+ "step": 1
17
+ },
18
+ {
19
+ "epoch": 0.06976744186046512,
20
+ "grad_norm": 44.429630279541016,
21
+ "learning_rate": 1.0000000000000001e-07,
22
+ "loss": 2.7809,
23
+ "step": 2
24
+ },
25
+ {
26
+ "epoch": 0.10465116279069768,
27
+ "grad_norm": 42.754234313964844,
28
+ "learning_rate": 1.5000000000000002e-07,
29
+ "loss": 2.6968,
30
+ "step": 3
31
+ },
32
+ {
33
+ "epoch": 0.13953488372093023,
34
+ "grad_norm": 45.27891159057617,
35
+ "learning_rate": 2.0000000000000002e-07,
36
+ "loss": 2.8076,
37
+ "step": 4
38
+ },
39
+ {
40
+ "epoch": 0.1744186046511628,
41
+ "grad_norm": 43.920860290527344,
42
+ "learning_rate": 2.5000000000000004e-07,
43
+ "loss": 2.7255,
44
+ "step": 5
45
+ },
46
+ {
47
+ "epoch": 0.20930232558139536,
48
+ "grad_norm": 44.61790466308594,
49
+ "learning_rate": 3.0000000000000004e-07,
50
+ "loss": 2.7178,
51
+ "step": 6
52
+ },
53
+ {
54
+ "epoch": 0.2441860465116279,
55
+ "grad_norm": 44.28205871582031,
56
+ "learning_rate": 3.5000000000000004e-07,
57
+ "loss": 2.7193,
58
+ "step": 7
59
+ },
60
+ {
61
+ "epoch": 0.27906976744186046,
62
+ "grad_norm": 45.1516227722168,
63
+ "learning_rate": 4.0000000000000003e-07,
64
+ "loss": 2.6846,
65
+ "step": 8
66
+ },
67
+ {
68
+ "epoch": 0.313953488372093,
69
+ "grad_norm": 46.8723258972168,
70
+ "learning_rate": 4.5000000000000003e-07,
71
+ "loss": 2.67,
72
+ "step": 9
73
+ },
74
+ {
75
+ "epoch": 0.3488372093023256,
76
+ "grad_norm": 48.33848571777344,
77
+ "learning_rate": 5.000000000000001e-07,
78
+ "loss": 2.6368,
79
+ "step": 10
80
+ },
81
+ {
82
+ "epoch": 0.38372093023255816,
83
+ "grad_norm": 46.772193908691406,
84
+ "learning_rate": 5.5e-07,
85
+ "loss": 2.5562,
86
+ "step": 11
87
+ },
88
+ {
89
+ "epoch": 0.4186046511627907,
90
+ "grad_norm": 48.12417984008789,
91
+ "learning_rate": 6.000000000000001e-07,
92
+ "loss": 2.5362,
93
+ "step": 12
94
+ },
95
+ {
96
+ "epoch": 0.45348837209302323,
97
+ "grad_norm": 49.0787239074707,
98
+ "learning_rate": 6.5e-07,
99
+ "loss": 2.3981,
100
+ "step": 13
101
+ },
102
+ {
103
+ "epoch": 0.4883720930232558,
104
+ "grad_norm": 52.367183685302734,
105
+ "learning_rate": 7.000000000000001e-07,
106
+ "loss": 2.3837,
107
+ "step": 14
108
+ },
109
+ {
110
+ "epoch": 0.5232558139534884,
111
+ "grad_norm": 53.57649230957031,
112
+ "learning_rate": 7.5e-07,
113
+ "loss": 2.2885,
114
+ "step": 15
115
+ },
116
+ {
117
+ "epoch": 0.5581395348837209,
118
+ "grad_norm": 56.58591079711914,
119
+ "learning_rate": 8.000000000000001e-07,
120
+ "loss": 2.1447,
121
+ "step": 16
122
+ },
123
+ {
124
+ "epoch": 0.5930232558139535,
125
+ "grad_norm": 55.064735412597656,
126
+ "learning_rate": 8.500000000000001e-07,
127
+ "loss": 2.0242,
128
+ "step": 17
129
+ },
130
+ {
131
+ "epoch": 0.627906976744186,
132
+ "grad_norm": 52.61149597167969,
133
+ "learning_rate": 9.000000000000001e-07,
134
+ "loss": 1.9263,
135
+ "step": 18
136
+ },
137
+ {
138
+ "epoch": 0.6627906976744186,
139
+ "grad_norm": 44.34925079345703,
140
+ "learning_rate": 9.500000000000001e-07,
141
+ "loss": 1.7466,
142
+ "step": 19
143
+ },
144
+ {
145
+ "epoch": 0.6976744186046512,
146
+ "grad_norm": 36.627296447753906,
147
+ "learning_rate": 1.0000000000000002e-06,
148
+ "loss": 1.6084,
149
+ "step": 20
150
+ },
151
+ {
152
+ "epoch": 0.7325581395348837,
153
+ "grad_norm": 30.89563751220703,
154
+ "learning_rate": 1.0500000000000001e-06,
155
+ "loss": 1.4701,
156
+ "step": 21
157
+ },
158
+ {
159
+ "epoch": 0.7674418604651163,
160
+ "grad_norm": 29.17167091369629,
161
+ "learning_rate": 1.1e-06,
162
+ "loss": 1.4218,
163
+ "step": 22
164
+ },
165
+ {
166
+ "epoch": 0.8023255813953488,
167
+ "grad_norm": 28.237022399902344,
168
+ "learning_rate": 1.1500000000000002e-06,
169
+ "loss": 1.3014,
170
+ "step": 23
171
+ },
172
+ {
173
+ "epoch": 0.8372093023255814,
174
+ "grad_norm": 28.778654098510742,
175
+ "learning_rate": 1.2000000000000002e-06,
176
+ "loss": 1.1857,
177
+ "step": 24
178
+ },
179
+ {
180
+ "epoch": 0.872093023255814,
181
+ "grad_norm": 29.368289947509766,
182
+ "learning_rate": 1.25e-06,
183
+ "loss": 1.082,
184
+ "step": 25
185
+ },
186
+ {
187
+ "epoch": 0.9069767441860465,
188
+ "grad_norm": 28.46448516845703,
189
+ "learning_rate": 1.3e-06,
190
+ "loss": 0.8716,
191
+ "step": 26
192
+ },
193
+ {
194
+ "epoch": 0.9418604651162791,
195
+ "grad_norm": 24.584318161010742,
196
+ "learning_rate": 1.3500000000000002e-06,
197
+ "loss": 0.6848,
198
+ "step": 27
199
+ },
200
+ {
201
+ "epoch": 0.9767441860465116,
202
+ "grad_norm": 21.166847229003906,
203
+ "learning_rate": 1.4000000000000001e-06,
204
+ "loss": 0.5438,
205
+ "step": 28
206
+ },
207
+ {
208
+ "epoch": 1.0,
209
+ "grad_norm": 21.166847229003906,
210
+ "learning_rate": 1.45e-06,
211
+ "loss": 0.4084,
212
+ "step": 29
213
+ },
214
+ {
215
+ "epoch": 1.0348837209302326,
216
+ "grad_norm": 24.098440170288086,
217
+ "learning_rate": 1.5e-06,
218
+ "loss": 0.3545,
219
+ "step": 30
220
+ },
221
+ {
222
+ "epoch": 1.069767441860465,
223
+ "grad_norm": 12.6813325881958,
224
+ "learning_rate": 1.5500000000000002e-06,
225
+ "loss": 0.2673,
226
+ "step": 31
227
+ },
228
+ {
229
+ "epoch": 1.1046511627906976,
230
+ "grad_norm": 7.6524128913879395,
231
+ "learning_rate": 1.6000000000000001e-06,
232
+ "loss": 0.2033,
233
+ "step": 32
234
+ },
235
+ {
236
+ "epoch": 1.1395348837209303,
237
+ "grad_norm": 5.23082160949707,
238
+ "learning_rate": 1.6500000000000003e-06,
239
+ "loss": 0.1887,
240
+ "step": 33
241
+ },
242
+ {
243
+ "epoch": 1.1744186046511629,
244
+ "grad_norm": 4.729929447174072,
245
+ "learning_rate": 1.7000000000000002e-06,
246
+ "loss": 0.1683,
247
+ "step": 34
248
+ },
249
+ {
250
+ "epoch": 1.2093023255813953,
251
+ "grad_norm": 5.831579208374023,
252
+ "learning_rate": 1.75e-06,
253
+ "loss": 0.1643,
254
+ "step": 35
255
+ },
256
+ {
257
+ "epoch": 1.244186046511628,
258
+ "grad_norm": 4.030057430267334,
259
+ "learning_rate": 1.8000000000000001e-06,
260
+ "loss": 0.1528,
261
+ "step": 36
262
+ },
263
+ {
264
+ "epoch": 1.2790697674418605,
265
+ "grad_norm": 3.560523509979248,
266
+ "learning_rate": 1.85e-06,
267
+ "loss": 0.1422,
268
+ "step": 37
269
+ },
270
+ {
271
+ "epoch": 1.3139534883720931,
272
+ "grad_norm": 3.3749780654907227,
273
+ "learning_rate": 1.9000000000000002e-06,
274
+ "loss": 0.1327,
275
+ "step": 38
276
+ },
277
+ {
278
+ "epoch": 1.3488372093023255,
279
+ "grad_norm": 2.7184131145477295,
280
+ "learning_rate": 1.9500000000000004e-06,
281
+ "loss": 0.1199,
282
+ "step": 39
283
+ },
284
+ {
285
+ "epoch": 1.3837209302325582,
286
+ "grad_norm": 2.8681583404541016,
287
+ "learning_rate": 2.0000000000000003e-06,
288
+ "loss": 0.1107,
289
+ "step": 40
290
+ },
291
+ {
292
+ "epoch": 1.4186046511627908,
293
+ "grad_norm": 2.8731987476348877,
294
+ "learning_rate": 2.05e-06,
295
+ "loss": 0.1063,
296
+ "step": 41
297
+ },
298
+ {
299
+ "epoch": 1.4534883720930232,
300
+ "grad_norm": 2.9522054195404053,
301
+ "learning_rate": 2.1000000000000002e-06,
302
+ "loss": 0.0989,
303
+ "step": 42
304
+ },
305
+ {
306
+ "epoch": 1.4883720930232558,
307
+ "grad_norm": 2.689038038253784,
308
+ "learning_rate": 2.15e-06,
309
+ "loss": 0.0949,
310
+ "step": 43
311
+ },
312
+ {
313
+ "epoch": 1.5232558139534884,
314
+ "grad_norm": 2.6328952312469482,
315
+ "learning_rate": 2.2e-06,
316
+ "loss": 0.1006,
317
+ "step": 44
318
+ },
319
+ {
320
+ "epoch": 1.558139534883721,
321
+ "grad_norm": 2.7056033611297607,
322
+ "learning_rate": 2.25e-06,
323
+ "loss": 0.0929,
324
+ "step": 45
325
+ },
326
+ {
327
+ "epoch": 1.5930232558139537,
328
+ "grad_norm": 2.568206310272217,
329
+ "learning_rate": 2.3000000000000004e-06,
330
+ "loss": 0.0881,
331
+ "step": 46
332
+ },
333
+ {
334
+ "epoch": 1.627906976744186,
335
+ "grad_norm": 2.714211940765381,
336
+ "learning_rate": 2.35e-06,
337
+ "loss": 0.096,
338
+ "step": 47
339
+ },
340
+ {
341
+ "epoch": 1.6627906976744184,
342
+ "grad_norm": 2.3650710582733154,
343
+ "learning_rate": 2.4000000000000003e-06,
344
+ "loss": 0.086,
345
+ "step": 48
346
+ },
347
+ {
348
+ "epoch": 1.697674418604651,
349
+ "grad_norm": 2.1763498783111572,
350
+ "learning_rate": 2.4500000000000003e-06,
351
+ "loss": 0.0858,
352
+ "step": 49
353
+ },
354
+ {
355
+ "epoch": 1.7325581395348837,
356
+ "grad_norm": 2.124727487564087,
357
+ "learning_rate": 2.5e-06,
358
+ "loss": 0.0843,
359
+ "step": 50
360
+ },
361
+ {
362
+ "epoch": 1.7674418604651163,
363
+ "grad_norm": 1.641269564628601,
364
+ "learning_rate": 2.55e-06,
365
+ "loss": 0.0803,
366
+ "step": 51
367
+ },
368
+ {
369
+ "epoch": 1.802325581395349,
370
+ "grad_norm": 1.5644842386245728,
371
+ "learning_rate": 2.6e-06,
372
+ "loss": 0.0779,
373
+ "step": 52
374
+ },
375
+ {
376
+ "epoch": 1.8372093023255816,
377
+ "grad_norm": 1.1404681205749512,
378
+ "learning_rate": 2.6500000000000005e-06,
379
+ "loss": 0.0765,
380
+ "step": 53
381
+ },
382
+ {
383
+ "epoch": 1.872093023255814,
384
+ "grad_norm": 0.8386123776435852,
385
+ "learning_rate": 2.7000000000000004e-06,
386
+ "loss": 0.0749,
387
+ "step": 54
388
+ },
389
+ {
390
+ "epoch": 1.9069767441860463,
391
+ "grad_norm": 1.8363338708877563,
392
+ "learning_rate": 2.7500000000000004e-06,
393
+ "loss": 0.0737,
394
+ "step": 55
395
+ },
396
+ {
397
+ "epoch": 1.941860465116279,
398
+ "grad_norm": 1.2105377912521362,
399
+ "learning_rate": 2.8000000000000003e-06,
400
+ "loss": 0.0762,
401
+ "step": 56
402
+ },
403
+ {
404
+ "epoch": 1.9767441860465116,
405
+ "grad_norm": 1.5502218008041382,
406
+ "learning_rate": 2.85e-06,
407
+ "loss": 0.07,
408
+ "step": 57
409
+ },
410
+ {
411
+ "epoch": 2.0,
412
+ "grad_norm": 1.0641449689865112,
413
+ "learning_rate": 2.9e-06,
414
+ "loss": 0.0679,
415
+ "step": 58
416
+ },
417
+ {
418
+ "epoch": 2.0348837209302326,
419
+ "grad_norm": 0.9201306104660034,
420
+ "learning_rate": 2.95e-06,
421
+ "loss": 0.0677,
422
+ "step": 59
423
+ },
424
+ {
425
+ "epoch": 2.0697674418604652,
426
+ "grad_norm": 0.5951386094093323,
427
+ "learning_rate": 3e-06,
428
+ "loss": 0.0654,
429
+ "step": 60
430
+ },
431
+ {
432
+ "epoch": 2.104651162790698,
433
+ "grad_norm": 0.8307608962059021,
434
+ "learning_rate": 3.05e-06,
435
+ "loss": 0.0706,
436
+ "step": 61
437
+ },
438
+ {
439
+ "epoch": 2.13953488372093,
440
+ "grad_norm": 0.7110892534255981,
441
+ "learning_rate": 3.1000000000000004e-06,
442
+ "loss": 0.0665,
443
+ "step": 62
444
+ },
445
+ {
446
+ "epoch": 2.1744186046511627,
447
+ "grad_norm": 0.6766234040260315,
448
+ "learning_rate": 3.1500000000000003e-06,
449
+ "loss": 0.0618,
450
+ "step": 63
451
+ },
452
+ {
453
+ "epoch": 2.2093023255813953,
454
+ "grad_norm": 0.3967410922050476,
455
+ "learning_rate": 3.2000000000000003e-06,
456
+ "loss": 0.0644,
457
+ "step": 64
458
+ },
459
+ {
460
+ "epoch": 2.244186046511628,
461
+ "grad_norm": 0.3713420331478119,
462
+ "learning_rate": 3.2500000000000002e-06,
463
+ "loss": 0.0607,
464
+ "step": 65
465
+ },
466
+ {
467
+ "epoch": 2.2790697674418605,
468
+ "grad_norm": 0.5613359212875366,
469
+ "learning_rate": 3.3000000000000006e-06,
470
+ "loss": 0.0623,
471
+ "step": 66
472
+ },
473
+ {
474
+ "epoch": 2.313953488372093,
475
+ "grad_norm": 0.5458635687828064,
476
+ "learning_rate": 3.3500000000000005e-06,
477
+ "loss": 0.0563,
478
+ "step": 67
479
+ },
480
+ {
481
+ "epoch": 2.3488372093023258,
482
+ "grad_norm": 0.4196176528930664,
483
+ "learning_rate": 3.4000000000000005e-06,
484
+ "loss": 0.0547,
485
+ "step": 68
486
+ },
487
+ {
488
+ "epoch": 2.383720930232558,
489
+ "grad_norm": 0.5757117867469788,
490
+ "learning_rate": 3.45e-06,
491
+ "loss": 0.0551,
492
+ "step": 69
493
+ },
494
+ {
495
+ "epoch": 2.4186046511627906,
496
+ "grad_norm": 0.45777687430381775,
497
+ "learning_rate": 3.5e-06,
498
+ "loss": 0.0575,
499
+ "step": 70
500
+ },
501
+ {
502
+ "epoch": 2.453488372093023,
503
+ "grad_norm": 0.5204553008079529,
504
+ "learning_rate": 3.5500000000000003e-06,
505
+ "loss": 0.057,
506
+ "step": 71
507
+ },
508
+ {
509
+ "epoch": 2.488372093023256,
510
+ "grad_norm": 0.6110821962356567,
511
+ "learning_rate": 3.6000000000000003e-06,
512
+ "loss": 0.0556,
513
+ "step": 72
514
+ },
515
+ {
516
+ "epoch": 2.5232558139534884,
517
+ "grad_norm": 0.45246991515159607,
518
+ "learning_rate": 3.65e-06,
519
+ "loss": 0.0564,
520
+ "step": 73
521
+ },
522
+ {
523
+ "epoch": 2.558139534883721,
524
+ "grad_norm": 0.441976398229599,
525
+ "learning_rate": 3.7e-06,
526
+ "loss": 0.0534,
527
+ "step": 74
528
+ },
529
+ {
530
+ "epoch": 2.5930232558139537,
531
+ "grad_norm": 0.46637651324272156,
532
+ "learning_rate": 3.7500000000000005e-06,
533
+ "loss": 0.0498,
534
+ "step": 75
535
+ },
536
+ {
537
+ "epoch": 2.6279069767441863,
538
+ "grad_norm": 0.482038289308548,
539
+ "learning_rate": 3.8000000000000005e-06,
540
+ "loss": 0.0521,
541
+ "step": 76
542
+ },
543
+ {
544
+ "epoch": 2.6627906976744184,
545
+ "grad_norm": 0.5914385318756104,
546
+ "learning_rate": 3.85e-06,
547
+ "loss": 0.0511,
548
+ "step": 77
549
+ },
550
+ {
551
+ "epoch": 2.697674418604651,
552
+ "grad_norm": 0.34532907605171204,
553
+ "learning_rate": 3.900000000000001e-06,
554
+ "loss": 0.0527,
555
+ "step": 78
556
+ },
557
+ {
558
+ "epoch": 2.7325581395348837,
559
+ "grad_norm": 0.35315006971359253,
560
+ "learning_rate": 3.95e-06,
561
+ "loss": 0.0486,
562
+ "step": 79
563
+ },
564
+ {
565
+ "epoch": 2.7674418604651163,
566
+ "grad_norm": 0.4521324634552002,
567
+ "learning_rate": 4.000000000000001e-06,
568
+ "loss": 0.0485,
569
+ "step": 80
570
+ },
571
+ {
572
+ "epoch": 2.802325581395349,
573
+ "grad_norm": 0.49457868933677673,
574
+ "learning_rate": 4.05e-06,
575
+ "loss": 0.0525,
576
+ "step": 81
577
+ },
578
+ {
579
+ "epoch": 2.8372093023255816,
580
+ "grad_norm": 0.7616601586341858,
581
+ "learning_rate": 4.1e-06,
582
+ "loss": 0.0528,
583
+ "step": 82
584
+ },
585
+ {
586
+ "epoch": 2.8720930232558137,
587
+ "grad_norm": 0.4791123569011688,
588
+ "learning_rate": 4.15e-06,
589
+ "loss": 0.0512,
590
+ "step": 83
591
+ },
592
+ {
593
+ "epoch": 2.9069767441860463,
594
+ "grad_norm": 0.38186997175216675,
595
+ "learning_rate": 4.2000000000000004e-06,
596
+ "loss": 0.0442,
597
+ "step": 84
598
+ },
599
+ {
600
+ "epoch": 2.941860465116279,
601
+ "grad_norm": 0.32648414373397827,
602
+ "learning_rate": 4.25e-06,
603
+ "loss": 0.0432,
604
+ "step": 85
605
+ },
606
+ {
607
+ "epoch": 2.9767441860465116,
608
+ "grad_norm": 0.410888671875,
609
+ "learning_rate": 4.3e-06,
610
+ "loss": 0.042,
611
+ "step": 86
612
+ },
613
+ {
614
+ "epoch": 3.0,
615
+ "grad_norm": 0.4732547104358673,
616
+ "learning_rate": 4.350000000000001e-06,
617
+ "loss": 0.0483,
618
+ "step": 87
619
+ },
620
+ {
621
+ "epoch": 3.0348837209302326,
622
+ "grad_norm": 0.6372231841087341,
623
+ "learning_rate": 4.4e-06,
624
+ "loss": 0.0382,
625
+ "step": 88
626
+ },
627
+ {
628
+ "epoch": 3.0697674418604652,
629
+ "grad_norm": 0.393078088760376,
630
+ "learning_rate": 4.450000000000001e-06,
631
+ "loss": 0.0403,
632
+ "step": 89
633
+ },
634
+ {
635
+ "epoch": 3.104651162790698,
636
+ "grad_norm": 0.4366248846054077,
637
+ "learning_rate": 4.5e-06,
638
+ "loss": 0.0367,
639
+ "step": 90
640
+ },
641
+ {
642
+ "epoch": 3.13953488372093,
643
+ "grad_norm": 0.301724374294281,
644
+ "learning_rate": 4.5500000000000005e-06,
645
+ "loss": 0.0351,
646
+ "step": 91
647
+ },
648
+ {
649
+ "epoch": 3.1744186046511627,
650
+ "grad_norm": 0.37665656208992004,
651
+ "learning_rate": 4.600000000000001e-06,
652
+ "loss": 0.03,
653
+ "step": 92
654
+ },
655
+ {
656
+ "epoch": 3.2093023255813953,
657
+ "grad_norm": 0.581331729888916,
658
+ "learning_rate": 4.65e-06,
659
+ "loss": 0.0316,
660
+ "step": 93
661
+ },
662
+ {
663
+ "epoch": 3.244186046511628,
664
+ "grad_norm": 0.44136878848075867,
665
+ "learning_rate": 4.7e-06,
666
+ "loss": 0.0312,
667
+ "step": 94
668
+ },
669
+ {
670
+ "epoch": 3.2790697674418605,
671
+ "grad_norm": 0.7624006271362305,
672
+ "learning_rate": 4.75e-06,
673
+ "loss": 0.0323,
674
+ "step": 95
675
+ },
676
+ {
677
+ "epoch": 3.313953488372093,
678
+ "grad_norm": 0.35294386744499207,
679
+ "learning_rate": 4.800000000000001e-06,
680
+ "loss": 0.0286,
681
+ "step": 96
682
+ },
683
+ {
684
+ "epoch": 3.3488372093023258,
685
+ "grad_norm": 0.6240035891532898,
686
+ "learning_rate": 4.85e-06,
687
+ "loss": 0.0289,
688
+ "step": 97
689
+ },
690
+ {
691
+ "epoch": 3.383720930232558,
692
+ "grad_norm": 0.40580570697784424,
693
+ "learning_rate": 4.9000000000000005e-06,
694
+ "loss": 0.0259,
695
+ "step": 98
696
+ },
697
+ {
698
+ "epoch": 3.4186046511627906,
699
+ "grad_norm": 0.2971636652946472,
700
+ "learning_rate": 4.95e-06,
701
+ "loss": 0.0251,
702
+ "step": 99
703
+ },
704
+ {
705
+ "epoch": 3.453488372093023,
706
+ "grad_norm": 0.3758476972579956,
707
+ "learning_rate": 5e-06,
708
+ "loss": 0.0244,
709
+ "step": 100
710
+ },
711
+ {
712
+ "epoch": 3.488372093023256,
713
+ "grad_norm": 0.3845921754837036,
714
+ "learning_rate": 4.997332437005932e-06,
715
+ "loss": 0.0242,
716
+ "step": 101
717
+ },
718
+ {
719
+ "epoch": 3.5232558139534884,
720
+ "grad_norm": 0.48894616961479187,
721
+ "learning_rate": 4.989335440737587e-06,
722
+ "loss": 0.0252,
723
+ "step": 102
724
+ },
725
+ {
726
+ "epoch": 3.558139534883721,
727
+ "grad_norm": 0.36844325065612793,
728
+ "learning_rate": 4.976026077188013e-06,
729
+ "loss": 0.0197,
730
+ "step": 103
731
+ },
732
+ {
733
+ "epoch": 3.5930232558139537,
734
+ "grad_norm": 0.2800253629684448,
735
+ "learning_rate": 4.957432749209755e-06,
736
+ "loss": 0.0187,
737
+ "step": 104
738
+ },
739
+ {
740
+ "epoch": 3.6279069767441863,
741
+ "grad_norm": 0.34042924642562866,
742
+ "learning_rate": 4.933595135901733e-06,
743
+ "loss": 0.0182,
744
+ "step": 105
745
+ },
746
+ {
747
+ "epoch": 3.6627906976744184,
748
+ "grad_norm": 0.33217477798461914,
749
+ "learning_rate": 4.904564107932048e-06,
750
+ "loss": 0.0166,
751
+ "step": 106
752
+ },
753
+ {
754
+ "epoch": 3.697674418604651,
755
+ "grad_norm": 0.32867398858070374,
756
+ "learning_rate": 4.870401618977415e-06,
757
+ "loss": 0.017,
758
+ "step": 107
759
+ },
760
+ {
761
+ "epoch": 3.7325581395348837,
762
+ "grad_norm": 0.24199356138706207,
763
+ "learning_rate": 4.83118057351089e-06,
764
+ "loss": 0.0138,
765
+ "step": 108
766
+ },
767
+ {
768
+ "epoch": 3.7674418604651163,
769
+ "grad_norm": 0.3216392695903778,
770
+ "learning_rate": 4.786984671220053e-06,
771
+ "loss": 0.0171,
772
+ "step": 109
773
+ },
774
+ {
775
+ "epoch": 3.802325581395349,
776
+ "grad_norm": 0.3574189841747284,
777
+ "learning_rate": 4.737908228387656e-06,
778
+ "loss": 0.0146,
779
+ "step": 110
780
+ },
781
+ {
782
+ "epoch": 3.8372093023255816,
783
+ "grad_norm": 0.30931738018989563,
784
+ "learning_rate": 4.684055976615924e-06,
785
+ "loss": 0.0128,
786
+ "step": 111
787
+ },
788
+ {
789
+ "epoch": 3.8720930232558137,
790
+ "grad_norm": 0.2074785828590393,
791
+ "learning_rate": 4.625542839324036e-06,
792
+ "loss": 0.0089,
793
+ "step": 112
794
+ }
795
+ ],
796
+ "logging_steps": 1,
797
+ "max_steps": 168,
798
+ "num_input_tokens_seen": 0,
799
+ "num_train_epochs": 6,
800
+ "save_steps": 28,
801
+ "stateful_callbacks": {
802
+ "TrainerControl": {
803
+ "args": {
804
+ "should_epoch_stop": false,
805
+ "should_evaluate": false,
806
+ "should_log": false,
807
+ "should_save": true,
808
+ "should_training_stop": false
809
+ },
810
+ "attributes": {}
811
+ }
812
+ },
813
+ "total_flos": 9.521111556499177e+18,
814
+ "train_batch_size": 4,
815
+ "trial_name": null,
816
+ "trial_params": null
817
+ }
checkpoint-112/training_args.bin ADDED
@@ -0,0 +1,3 @@
 
 
 
 
1
+ version https://git-lfs.github.com/spec/v1
2
+ oid sha256:cb267fa439e36c44dd8202fc6ee1621eee6c4bd83bf7c45ed193d0b07b59aaca
3
+ size 8056
checkpoint-112/zero_to_fp32.py ADDED
@@ -0,0 +1,604 @@
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
1
+ #!/usr/bin/env python
2
+
3
+ # Copyright (c) Microsoft Corporation.
4
+ # SPDX-License-Identifier: Apache-2.0
5
+
6
+ # DeepSpeed Team
7
+
8
+ # This script extracts fp32 consolidated weights from a zero 1, 2 and 3 DeepSpeed checkpoints. It gets
9
+ # copied into the top level checkpoint dir, so the user can easily do the conversion at any point in
10
+ # the future. Once extracted, the weights don't require DeepSpeed and can be used in any
11
+ # application.
12
+ #
13
+ # example: python zero_to_fp32.py . pytorch_model.bin
14
+
15
+ import argparse
16
+ import torch
17
+ import glob
18
+ import math
19
+ import os
20
+ import re
21
+ from collections import OrderedDict
22
+ from dataclasses import dataclass
23
+
24
+ # while this script doesn't use deepspeed to recover data, since the checkpoints are pickled with
25
+ # DeepSpeed data structures it has to be available in the current python environment.
26
+ from deepspeed.utils import logger
27
+ from deepspeed.checkpoint.constants import (DS_VERSION, OPTIMIZER_STATE_DICT, SINGLE_PARTITION_OF_FP32_GROUPS,
28
+ FP32_FLAT_GROUPS, ZERO_STAGE, PARTITION_COUNT, PARAM_SHAPES, BUFFER_NAMES,
29
+ FROZEN_PARAM_SHAPES, FROZEN_PARAM_FRAGMENTS)
30
+
31
+
32
+ @dataclass
33
+ class zero_model_state:
34
+ buffers: dict()
35
+ param_shapes: dict()
36
+ shared_params: list
37
+ ds_version: int
38
+ frozen_param_shapes: dict()
39
+ frozen_param_fragments: dict()
40
+
41
+
42
+ debug = 0
43
+
44
+ # load to cpu
45
+ device = torch.device('cpu')
46
+
47
+
48
+ def atoi(text):
49
+ return int(text) if text.isdigit() else text
50
+
51
+
52
+ def natural_keys(text):
53
+ '''
54
+ alist.sort(key=natural_keys) sorts in human order
55
+ http://nedbatchelder.com/blog/200712/human_sorting.html
56
+ (See Toothy's implementation in the comments)
57
+ '''
58
+ return [atoi(c) for c in re.split(r'(\d+)', text)]
59
+
60
+
61
+ def get_model_state_file(checkpoint_dir, zero_stage):
62
+ if not os.path.isdir(checkpoint_dir):
63
+ raise FileNotFoundError(f"Directory '{checkpoint_dir}' doesn't exist")
64
+
65
+ # there should be only one file
66
+ if zero_stage <= 2:
67
+ file = os.path.join(checkpoint_dir, "mp_rank_00_model_states.pt")
68
+ elif zero_stage == 3:
69
+ file = os.path.join(checkpoint_dir, "zero_pp_rank_0_mp_rank_00_model_states.pt")
70
+
71
+ if not os.path.exists(file):
72
+ raise FileNotFoundError(f"can't find model states file at '{file}'")
73
+
74
+ return file
75
+
76
+
77
+ def get_checkpoint_files(checkpoint_dir, glob_pattern):
78
+ # XXX: need to test that this simple glob rule works for multi-node setup too
79
+ ckpt_files = sorted(glob.glob(os.path.join(checkpoint_dir, glob_pattern)), key=natural_keys)
80
+
81
+ if len(ckpt_files) == 0:
82
+ raise FileNotFoundError(f"can't find {glob_pattern} files in directory '{checkpoint_dir}'")
83
+
84
+ return ckpt_files
85
+
86
+
87
+ def get_optim_files(checkpoint_dir):
88
+ return get_checkpoint_files(checkpoint_dir, "*_optim_states.pt")
89
+
90
+
91
+ def get_model_state_files(checkpoint_dir):
92
+ return get_checkpoint_files(checkpoint_dir, "*_model_states.pt")
93
+
94
+
95
+ def parse_model_states(files):
96
+ zero_model_states = []
97
+ for file in files:
98
+ state_dict = torch.load(file, map_location=device)
99
+
100
+ if BUFFER_NAMES not in state_dict:
101
+ raise ValueError(f"{file} is not a model state checkpoint")
102
+ buffer_names = state_dict[BUFFER_NAMES]
103
+ if debug:
104
+ print("Found buffers:", buffer_names)
105
+
106
+ # recover just the buffers while restoring them to fp32 if they were saved in fp16
107
+ buffers = {k: v.float() for k, v in state_dict["module"].items() if k in buffer_names}
108
+ param_shapes = state_dict[PARAM_SHAPES]
109
+
110
+ # collect parameters that are included in param_shapes
111
+ param_names = []
112
+ for s in param_shapes:
113
+ for name in s.keys():
114
+ param_names.append(name)
115
+
116
+ # update with frozen parameters
117
+ frozen_param_shapes = state_dict.get(FROZEN_PARAM_SHAPES, None)
118
+ if frozen_param_shapes is not None:
119
+ if debug:
120
+ print(f"Found frozen_param_shapes: {frozen_param_shapes}")
121
+ param_names += list(frozen_param_shapes.keys())
122
+
123
+ # handle shared params
124
+ shared_params = [[k, v] for k, v in state_dict["shared_params"].items()]
125
+
126
+ ds_version = state_dict.get(DS_VERSION, None)
127
+
128
+ frozen_param_fragments = state_dict.get(FROZEN_PARAM_FRAGMENTS, None)
129
+
130
+ z_model_state = zero_model_state(buffers=buffers,
131
+ param_shapes=param_shapes,
132
+ shared_params=shared_params,
133
+ ds_version=ds_version,
134
+ frozen_param_shapes=frozen_param_shapes,
135
+ frozen_param_fragments=frozen_param_fragments)
136
+ zero_model_states.append(z_model_state)
137
+
138
+ return zero_model_states
139
+
140
+
141
+ def parse_optim_states(files, ds_checkpoint_dir):
142
+
143
+ total_files = len(files)
144
+ state_dicts = []
145
+ for f in files:
146
+ state_dict = torch.load(f, map_location=device)
147
+ # immediately discard the potentially huge 2 optimizer states as we only care for fp32 master weights
148
+ # and also handle the case where it was already removed by another helper script
149
+ state_dict["optimizer_state_dict"].pop("optimizer_state_dict", None)
150
+ state_dicts.append(state_dict)
151
+
152
+ if not ZERO_STAGE in state_dicts[0][OPTIMIZER_STATE_DICT]:
153
+ raise ValueError(f"{files[0]} is not a zero checkpoint")
154
+ zero_stage = state_dicts[0][OPTIMIZER_STATE_DICT][ZERO_STAGE]
155
+ world_size = state_dicts[0][OPTIMIZER_STATE_DICT][PARTITION_COUNT]
156
+
157
+ # For ZeRO-2 each param group can have different partition_count as data parallelism for expert
158
+ # parameters can be different from data parallelism for non-expert parameters. So we can just
159
+ # use the max of the partition_count to get the dp world_size.
160
+
161
+ if type(world_size) is list:
162
+ world_size = max(world_size)
163
+
164
+ if world_size != total_files:
165
+ raise ValueError(
166
+ f"Expected {world_size} of '*_optim_states.pt' under '{ds_checkpoint_dir}' but found {total_files} files. "
167
+ "Possibly due to an overwrite of an old checkpoint, or a checkpoint didn't get saved by one or more processes."
168
+ )
169
+
170
+ # the groups are named differently in each stage
171
+ if zero_stage <= 2:
172
+ fp32_groups_key = SINGLE_PARTITION_OF_FP32_GROUPS
173
+ elif zero_stage == 3:
174
+ fp32_groups_key = FP32_FLAT_GROUPS
175
+ else:
176
+ raise ValueError(f"unknown zero stage {zero_stage}")
177
+
178
+ if zero_stage <= 2:
179
+ fp32_flat_groups = [state_dicts[i][OPTIMIZER_STATE_DICT][fp32_groups_key] for i in range(len(state_dicts))]
180
+ elif zero_stage == 3:
181
+ # if there is more than one param group, there will be multiple flattened tensors - one
182
+ # flattened tensor per group - for simplicity merge them into a single tensor
183
+ #
184
+ # XXX: could make the script more memory efficient for when there are multiple groups - it
185
+ # will require matching the sub-lists of param_shapes for each param group flattened tensor
186
+
187
+ fp32_flat_groups = [
188
+ torch.cat(state_dicts[i][OPTIMIZER_STATE_DICT][fp32_groups_key], 0) for i in range(len(state_dicts))
189
+ ]
190
+
191
+ return zero_stage, world_size, fp32_flat_groups
192
+
193
+
194
+ def _get_fp32_state_dict_from_zero_checkpoint(ds_checkpoint_dir, exclude_frozen_parameters):
195
+ """
196
+ Returns fp32 state_dict reconstructed from ds checkpoint
197
+
198
+ Args:
199
+ - ``ds_checkpoint_dir``: path to the deepspeed checkpoint folder (where the optimizer files are)
200
+
201
+ """
202
+ print(f"Processing zero checkpoint '{ds_checkpoint_dir}'")
203
+
204
+ optim_files = get_optim_files(ds_checkpoint_dir)
205
+ zero_stage, world_size, fp32_flat_groups = parse_optim_states(optim_files, ds_checkpoint_dir)
206
+ print(f"Detected checkpoint of type zero stage {zero_stage}, world_size: {world_size}")
207
+
208
+ model_files = get_model_state_files(ds_checkpoint_dir)
209
+
210
+ zero_model_states = parse_model_states(model_files)
211
+ print(f'Parsing checkpoint created by deepspeed=={zero_model_states[0].ds_version}')
212
+
213
+ if zero_stage <= 2:
214
+ return _get_fp32_state_dict_from_zero2_checkpoint(world_size, fp32_flat_groups, zero_model_states,
215
+ exclude_frozen_parameters)
216
+ elif zero_stage == 3:
217
+ return _get_fp32_state_dict_from_zero3_checkpoint(world_size, fp32_flat_groups, zero_model_states,
218
+ exclude_frozen_parameters)
219
+
220
+
221
+ def _zero2_merge_frozen_params(state_dict, zero_model_states):
222
+ if zero_model_states[0].frozen_param_shapes is None or len(zero_model_states[0].frozen_param_shapes) == 0:
223
+ return
224
+
225
+ frozen_param_shapes = zero_model_states[0].frozen_param_shapes
226
+ frozen_param_fragments = zero_model_states[0].frozen_param_fragments
227
+
228
+ if debug:
229
+ num_elem = sum(s.numel() for s in frozen_param_shapes.values())
230
+ print(f'rank 0: {FROZEN_PARAM_SHAPES}.numel = {num_elem}')
231
+
232
+ wanted_params = len(frozen_param_shapes)
233
+ wanted_numel = sum(s.numel() for s in frozen_param_shapes.values())
234
+ avail_numel = sum([p.numel() for p in frozen_param_fragments.values()])
235
+ print(f'Frozen params: Have {avail_numel} numels to process.')
236
+ print(f'Frozen params: Need {wanted_numel} numels in {wanted_params} params')
237
+
238
+ total_params = 0
239
+ total_numel = 0
240
+ for name, shape in frozen_param_shapes.items():
241
+ total_params += 1
242
+ unpartitioned_numel = shape.numel()
243
+ total_numel += unpartitioned_numel
244
+
245
+ state_dict[name] = frozen_param_fragments[name]
246
+
247
+ if debug:
248
+ print(f"{name} full shape: {shape} unpartitioned numel {unpartitioned_numel} ")
249
+
250
+ print(f"Reconstructed Frozen fp32 state dict with {total_params} params {total_numel} elements")
251
+
252
+
253
+ def _has_callable(obj, fn):
254
+ attr = getattr(obj, fn, None)
255
+ return callable(attr)
256
+
257
+
258
+ def _zero2_merge_trainable_params(state_dict, world_size, fp32_flat_groups, zero_model_states):
259
+ param_shapes = zero_model_states[0].param_shapes
260
+
261
+ # Reconstruction protocol:
262
+ #
263
+ # XXX: document this
264
+
265
+ if debug:
266
+ for i in range(world_size):
267
+ for j in range(len(fp32_flat_groups[0])):
268
+ print(f"{FP32_FLAT_GROUPS}[{i}][{j}].shape={fp32_flat_groups[i][j].shape}")
269
+
270
+ # XXX: memory usage doubles here (zero2)
271
+ num_param_groups = len(fp32_flat_groups[0])
272
+ merged_single_partition_of_fp32_groups = []
273
+ for i in range(num_param_groups):
274
+ merged_partitions = [sd[i] for sd in fp32_flat_groups]
275
+ full_single_fp32_vector = torch.cat(merged_partitions, 0)
276
+ merged_single_partition_of_fp32_groups.append(full_single_fp32_vector)
277
+ avail_numel = sum(
278
+ [full_single_fp32_vector.numel() for full_single_fp32_vector in merged_single_partition_of_fp32_groups])
279
+
280
+ if debug:
281
+ wanted_params = sum([len(shapes) for shapes in param_shapes])
282
+ wanted_numel = sum([sum(shape.numel() for shape in shapes.values()) for shapes in param_shapes])
283
+ # not asserting if there is a mismatch due to possible padding
284
+ print(f"Have {avail_numel} numels to process.")
285
+ print(f"Need {wanted_numel} numels in {wanted_params} params.")
286
+
287
+ # params
288
+ # XXX: for huge models that can't fit into the host's RAM we will have to recode this to support
289
+ # out-of-core computing solution
290
+ total_numel = 0
291
+ total_params = 0
292
+ for shapes, full_single_fp32_vector in zip(param_shapes, merged_single_partition_of_fp32_groups):
293
+ offset = 0
294
+ avail_numel = full_single_fp32_vector.numel()
295
+ for name, shape in shapes.items():
296
+
297
+ unpartitioned_numel = shape.numel() if _has_callable(shape, 'numel') else math.prod(shape)
298
+ total_numel += unpartitioned_numel
299
+ total_params += 1
300
+
301
+ if debug:
302
+ print(f"{name} full shape: {shape} unpartitioned numel {unpartitioned_numel} ")
303
+ state_dict[name] = full_single_fp32_vector.narrow(0, offset, unpartitioned_numel).view(shape)
304
+ offset += unpartitioned_numel
305
+
306
+ # Z2 started to align to 2*world_size to improve nccl performance. Therefore both offset and
307
+ # avail_numel can differ by anywhere between 0..2*world_size. Due to two unrelated complex
308
+ # paddings performed in the code it's almost impossible to predict the exact numbers w/o the
309
+ # live optimizer object, so we are checking that the numbers are within the right range
310
+ align_to = 2 * world_size
311
+
312
+ def zero2_align(x):
313
+ return align_to * math.ceil(x / align_to)
314
+
315
+ if debug:
316
+ print(f"original offset={offset}, avail_numel={avail_numel}")
317
+
318
+ offset = zero2_align(offset)
319
+ avail_numel = zero2_align(avail_numel)
320
+
321
+ if debug:
322
+ print(f"aligned offset={offset}, avail_numel={avail_numel}")
323
+
324
+ # Sanity check
325
+ if offset != avail_numel:
326
+ raise ValueError(f"consumed {offset} numels out of {avail_numel} - something is wrong")
327
+
328
+ print(f"Reconstructed fp32 state dict with {total_params} params {total_numel} elements")
329
+
330
+
331
+ def _get_fp32_state_dict_from_zero2_checkpoint(world_size, fp32_flat_groups, zero_model_states,
332
+ exclude_frozen_parameters):
333
+ state_dict = OrderedDict()
334
+
335
+ # buffers
336
+ buffers = zero_model_states[0].buffers
337
+ state_dict.update(buffers)
338
+ if debug:
339
+ print(f"added {len(buffers)} buffers")
340
+
341
+ if not exclude_frozen_parameters:
342
+ _zero2_merge_frozen_params(state_dict, zero_model_states)
343
+
344
+ _zero2_merge_trainable_params(state_dict, world_size, fp32_flat_groups, zero_model_states)
345
+
346
+ # recover shared parameters
347
+ for pair in zero_model_states[0].shared_params:
348
+ if pair[1] in state_dict:
349
+ state_dict[pair[0]] = state_dict[pair[1]]
350
+
351
+ return state_dict
352
+
353
+
354
+ def zero3_partitioned_param_info(unpartitioned_numel, world_size):
355
+ remainder = unpartitioned_numel % world_size
356
+ padding_numel = (world_size - remainder) if remainder else 0
357
+ partitioned_numel = math.ceil(unpartitioned_numel / world_size)
358
+ return partitioned_numel, padding_numel
359
+
360
+
361
+ def _zero3_merge_frozen_params(state_dict, world_size, zero_model_states):
362
+ if zero_model_states[0].frozen_param_shapes is None or len(zero_model_states[0].frozen_param_shapes) == 0:
363
+ return
364
+
365
+ if debug:
366
+ for i in range(world_size):
367
+ num_elem = sum(s.numel() for s in zero_model_states[i].frozen_param_fragments.values())
368
+ print(f'rank {i}: {FROZEN_PARAM_SHAPES}.numel = {num_elem}')
369
+
370
+ frozen_param_shapes = zero_model_states[0].frozen_param_shapes
371
+ wanted_params = len(frozen_param_shapes)
372
+ wanted_numel = sum(s.numel() for s in frozen_param_shapes.values())
373
+ avail_numel = sum([p.numel() for p in zero_model_states[0].frozen_param_fragments.values()]) * world_size
374
+ print(f'Frozen params: Have {avail_numel} numels to process.')
375
+ print(f'Frozen params: Need {wanted_numel} numels in {wanted_params} params')
376
+
377
+ total_params = 0
378
+ total_numel = 0
379
+ for name, shape in zero_model_states[0].frozen_param_shapes.items():
380
+ total_params += 1
381
+ unpartitioned_numel = shape.numel()
382
+ total_numel += unpartitioned_numel
383
+
384
+ param_frags = tuple(model_state.frozen_param_fragments[name] for model_state in zero_model_states)
385
+ state_dict[name] = torch.cat(param_frags, 0).narrow(0, 0, unpartitioned_numel).view(shape)
386
+
387
+ partitioned_numel, partitioned_padding_numel = zero3_partitioned_param_info(unpartitioned_numel, world_size)
388
+
389
+ if debug:
390
+ print(
391
+ f"Frozen params: {total_params} {name} full shape: {shape} partition0 numel={partitioned_numel} partitioned_padding_numel={partitioned_padding_numel}"
392
+ )
393
+
394
+ print(f"Reconstructed Frozen fp32 state dict with {total_params} params {total_numel} elements")
395
+
396
+
397
+ def _zero3_merge_trainable_params(state_dict, world_size, fp32_flat_groups, zero_model_states):
398
+ param_shapes = zero_model_states[0].param_shapes
399
+ avail_numel = fp32_flat_groups[0].numel() * world_size
400
+ # Reconstruction protocol: For zero3 we need to zip the partitions together at boundary of each
401
+ # param, re-consolidating each param, while dealing with padding if any
402
+
403
+ # merge list of dicts, preserving order
404
+ param_shapes = {k: v for d in param_shapes for k, v in d.items()}
405
+
406
+ if debug:
407
+ for i in range(world_size):
408
+ print(f"{FP32_FLAT_GROUPS}[{i}].shape={fp32_flat_groups[i].shape}")
409
+
410
+ wanted_params = len(param_shapes)
411
+ wanted_numel = sum(shape.numel() for shape in param_shapes.values())
412
+ # not asserting if there is a mismatch due to possible padding
413
+ avail_numel = fp32_flat_groups[0].numel() * world_size
414
+ print(f"Trainable params: Have {avail_numel} numels to process.")
415
+ print(f"Trainable params: Need {wanted_numel} numels in {wanted_params} params.")
416
+
417
+ # params
418
+ # XXX: for huge models that can't fit into the host's RAM we will have to recode this to support
419
+ # out-of-core computing solution
420
+ offset = 0
421
+ total_numel = 0
422
+ total_params = 0
423
+ for name, shape in param_shapes.items():
424
+
425
+ unpartitioned_numel = shape.numel()
426
+ total_numel += unpartitioned_numel
427
+ total_params += 1
428
+
429
+ partitioned_numel, partitioned_padding_numel = zero3_partitioned_param_info(unpartitioned_numel, world_size)
430
+
431
+ if debug:
432
+ print(
433
+ f"Trainable params: {total_params} {name} full shape: {shape} partition0 numel={partitioned_numel} partitioned_padding_numel={partitioned_padding_numel}"
434
+ )
435
+
436
+ # XXX: memory usage doubles here
437
+ state_dict[name] = torch.cat(
438
+ tuple(fp32_flat_groups[i].narrow(0, offset, partitioned_numel) for i in range(world_size)),
439
+ 0).narrow(0, 0, unpartitioned_numel).view(shape)
440
+ offset += partitioned_numel
441
+
442
+ offset *= world_size
443
+
444
+ # Sanity check
445
+ if offset != avail_numel:
446
+ raise ValueError(f"consumed {offset} numels out of {avail_numel} - something is wrong")
447
+
448
+ print(f"Reconstructed Trainable fp32 state dict with {total_params} params {total_numel} elements")
449
+
450
+
451
+ def _get_fp32_state_dict_from_zero3_checkpoint(world_size, fp32_flat_groups, zero_model_states,
452
+ exclude_frozen_parameters):
453
+ state_dict = OrderedDict()
454
+
455
+ # buffers
456
+ buffers = zero_model_states[0].buffers
457
+ state_dict.update(buffers)
458
+ if debug:
459
+ print(f"added {len(buffers)} buffers")
460
+
461
+ if not exclude_frozen_parameters:
462
+ _zero3_merge_frozen_params(state_dict, world_size, zero_model_states)
463
+
464
+ _zero3_merge_trainable_params(state_dict, world_size, fp32_flat_groups, zero_model_states)
465
+
466
+ # recover shared parameters
467
+ for pair in zero_model_states[0].shared_params:
468
+ if pair[1] in state_dict:
469
+ state_dict[pair[0]] = state_dict[pair[1]]
470
+
471
+ return state_dict
472
+
473
+
474
+ def get_fp32_state_dict_from_zero_checkpoint(checkpoint_dir, tag=None, exclude_frozen_parameters=False):
475
+ """
476
+ Convert ZeRO 2 or 3 checkpoint into a single fp32 consolidated state_dict that can be loaded with
477
+ ``load_state_dict()`` and used for training without DeepSpeed or shared with others, for example
478
+ via a model hub.
479
+
480
+ Args:
481
+ - ``checkpoint_dir``: path to the desired checkpoint folder
482
+ - ``tag``: checkpoint tag used as a unique identifier for checkpoint. If not provided will attempt to load tag in 'latest' file. e.g., ``global_step14``
483
+ - ``exclude_frozen_parameters``: exclude frozen parameters
484
+
485
+ Returns:
486
+ - pytorch ``state_dict``
487
+
488
+ Note: this approach may not work if your application doesn't have sufficient free CPU memory and
489
+ you may need to use the offline approach using the ``zero_to_fp32.py`` script that is saved with
490
+ the checkpoint.
491
+
492
+ A typical usage might be ::
493
+
494
+ from deepspeed.utils.zero_to_fp32 import get_fp32_state_dict_from_zero_checkpoint
495
+ # do the training and checkpoint saving
496
+ state_dict = get_fp32_state_dict_from_zero_checkpoint(checkpoint_dir) # already on cpu
497
+ model = model.cpu() # move to cpu
498
+ model.load_state_dict(state_dict)
499
+ # submit to model hub or save the model to share with others
500
+
501
+ In this example the ``model`` will no longer be usable in the deepspeed context of the same
502
+ application. i.e. you will need to re-initialize the deepspeed engine, since
503
+ ``model.load_state_dict(state_dict)`` will remove all the deepspeed magic from it.
504
+
505
+ If you want it all done for you, use ``load_state_dict_from_zero_checkpoint`` instead.
506
+
507
+ """
508
+ if tag is None:
509
+ latest_path = os.path.join(checkpoint_dir, 'latest')
510
+ if os.path.isfile(latest_path):
511
+ with open(latest_path, 'r') as fd:
512
+ tag = fd.read().strip()
513
+ else:
514
+ raise ValueError(f"Unable to find 'latest' file at {latest_path}")
515
+
516
+ ds_checkpoint_dir = os.path.join(checkpoint_dir, tag)
517
+
518
+ if not os.path.isdir(ds_checkpoint_dir):
519
+ raise FileNotFoundError(f"Directory '{ds_checkpoint_dir}' doesn't exist")
520
+
521
+ return _get_fp32_state_dict_from_zero_checkpoint(ds_checkpoint_dir, exclude_frozen_parameters)
522
+
523
+
524
+ def convert_zero_checkpoint_to_fp32_state_dict(checkpoint_dir, output_file, tag=None, exclude_frozen_parameters=False):
525
+ """
526
+ Convert ZeRO 2 or 3 checkpoint into a single fp32 consolidated ``state_dict`` file that can be
527
+ loaded with ``torch.load(file)`` + ``load_state_dict()`` and used for training without DeepSpeed.
528
+
529
+ Args:
530
+ - ``checkpoint_dir``: path to the desired checkpoint folder. (one that contains the tag-folder, like ``global_step14``)
531
+ - ``output_file``: path to the pytorch fp32 state_dict output file (e.g. path/pytorch_model.bin)
532
+ - ``tag``: checkpoint tag used as a unique identifier for checkpoint. If not provided will attempt to load tag in the file named ``latest`` in the checkpoint folder, e.g., ``global_step14``
533
+ - ``exclude_frozen_parameters``: exclude frozen parameters
534
+ """
535
+
536
+ state_dict = get_fp32_state_dict_from_zero_checkpoint(checkpoint_dir, tag, exclude_frozen_parameters)
537
+ print(f"Saving fp32 state dict to {output_file}")
538
+ torch.save(state_dict, output_file)
539
+
540
+
541
+ def load_state_dict_from_zero_checkpoint(model, checkpoint_dir, tag=None):
542
+ """
543
+ 1. Put the provided model to cpu
544
+ 2. Convert ZeRO 2 or 3 checkpoint into a single fp32 consolidated ``state_dict``
545
+ 3. Load it into the provided model
546
+
547
+ Args:
548
+ - ``model``: the model object to update
549
+ - ``checkpoint_dir``: path to the desired checkpoint folder. (one that contains the tag-folder, like ``global_step14``)
550
+ - ``tag``: checkpoint tag used as a unique identifier for checkpoint. If not provided will attempt to load tag in the file named ``latest`` in the checkpoint folder, e.g., ``global_step14``
551
+
552
+ Returns:
553
+ - ``model`: modified model
554
+
555
+ Make sure you have plenty of CPU memory available before you call this function. If you don't
556
+ have enough use the ``zero_to_fp32.py`` utility to do the conversion. You will find it
557
+ conveniently placed for you in the checkpoint folder.
558
+
559
+ A typical usage might be ::
560
+
561
+ from deepspeed.utils.zero_to_fp32 import load_state_dict_from_zero_checkpoint
562
+ model = load_state_dict_from_zero_checkpoint(trainer.model, checkpoint_dir)
563
+ # submit to model hub or save the model to share with others
564
+
565
+ Note, that once this was run, the ``model`` will no longer be usable in the deepspeed context
566
+ of the same application. i.e. you will need to re-initialize the deepspeed engine, since
567
+ ``model.load_state_dict(state_dict)`` will remove all the deepspeed magic from it.
568
+
569
+ """
570
+ logger.info(f"Extracting fp32 weights")
571
+ state_dict = get_fp32_state_dict_from_zero_checkpoint(checkpoint_dir, tag)
572
+
573
+ logger.info(f"Overwriting model with fp32 weights")
574
+ model = model.cpu()
575
+ model.load_state_dict(state_dict, strict=False)
576
+
577
+ return model
578
+
579
+
580
+ if __name__ == "__main__":
581
+
582
+ parser = argparse.ArgumentParser()
583
+ parser.add_argument("checkpoint_dir",
584
+ type=str,
585
+ help="path to the desired checkpoint folder, e.g., path/checkpoint-12")
586
+ parser.add_argument(
587
+ "output_file",
588
+ type=str,
589
+ help="path to the pytorch fp32 state_dict output file (e.g. path/checkpoint-12/pytorch_model.bin)")
590
+ parser.add_argument("-t",
591
+ "--tag",
592
+ type=str,
593
+ default=None,
594
+ help="checkpoint tag used as a unique identifier for checkpoint. e.g., global_step1")
595
+ parser.add_argument("--exclude_frozen_parameters", action='store_true', help="exclude frozen parameters")
596
+ parser.add_argument("-d", "--debug", action='store_true', help="enable debug")
597
+ args = parser.parse_args()
598
+
599
+ debug = args.debug
600
+
601
+ convert_zero_checkpoint_to_fp32_state_dict(args.checkpoint_dir,
602
+ args.output_file,
603
+ tag=args.tag,
604
+ exclude_frozen_parameters=args.exclude_frozen_parameters)
checkpoint-140/README.md ADDED
@@ -0,0 +1,202 @@
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
1
+ ---
2
+ base_model: deepseek-ai/DeepSeek-R1-Distill-Llama-70B
3
+ library_name: peft
4
+ ---
5
+
6
+ # Model Card for Model ID
7
+
8
+ <!-- Provide a quick summary of what the model is/does. -->
9
+
10
+
11
+
12
+ ## Model Details
13
+
14
+ ### Model Description
15
+
16
+ <!-- Provide a longer summary of what this model is. -->
17
+
18
+
19
+
20
+ - **Developed by:** [More Information Needed]
21
+ - **Funded by [optional]:** [More Information Needed]
22
+ - **Shared by [optional]:** [More Information Needed]
23
+ - **Model type:** [More Information Needed]
24
+ - **Language(s) (NLP):** [More Information Needed]
25
+ - **License:** [More Information Needed]
26
+ - **Finetuned from model [optional]:** [More Information Needed]
27
+
28
+ ### Model Sources [optional]
29
+
30
+ <!-- Provide the basic links for the model. -->
31
+
32
+ - **Repository:** [More Information Needed]
33
+ - **Paper [optional]:** [More Information Needed]
34
+ - **Demo [optional]:** [More Information Needed]
35
+
36
+ ## Uses
37
+
38
+ <!-- Address questions around how the model is intended to be used, including the foreseeable users of the model and those affected by the model. -->
39
+
40
+ ### Direct Use
41
+
42
+ <!-- This section is for the model use without fine-tuning or plugging into a larger ecosystem/app. -->
43
+
44
+ [More Information Needed]
45
+
46
+ ### Downstream Use [optional]
47
+
48
+ <!-- This section is for the model use when fine-tuned for a task, or when plugged into a larger ecosystem/app -->
49
+
50
+ [More Information Needed]
51
+
52
+ ### Out-of-Scope Use
53
+
54
+ <!-- This section addresses misuse, malicious use, and uses that the model will not work well for. -->
55
+
56
+ [More Information Needed]
57
+
58
+ ## Bias, Risks, and Limitations
59
+
60
+ <!-- This section is meant to convey both technical and sociotechnical limitations. -->
61
+
62
+ [More Information Needed]
63
+
64
+ ### Recommendations
65
+
66
+ <!-- This section is meant to convey recommendations with respect to the bias, risk, and technical limitations. -->
67
+
68
+ Users (both direct and downstream) should be made aware of the risks, biases and limitations of the model. More information needed for further recommendations.
69
+
70
+ ## How to Get Started with the Model
71
+
72
+ Use the code below to get started with the model.
73
+
74
+ [More Information Needed]
75
+
76
+ ## Training Details
77
+
78
+ ### Training Data
79
+
80
+ <!-- This should link to a Dataset Card, perhaps with a short stub of information on what the training data is all about as well as documentation related to data pre-processing or additional filtering. -->
81
+
82
+ [More Information Needed]
83
+
84
+ ### Training Procedure
85
+
86
+ <!-- This relates heavily to the Technical Specifications. Content here should link to that section when it is relevant to the training procedure. -->
87
+
88
+ #### Preprocessing [optional]
89
+
90
+ [More Information Needed]
91
+
92
+
93
+ #### Training Hyperparameters
94
+
95
+ - **Training regime:** [More Information Needed] <!--fp32, fp16 mixed precision, bf16 mixed precision, bf16 non-mixed precision, fp16 non-mixed precision, fp8 mixed precision -->
96
+
97
+ #### Speeds, Sizes, Times [optional]
98
+
99
+ <!-- This section provides information about throughput, start/end time, checkpoint size if relevant, etc. -->
100
+
101
+ [More Information Needed]
102
+
103
+ ## Evaluation
104
+
105
+ <!-- This section describes the evaluation protocols and provides the results. -->
106
+
107
+ ### Testing Data, Factors & Metrics
108
+
109
+ #### Testing Data
110
+
111
+ <!-- This should link to a Dataset Card if possible. -->
112
+
113
+ [More Information Needed]
114
+
115
+ #### Factors
116
+
117
+ <!-- These are the things the evaluation is disaggregating by, e.g., subpopulations or domains. -->
118
+
119
+ [More Information Needed]
120
+
121
+ #### Metrics
122
+
123
+ <!-- These are the evaluation metrics being used, ideally with a description of why. -->
124
+
125
+ [More Information Needed]
126
+
127
+ ### Results
128
+
129
+ [More Information Needed]
130
+
131
+ #### Summary
132
+
133
+
134
+
135
+ ## Model Examination [optional]
136
+
137
+ <!-- Relevant interpretability work for the model goes here -->
138
+
139
+ [More Information Needed]
140
+
141
+ ## Environmental Impact
142
+
143
+ <!-- Total emissions (in grams of CO2eq) and additional considerations, such as electricity usage, go here. Edit the suggested text below accordingly -->
144
+
145
+ Carbon emissions can be estimated using the [Machine Learning Impact calculator](https://mlco2.github.io/impact#compute) presented in [Lacoste et al. (2019)](https://arxiv.org/abs/1910.09700).
146
+
147
+ - **Hardware Type:** [More Information Needed]
148
+ - **Hours used:** [More Information Needed]
149
+ - **Cloud Provider:** [More Information Needed]
150
+ - **Compute Region:** [More Information Needed]
151
+ - **Carbon Emitted:** [More Information Needed]
152
+
153
+ ## Technical Specifications [optional]
154
+
155
+ ### Model Architecture and Objective
156
+
157
+ [More Information Needed]
158
+
159
+ ### Compute Infrastructure
160
+
161
+ [More Information Needed]
162
+
163
+ #### Hardware
164
+
165
+ [More Information Needed]
166
+
167
+ #### Software
168
+
169
+ [More Information Needed]
170
+
171
+ ## Citation [optional]
172
+
173
+ <!-- If there is a paper or blog post introducing the model, the APA and Bibtex information for that should go in this section. -->
174
+
175
+ **BibTeX:**
176
+
177
+ [More Information Needed]
178
+
179
+ **APA:**
180
+
181
+ [More Information Needed]
182
+
183
+ ## Glossary [optional]
184
+
185
+ <!-- If relevant, include terms and calculations in this section that can help readers understand the model or model card. -->
186
+
187
+ [More Information Needed]
188
+
189
+ ## More Information [optional]
190
+
191
+ [More Information Needed]
192
+
193
+ ## Model Card Authors [optional]
194
+
195
+ [More Information Needed]
196
+
197
+ ## Model Card Contact
198
+
199
+ [More Information Needed]
200
+ ### Framework versions
201
+
202
+ - PEFT 0.14.0
checkpoint-140/adapter_config.json ADDED
@@ -0,0 +1,40 @@
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
1
+ {
2
+ "alpha_pattern": {},
3
+ "auto_mapping": null,
4
+ "base_model_name_or_path": "deepseek-ai/DeepSeek-R1-Distill-Llama-70B",
5
+ "bias": "none",
6
+ "eva_config": null,
7
+ "exclude_modules": null,
8
+ "fan_in_fan_out": null,
9
+ "inference_mode": true,
10
+ "init_lora_weights": true,
11
+ "layer_replication": null,
12
+ "layers_pattern": null,
13
+ "layers_to_transform": null,
14
+ "loftq_config": {},
15
+ "lora_alpha": 512,
16
+ "lora_bias": false,
17
+ "lora_dropout": 0.05,
18
+ "megatron_config": null,
19
+ "megatron_core": "megatron.core",
20
+ "modules_to_save": [
21
+ "embed_tokens",
22
+ "lm_head"
23
+ ],
24
+ "peft_type": "LORA",
25
+ "r": 256,
26
+ "rank_pattern": {},
27
+ "revision": null,
28
+ "target_modules": [
29
+ "q_proj",
30
+ "o_proj",
31
+ "down_proj",
32
+ "gate_proj",
33
+ "v_proj",
34
+ "k_proj",
35
+ "up_proj"
36
+ ],
37
+ "task_type": "CAUSAL_LM",
38
+ "use_dora": false,
39
+ "use_rslora": false
40
+ }
checkpoint-140/adapter_model.safetensors ADDED
@@ -0,0 +1,3 @@
 
 
 
 
1
+ version https://git-lfs.github.com/spec/v1
2
+ oid sha256:fb3591d8feb315fa4922eab3c265cec4b288553c3e8e6a112f8335d49c2c7407
3
+ size 10829849744
checkpoint-140/global_step138/bf16_zero_pp_rank_0_mp_rank_00_optim_states.pt ADDED
@@ -0,0 +1,3 @@
 
 
 
 
1
+ version https://git-lfs.github.com/spec/v1
2
+ oid sha256:d79e4da80f0e08b0635d3a9b92ccf9e126d3404cfe85af76913d991cc349c27a
3
+ size 16244554780
checkpoint-140/global_step138/bf16_zero_pp_rank_1_mp_rank_00_optim_states.pt ADDED
@@ -0,0 +1,3 @@
 
 
 
 
1
+ version https://git-lfs.github.com/spec/v1
2
+ oid sha256:ff75d7fa5d713c428900bb32839809a814b7034a4f06d233fc490a43241cb33d
3
+ size 16244593500
checkpoint-140/global_step138/bf16_zero_pp_rank_2_mp_rank_00_optim_states.pt ADDED
@@ -0,0 +1,3 @@
 
 
 
 
1
+ version https://git-lfs.github.com/spec/v1
2
+ oid sha256:6d3a882a6307cd6917128579e6035298a09c8593633248896d5377b823180446
3
+ size 16244594204
checkpoint-140/global_step138/bf16_zero_pp_rank_3_mp_rank_00_optim_states.pt ADDED
@@ -0,0 +1,3 @@
 
 
 
 
1
+ version https://git-lfs.github.com/spec/v1
2
+ oid sha256:70e0202e12f75e0b37392de2a307636fa35e3309feba9841c6243cd039d8ca91
3
+ size 16244554268
checkpoint-140/global_step138/mp_rank_00_model_states.pt ADDED
@@ -0,0 +1,3 @@
 
 
 
 
1
+ version https://git-lfs.github.com/spec/v1
2
+ oid sha256:ffb694071a45203aaa9c078ff65b0dce6c031e524b9ed547e361f22277603ab6
3
+ size 11918643933
checkpoint-140/latest ADDED
@@ -0,0 +1 @@
 
 
1
+ global_step138
checkpoint-140/rng_state_0.pth ADDED
@@ -0,0 +1,3 @@
 
 
 
 
1
+ version https://git-lfs.github.com/spec/v1
2
+ oid sha256:d6cf0d031ea16d1186e8d00e581ec810ec178071afd1cd7cd5449503f613d25d
3
+ size 15024
checkpoint-140/rng_state_1.pth ADDED
@@ -0,0 +1,3 @@
 
 
 
 
1
+ version https://git-lfs.github.com/spec/v1
2
+ oid sha256:70a6c20a6a683c4eee9d41647d74049432b2c874d414604368e0ff9ef2da0d45
3
+ size 15024
checkpoint-140/rng_state_2.pth ADDED
@@ -0,0 +1,3 @@
 
 
 
 
1
+ version https://git-lfs.github.com/spec/v1
2
+ oid sha256:5ef96ee261fd3356adf87507260a418cc9ff9a0a027eef95f547cd55245b4940
3
+ size 15024
checkpoint-140/rng_state_3.pth ADDED
@@ -0,0 +1,3 @@
 
 
 
 
1
+ version https://git-lfs.github.com/spec/v1
2
+ oid sha256:b31e87cb10ffa753df310edf8561f565941df9555c8f4e5d7198574ae74053b5
3
+ size 15024
checkpoint-140/scheduler.pt ADDED
@@ -0,0 +1,3 @@
 
 
 
 
1
+ version https://git-lfs.github.com/spec/v1
2
+ oid sha256:b93fd6b58aedcf2b17639c0e339e842aa4157a5f3e34b1f119941ae4b9872de0
3
+ size 1064
checkpoint-140/special_tokens_map.json ADDED
@@ -0,0 +1,23 @@
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
1
+ {
2
+ "bos_token": {
3
+ "content": "<｜begin▁of▁sentence｜>",
4
+ "lstrip": false,
5
+ "normalized": false,
6
+ "rstrip": false,
7
+ "single_word": false
8
+ },
9
+ "eos_token": {
10
+ "content": "<｜end▁of▁sentence｜>",
11
+ "lstrip": false,
12
+ "normalized": false,
13
+ "rstrip": false,
14
+ "single_word": false
15
+ },
16
+ "pad_token": {
17
+ "content": "<｜end▁of▁sentence｜>",
18
+ "lstrip": false,
19
+ "normalized": false,
20
+ "rstrip": false,
21
+ "single_word": false
22
+ }
23
+ }
checkpoint-140/tokenizer.json ADDED
@@ -0,0 +1,3 @@
 
 
 
 
1
+ version https://git-lfs.github.com/spec/v1
2
+ oid sha256:d91915040cfac999d8c55f4b5bc6e67367c065e3a7a4e4b9438ce1f256addd86
3
+ size 17209530
checkpoint-140/tokenizer_config.json ADDED
@@ -0,0 +1,2067 @@
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
1
+ {
2
+ "add_bos_token": true,
3
+ "add_eos_token": false,
4
+ "add_prefix_space": null,
5
+ "added_tokens_decoder": {
6
+ "128000": {
7
+ "content": "<｜begin▁of▁sentence｜>",
8
+ "lstrip": false,
9
+ "normalized": false,
10
+ "rstrip": false,
11
+ "single_word": false,
12
+ "special": true
13
+ },
14
+ "128001": {
15
+ "content": "<｜end▁of▁sentence｜>",
16
+ "lstrip": false,
17
+ "normalized": false,
18
+ "rstrip": false,
19
+ "single_word": false,
20
+ "special": true
21
+ },
22
+ "128002": {
23
+ "content": "<|reserved_special_token_0|>",
24
+ "lstrip": false,
25
+ "normalized": false,
26
+ "rstrip": false,
27
+ "single_word": false,
28
+ "special": true
29
+ },
30
+ "128003": {
31
+ "content": "<|reserved_special_token_1|>",
32
+ "lstrip": false,
33
+ "normalized": false,
34
+ "rstrip": false,
35
+ "single_word": false,
36
+ "special": true
37
+ },
38
+ "128004": {
39
+ "content": "<|finetune_right_pad_id|>",
40
+ "lstrip": false,
41
+ "normalized": false,
42
+ "rstrip": false,
43
+ "single_word": false,
44
+ "special": true
45
+ },
46
+ "128005": {
47
+ "content": "<|reserved_special_token_2|>",
48
+ "lstrip": false,
49
+ "normalized": false,
50
+ "rstrip": false,
51
+ "single_word": false,
52
+ "special": true
53
+ },
54
+ "128006": {
55
+ "content": "<|start_header_id|>",
56
+ "lstrip": false,
57
+ "normalized": false,
58
+ "rstrip": false,
59
+ "single_word": false,
60
+ "special": true
61
+ },
62
+ "128007": {
63
+ "content": "<|end_header_id|>",
64
+ "lstrip": false,
65
+ "normalized": false,
66
+ "rstrip": false,
67
+ "single_word": false,
68
+ "special": true
69
+ },
70
+ "128008": {
71
+ "content": "<|eom_id|>",
72
+ "lstrip": false,
73
+ "normalized": false,
74
+ "rstrip": false,
75
+ "single_word": false,
76
+ "special": true
77
+ },
78
+ "128009": {
79
+ "content": "<|eot_id|>",
80
+ "lstrip": false,
81
+ "normalized": false,
82
+ "rstrip": false,
83
+ "single_word": false,
84
+ "special": true
85
+ },
86
+ "128010": {
87
+ "content": "<|python_tag|>",
88
+ "lstrip": false,
89
+ "normalized": false,
90
+ "rstrip": false,
91
+ "single_word": false,
92
+ "special": true
93
+ },
94
+ "128011": {
95
+ "content": "<｜User｜>",
96
+ "lstrip": false,
97
+ "normalized": false,
98
+ "rstrip": false,
99
+ "single_word": false,
100
+ "special": false
101
+ },
102
+ "128012": {
103
+ "content": "<｜Assistant｜>",
104
+ "lstrip": false,
105
+ "normalized": false,
106
+ "rstrip": false,
107
+ "single_word": false,
108
+ "special": false
109
+ },
110
+ "128013": {
111
+ "content": "<think>",
112
+ "lstrip": false,
113
+ "normalized": false,
114
+ "rstrip": false,
115
+ "single_word": false,
116
+ "special": false
117
+ },
118
+ "128014": {
119
+ "content": "</think>",
120
+ "lstrip": false,
121
+ "normalized": false,
122
+ "rstrip": false,
123
+ "single_word": false,
124
+ "special": false
125
+ },
126
+ "128015": {
127
+ "content": "<｜▁pad▁｜>",
128
+ "lstrip": false,
129
+ "normalized": false,
130
+ "rstrip": false,
131
+ "single_word": false,
132
+ "special": true
133
+ },
134
+ "128016": {
135
+ "content": "<|reserved_special_token_8|>",
136
+ "lstrip": false,
137
+ "normalized": false,
138
+ "rstrip": false,
139
+ "single_word": false,
140
+ "special": true
141
+ },
142
+ "128017": {
143
+ "content": "<|reserved_special_token_9|>",
144
+ "lstrip": false,
145
+ "normalized": false,
146
+ "rstrip": false,
147
+ "single_word": false,
148
+ "special": true
149
+ },
150
+ "128018": {
151
+ "content": "<|reserved_special_token_10|>",
152
+ "lstrip": false,
153
+ "normalized": false,
154
+ "rstrip": false,
155
+ "single_word": false,
156
+ "special": true
157
+ },
158
+ "128019": {
159
+ "content": "<|reserved_special_token_11|>",
160
+ "lstrip": false,
161
+ "normalized": false,
162
+ "rstrip": false,
163
+ "single_word": false,
164
+ "special": true
165
+ },
166
+ "128020": {
167
+ "content": "<|reserved_special_token_12|>",
168
+ "lstrip": false,
169
+ "normalized": false,
170
+ "rstrip": false,
171
+ "single_word": false,
172
+ "special": true
173
+ },
174
+ "128021": {
175
+ "content": "<|reserved_special_token_13|>",
176
+ "lstrip": false,
177
+ "normalized": false,
178
+ "rstrip": false,
179
+ "single_word": false,
180
+ "special": true
181
+ },
182
+ "128022": {
183
+ "content": "<|reserved_special_token_14|>",
184
+ "lstrip": false,
185
+ "normalized": false,
186
+ "rstrip": false,
187
+ "single_word": false,
188
+ "special": true
189
+ },
190
+ "128023": {
191
+ "content": "<|reserved_special_token_15|>",
192
+ "lstrip": false,
193
+ "normalized": false,
194
+ "rstrip": false,
195
+ "single_word": false,
196
+ "special": true
197
+ },
198
+ "128024": {
199
+ "content": "<|reserved_special_token_16|>",
200
+ "lstrip": false,
201
+ "normalized": false,
202
+ "rstrip": false,
203
+ "single_word": false,
204
+ "special": true
205
+ },
206
+ "128025": {
207
+ "content": "<|reserved_special_token_17|>",
208
+ "lstrip": false,
209
+ "normalized": false,
210
+ "rstrip": false,
211
+ "single_word": false,
212
+ "special": true
213
+ },
214
+ "128026": {
215
+ "content": "<|reserved_special_token_18|>",
216
+ "lstrip": false,
217
+ "normalized": false,
218
+ "rstrip": false,
219
+ "single_word": false,
220
+ "special": true
221
+ },
222
+ "128027": {
223
+ "content": "<|reserved_special_token_19|>",
224
+ "lstrip": false,
225
+ "normalized": false,
226
+ "rstrip": false,
227
+ "single_word": false,
228
+ "special": true
229
+ },
230
+ "128028": {
231
+ "content": "<|reserved_special_token_20|>",
232
+ "lstrip": false,
233
+ "normalized": false,
234
+ "rstrip": false,
235
+ "single_word": false,
236
+ "special": true
237
+ },
238
+ "128029": {
239
+ "content": "<|reserved_special_token_21|>",
240
+ "lstrip": false,
241
+ "normalized": false,
242
+ "rstrip": false,
243
+ "single_word": false,
244
+ "special": true
245
+ },
246
+ "128030": {
247
+ "content": "<|reserved_special_token_22|>",
248
+ "lstrip": false,
249
+ "normalized": false,
250
+ "rstrip": false,
251
+ "single_word": false,
252
+ "special": true
253
+ },
254
+ "128031": {
255
+ "content": "<|reserved_special_token_23|>",
256
+ "lstrip": false,
257
+ "normalized": false,
258
+ "rstrip": false,
259
+ "single_word": false,
260
+ "special": true
261
+ },
262
+ "128032": {
263
+ "content": "<|reserved_special_token_24|>",
264
+ "lstrip": false,
265
+ "normalized": false,
266
+ "rstrip": false,
267
+ "single_word": false,
268
+ "special": true
269
+ },
270
+ "128033": {
271
+ "content": "<|reserved_special_token_25|>",
272
+ "lstrip": false,
273
+ "normalized": false,
274
+ "rstrip": false,
275
+ "single_word": false,
276
+ "special": true
277
+ },
278
+ "128034": {
279
+ "content": "<|reserved_special_token_26|>",
280
+ "lstrip": false,
281
+ "normalized": false,
282
+ "rstrip": false,
283
+ "single_word": false,
284
+ "special": true
285
+ },
286
+ "128035": {
287
+ "content": "<|reserved_special_token_27|>",
288
+ "lstrip": false,
289
+ "normalized": false,
290
+ "rstrip": false,
291
+ "single_word": false,
292
+ "special": true
293
+ },
294
+ "128036": {
295
+ "content": "<|reserved_special_token_28|>",
296
+ "lstrip": false,
297
+ "normalized": false,
298
+ "rstrip": false,
299
+ "single_word": false,
300
+ "special": true
301
+ },
302
+ "128037": {
303
+ "content": "<|reserved_special_token_29|>",
304
+ "lstrip": false,
305
+ "normalized": false,
306
+ "rstrip": false,
307
+ "single_word": false,
308
+ "special": true
309
+ },
310
+ "128038": {
311
+ "content": "<|reserved_special_token_30|>",
312
+ "lstrip": false,
313
+ "normalized": false,
314
+ "rstrip": false,
315
+ "single_word": false,
316
+ "special": true
317
+ },
318
+ "128039": {
319
+ "content": "<|reserved_special_token_31|>",
320
+ "lstrip": false,
321
+ "normalized": false,
322
+ "rstrip": false,
323
+ "single_word": false,
324
+ "special": true
325
+ },
326
+ "128040": {
327
+ "content": "<|reserved_special_token_32|>",
328
+ "lstrip": false,
329
+ "normalized": false,
330
+ "rstrip": false,
331
+ "single_word": false,
332
+ "special": true
333
+ },
334
+ "128041": {
335
+ "content": "<|reserved_special_token_33|>",
336
+ "lstrip": false,
337
+ "normalized": false,
338
+ "rstrip": false,
339
+ "single_word": false,
340
+ "special": true
341
+ },
342
+ "128042": {
343
+ "content": "<|reserved_special_token_34|>",
344
+ "lstrip": false,
345
+ "normalized": false,
346
+ "rstrip": false,
347
+ "single_word": false,
348
+ "special": true
349
+ },
350
+ "128043": {
351
+ "content": "<|reserved_special_token_35|>",
352
+ "lstrip": false,
353
+ "normalized": false,
354
+ "rstrip": false,
355
+ "single_word": false,
356
+ "special": true
357
+ },
358
+ "128044": {
359
+ "content": "<|reserved_special_token_36|>",
360
+ "lstrip": false,
361
+ "normalized": false,
362
+ "rstrip": false,
363
+ "single_word": false,
364
+ "special": true
365
+ },
366
+ "128045": {
367
+ "content": "<|reserved_special_token_37|>",
368
+ "lstrip": false,
369
+ "normalized": false,
370
+ "rstrip": false,
371
+ "single_word": false,
372
+ "special": true
373
+ },
374
+ "128046": {
375
+ "content": "<|reserved_special_token_38|>",
376
+ "lstrip": false,
377
+ "normalized": false,
378
+ "rstrip": false,
379
+ "single_word": false,
380
+ "special": true
381
+ },
382
+ "128047": {
383
+ "content": "<|reserved_special_token_39|>",
384
+ "lstrip": false,
385
+ "normalized": false,
386
+ "rstrip": false,
387
+ "single_word": false,
388
+ "special": true
389
+ },
390
+ "128048": {
391
+ "content": "<|reserved_special_token_40|>",
392
+ "lstrip": false,
393
+ "normalized": false,
394
+ "rstrip": false,
395
+ "single_word": false,
396
+ "special": true
397
+ },
398
+ "128049": {
399
+ "content": "<|reserved_special_token_41|>",
400
+ "lstrip": false,
401
+ "normalized": false,
402
+ "rstrip": false,
403
+ "single_word": false,
404
+ "special": true
405
+ },
406
+ "128050": {
407
+ "content": "<|reserved_special_token_42|>",
408
+ "lstrip": false,
409
+ "normalized": false,
410
+ "rstrip": false,
411
+ "single_word": false,
412
+ "special": true
413
+ },
414
+ "128051": {
415
+ "content": "<|reserved_special_token_43|>",
416
+ "lstrip": false,
417
+ "normalized": false,
418
+ "rstrip": false,
419
+ "single_word": false,
420
+ "special": true
421
+ },
422
+ "128052": {
423
+ "content": "<|reserved_special_token_44|>",
424
+ "lstrip": false,
425
+ "normalized": false,
426
+ "rstrip": false,
427
+ "single_word": false,
428
+ "special": true
429
+ },
430
+ "128053": {
431
+ "content": "<|reserved_special_token_45|>",
432
+ "lstrip": false,
433
+ "normalized": false,
434
+ "rstrip": false,
435
+ "single_word": false,
436
+ "special": true
437
+ },
438
+ "128054": {
439
+ "content": "<|reserved_special_token_46|>",
440
+ "lstrip": false,
441
+ "normalized": false,
442
+ "rstrip": false,
443
+ "single_word": false,
444
+ "special": true
445
+ },
446
+ "128055": {
447
+ "content": "<|reserved_special_token_47|>",
448
+ "lstrip": false,
449
+ "normalized": false,
450
+ "rstrip": false,
451
+ "single_word": false,
452
+ "special": true
453
+ },
454
+ "128056": {
455
+ "content": "<|reserved_special_token_48|>",
456
+ "lstrip": false,
457
+ "normalized": false,
458
+ "rstrip": false,
459
+ "single_word": false,
460
+ "special": true
461
+ },
462
+ "128057": {
463
+ "content": "<|reserved_special_token_49|>",
464
+ "lstrip": false,
465
+ "normalized": false,
466
+ "rstrip": false,
467
+ "single_word": false,
468
+ "special": true
469
+ },
470
+ "128058": {
471
+ "content": "<|reserved_special_token_50|>",
472
+ "lstrip": false,
473
+ "normalized": false,
474
+ "rstrip": false,
475
+ "single_word": false,
476
+ "special": true
477
+ },
478
+ "128059": {
479
+ "content": "<|reserved_special_token_51|>",
480
+ "lstrip": false,
481
+ "normalized": false,
482
+ "rstrip": false,
483
+ "single_word": false,
484
+ "special": true
485
+ },
486
+ "128060": {
487
+ "content": "<|reserved_special_token_52|>",
488
+ "lstrip": false,
489
+ "normalized": false,
490
+ "rstrip": false,
491
+ "single_word": false,
492
+ "special": true
493
+ },
494
+ "128061": {
495
+ "content": "<|reserved_special_token_53|>",
496
+ "lstrip": false,
497
+ "normalized": false,
498
+ "rstrip": false,
499
+ "single_word": false,
500
+ "special": true
501
+ },
502
+ "128062": {
503
+ "content": "<|reserved_special_token_54|>",
504
+ "lstrip": false,
505
+ "normalized": false,
506
+ "rstrip": false,
507
+ "single_word": false,
508
+ "special": true
509
+ },
510
+ "128063": {
511
+ "content": "<|reserved_special_token_55|>",
512
+ "lstrip": false,
513
+ "normalized": false,
514
+ "rstrip": false,
515
+ "single_word": false,
516
+ "special": true
517
+ },
518
+ "128064": {
519
+ "content": "<|reserved_special_token_56|>",
520
+ "lstrip": false,
521
+ "normalized": false,
522
+ "rstrip": false,
523
+ "single_word": false,
524
+ "special": true
525
+ },
526
+ "128065": {
527
+ "content": "<|reserved_special_token_57|>",
528
+ "lstrip": false,
529
+ "normalized": false,
530
+ "rstrip": false,
531
+ "single_word": false,
532
+ "special": true
533
+ },
534
+ "128066": {
535
+ "content": "<|reserved_special_token_58|>",
536
+ "lstrip": false,
537
+ "normalized": false,
538
+ "rstrip": false,
539
+ "single_word": false,
540
+ "special": true
541
+ },
542
+ "128067": {
543
+ "content": "<|reserved_special_token_59|>",
544
+ "lstrip": false,
545
+ "normalized": false,
546
+ "rstrip": false,
547
+ "single_word": false,
548
+ "special": true
549
+ },
550
+ "128068": {
551
+ "content": "<|reserved_special_token_60|>",
552
+ "lstrip": false,
553
+ "normalized": false,
554
+ "rstrip": false,
555
+ "single_word": false,
556
+ "special": true
557
+ },
558
+ "128069": {
559
+ "content": "<|reserved_special_token_61|>",
560
+ "lstrip": false,
561
+ "normalized": false,
562
+ "rstrip": false,
563
+ "single_word": false,
564
+ "special": true
565
+ },
566
+ "128070": {
567
+ "content": "<|reserved_special_token_62|>",
568
+ "lstrip": false,
569
+ "normalized": false,
570
+ "rstrip": false,
571
+ "single_word": false,
572
+ "special": true
573
+ },
574
+ "128071": {
575
+ "content": "<|reserved_special_token_63|>",
576
+ "lstrip": false,
577
+ "normalized": false,
578
+ "rstrip": false,
579
+ "single_word": false,
580
+ "special": true
581
+ },
582
+ "128072": {
583
+ "content": "<|reserved_special_token_64|>",
584
+ "lstrip": false,
585
+ "normalized": false,
586
+ "rstrip": false,
587
+ "single_word": false,
588
+ "special": true
589
+ },
590
+ "128073": {
591
+ "content": "<|reserved_special_token_65|>",
592
+ "lstrip": false,
593
+ "normalized": false,
594
+ "rstrip": false,
595
+ "single_word": false,
596
+ "special": true
597
+ },
598
+ "128074": {
599
+ "content": "<|reserved_special_token_66|>",
600
+ "lstrip": false,
601
+ "normalized": false,
602
+ "rstrip": false,
603
+ "single_word": false,
604
+ "special": true
605
+ },
606
+ "128075": {
607
+ "content": "<|reserved_special_token_67|>",
608
+ "lstrip": false,
609
+ "normalized": false,
610
+ "rstrip": false,
611
+ "single_word": false,
612
+ "special": true
613
+ },
614
+ "128076": {
615
+ "content": "<|reserved_special_token_68|>",
616
+ "lstrip": false,
617
+ "normalized": false,
618
+ "rstrip": false,
619
+ "single_word": false,
620
+ "special": true
621
+ },
622
+ "128077": {
623
+ "content": "<|reserved_special_token_69|>",
624
+ "lstrip": false,
625
+ "normalized": false,
626
+ "rstrip": false,
627
+ "single_word": false,
628
+ "special": true
629
+ },
630
+ "128078": {
631
+ "content": "<|reserved_special_token_70|>",
632
+ "lstrip": false,
633
+ "normalized": false,
634
+ "rstrip": false,
635
+ "single_word": false,
636
+ "special": true
637
+ },
638
+ "128079": {
639
+ "content": "<|reserved_special_token_71|>",
640
+ "lstrip": false,
641
+ "normalized": false,
642
+ "rstrip": false,
643
+ "single_word": false,
644
+ "special": true
645
+ },
646
+ "128080": {
647
+ "content": "<|reserved_special_token_72|>",
648
+ "lstrip": false,
649
+ "normalized": false,
650
+ "rstrip": false,
651
+ "single_word": false,
652
+ "special": true
653
+ },
654
+ "128081": {
655
+ "content": "<|reserved_special_token_73|>",
656
+ "lstrip": false,
657
+ "normalized": false,
658
+ "rstrip": false,
659
+ "single_word": false,
660
+ "special": true
661
+ },
662
+ "128082": {
663
+ "content": "<|reserved_special_token_74|>",
664
+ "lstrip": false,
665
+ "normalized": false,
666
+ "rstrip": false,
667
+ "single_word": false,
668
+ "special": true
669
+ },
670
+ "128083": {
671
+ "content": "<|reserved_special_token_75|>",
672
+ "lstrip": false,
673
+ "normalized": false,
674
+ "rstrip": false,
675
+ "single_word": false,
676
+ "special": true
677
+ },
678
+ "128084": {
679
+ "content": "<|reserved_special_token_76|>",
680
+ "lstrip": false,
681
+ "normalized": false,
682
+ "rstrip": false,
683
+ "single_word": false,
684
+ "special": true
685
+ },
686
+ "128085": {
687
+ "content": "<|reserved_special_token_77|>",
688
+ "lstrip": false,
689
+ "normalized": false,
690
+ "rstrip": false,
691
+ "single_word": false,
692
+ "special": true
693
+ },
694
+ "128086": {
695
+ "content": "<|reserved_special_token_78|>",
696
+ "lstrip": false,
697
+ "normalized": false,
698
+ "rstrip": false,
699
+ "single_word": false,
700
+ "special": true
701
+ },
702
+ "128087": {
703
+ "content": "<|reserved_special_token_79|>",
704
+ "lstrip": false,
705
+ "normalized": false,
706
+ "rstrip": false,
707
+ "single_word": false,
708
+ "special": true
709
+ },
710
+ "128088": {
711
+ "content": "<|reserved_special_token_80|>",
712
+ "lstrip": false,
713
+ "normalized": false,
714
+ "rstrip": false,
715
+ "single_word": false,
716
+ "special": true
717
+ },
718
+ "128089": {
719
+ "content": "<|reserved_special_token_81|>",
720
+ "lstrip": false,
721
+ "normalized": false,
722
+ "rstrip": false,
723
+ "single_word": false,
724
+ "special": true
725
+ },
726
+ "128090": {
727
+ "content": "<|reserved_special_token_82|>",
728
+ "lstrip": false,
729
+ "normalized": false,
730
+ "rstrip": false,
731
+ "single_word": false,
732
+ "special": true
733
+ },
734
+ "128091": {
735
+ "content": "<|reserved_special_token_83|>",
736
+ "lstrip": false,
737
+ "normalized": false,
738
+ "rstrip": false,
739
+ "single_word": false,
740
+ "special": true
741
+ },
742
+ "128092": {
743
+ "content": "<|reserved_special_token_84|>",
744
+ "lstrip": false,
745
+ "normalized": false,
746
+ "rstrip": false,
747
+ "single_word": false,
748
+ "special": true
749
+ },
750
+ "128093": {
751
+ "content": "<|reserved_special_token_85|>",
752
+ "lstrip": false,
753
+ "normalized": false,
754
+ "rstrip": false,
755
+ "single_word": false,
756
+ "special": true
757
+ },
758
+ "128094": {
759
+ "content": "<|reserved_special_token_86|>",
760
+ "lstrip": false,
761
+ "normalized": false,
762
+ "rstrip": false,
763
+ "single_word": false,
764
+ "special": true
765
+ },
766
+ "128095": {
767
+ "content": "<|reserved_special_token_87|>",
768
+ "lstrip": false,
769
+ "normalized": false,
770
+ "rstrip": false,
771
+ "single_word": false,
772
+ "special": true
773
+ },
774
+ "128096": {
775
+ "content": "<|reserved_special_token_88|>",
776
+ "lstrip": false,
777
+ "normalized": false,
778
+ "rstrip": false,
779
+ "single_word": false,
780
+ "special": true
781
+ },
782
+ "128097": {
783
+ "content": "<|reserved_special_token_89|>",
784
+ "lstrip": false,
785
+ "normalized": false,
786
+ "rstrip": false,
787
+ "single_word": false,
788
+ "special": true
789
+ },
790
+ "128098": {
791
+ "content": "<|reserved_special_token_90|>",
792
+ "lstrip": false,
793
+ "normalized": false,
794
+ "rstrip": false,
795
+ "single_word": false,
796
+ "special": true
797
+ },
798
+ "128099": {
799
+ "content": "<|reserved_special_token_91|>",
800
+ "lstrip": false,
801
+ "normalized": false,
802
+ "rstrip": false,
803
+ "single_word": false,
804
+ "special": true
805
+ },
806
+ "128100": {
807
+ "content": "<|reserved_special_token_92|>",
808
+ "lstrip": false,
809
+ "normalized": false,
810
+ "rstrip": false,
811
+ "single_word": false,
812
+ "special": true
813
+ },
814
+ "128101": {
815
+ "content": "<|reserved_special_token_93|>",
816
+ "lstrip": false,
817
+ "normalized": false,
818
+ "rstrip": false,
819
+ "single_word": false,
820
+ "special": true
821
+ },
822
+ "128102": {
823
+ "content": "<|reserved_special_token_94|>",
824
+ "lstrip": false,
825
+ "normalized": false,
826
+ "rstrip": false,
827
+ "single_word": false,
828
+ "special": true
829
+ },
830
+ "128103": {
831
+ "content": "<|reserved_special_token_95|>",
832
+ "lstrip": false,
833
+ "normalized": false,
834
+ "rstrip": false,
835
+ "single_word": false,
836
+ "special": true
837
+ },
838
+ "128104": {
839
+ "content": "<|reserved_special_token_96|>",
840
+ "lstrip": false,
841
+ "normalized": false,
842
+ "rstrip": false,
843
+ "single_word": false,
844
+ "special": true
845
+ },
846
+ "128105": {
847
+ "content": "<|reserved_special_token_97|>",
848
+ "lstrip": false,
849
+ "normalized": false,
850
+ "rstrip": false,
851
+ "single_word": false,
852
+ "special": true
853
+ },
854
+ "128106": {
855
+ "content": "<|reserved_special_token_98|>",
856
+ "lstrip": false,
857
+ "normalized": false,
858
+ "rstrip": false,
859
+ "single_word": false,
860
+ "special": true
861
+ },
862
+ "128107": {
863
+ "content": "<|reserved_special_token_99|>",
864
+ "lstrip": false,
865
+ "normalized": false,
866
+ "rstrip": false,
867
+ "single_word": false,
868
+ "special": true
869
+ },
870
+ "128108": {
871
+ "content": "<|reserved_special_token_100|>",
872
+ "lstrip": false,
873
+ "normalized": false,
874
+ "rstrip": false,
875
+ "single_word": false,
876
+ "special": true
877
+ },
878
+ "128109": {
879
+ "content": "<|reserved_special_token_101|>",
880
+ "lstrip": false,
881
+ "normalized": false,
882
+ "rstrip": false,
883
+ "single_word": false,
884
+ "special": true
885
+ },
886
+ "128110": {
887
+ "content": "<|reserved_special_token_102|>",
888
+ "lstrip": false,
889
+ "normalized": false,
890
+ "rstrip": false,
891
+ "single_word": false,
892
+ "special": true
893
+ },
894
+ "128111": {
895
+ "content": "<|reserved_special_token_103|>",
896
+ "lstrip": false,
897
+ "normalized": false,
898
+ "rstrip": false,
899
+ "single_word": false,
900
+ "special": true
901
+ },
902
+ "128112": {
903
+ "content": "<|reserved_special_token_104|>",
904
+ "lstrip": false,
905
+ "normalized": false,
906
+ "rstrip": false,
907
+ "single_word": false,
908
+ "special": true
909
+ },
910
+ "128113": {
911
+ "content": "<|reserved_special_token_105|>",
912
+ "lstrip": false,
913
+ "normalized": false,
914
+ "rstrip": false,
915
+ "single_word": false,
916
+ "special": true
917
+ },
918
+ "128114": {
919
+ "content": "<|reserved_special_token_106|>",
920
+ "lstrip": false,
921
+ "normalized": false,
922
+ "rstrip": false,
923
+ "single_word": false,
924
+ "special": true
925
+ },
926
+ "128115": {
927
+ "content": "<|reserved_special_token_107|>",
928
+ "lstrip": false,
929
+ "normalized": false,
930
+ "rstrip": false,
931
+ "single_word": false,
932
+ "special": true
933
+ },
934
+ "128116": {
935
+ "content": "<|reserved_special_token_108|>",
936
+ "lstrip": false,
937
+ "normalized": false,
938
+ "rstrip": false,
939
+ "single_word": false,
940
+ "special": true
941
+ },
942
+ "128117": {
943
+ "content": "<|reserved_special_token_109|>",
944
+ "lstrip": false,
945
+ "normalized": false,
946
+ "rstrip": false,
947
+ "single_word": false,
948
+ "special": true
949
+ },
950
+ "128118": {
951
+ "content": "<|reserved_special_token_110|>",
952
+ "lstrip": false,
953
+ "normalized": false,
954
+ "rstrip": false,
955
+ "single_word": false,
956
+ "special": true
957
+ },
958
+ "128119": {
959
+ "content": "<|reserved_special_token_111|>",
960
+ "lstrip": false,
961
+ "normalized": false,
962
+ "rstrip": false,
963
+ "single_word": false,
964
+ "special": true
965
+ },
966
+ "128120": {
967
+ "content": "<|reserved_special_token_112|>",
968
+ "lstrip": false,
969
+ "normalized": false,
970
+ "rstrip": false,
971
+ "single_word": false,
972
+ "special": true
973
+ },
974
+ "128121": {
975
+ "content": "<|reserved_special_token_113|>",
976
+ "lstrip": false,
977
+ "normalized": false,
978
+ "rstrip": false,
979
+ "single_word": false,
980
+ "special": true
981
+ },
982
+ "128122": {
983
+ "content": "<|reserved_special_token_114|>",
984
+ "lstrip": false,
985
+ "normalized": false,
986
+ "rstrip": false,
987
+ "single_word": false,
988
+ "special": true
989
+ },
990
+ "128123": {
991
+ "content": "<|reserved_special_token_115|>",
992
+ "lstrip": false,
993
+ "normalized": false,
994
+ "rstrip": false,
995
+ "single_word": false,
996
+ "special": true
997
+ },
998
+ "128124": {
999
+ "content": "<|reserved_special_token_116|>",
1000
+ "lstrip": false,
1001
+ "normalized": false,
1002
+ "rstrip": false,
1003
+ "single_word": false,
1004
+ "special": true
1005
+ },
1006
+ "128125": {
1007
+ "content": "<|reserved_special_token_117|>",
1008
+ "lstrip": false,
1009
+ "normalized": false,
1010
+ "rstrip": false,
1011
+ "single_word": false,
1012
+ "special": true
1013
+ },
1014
+ "128126": {
1015
+ "content": "<|reserved_special_token_118|>",
1016
+ "lstrip": false,
1017
+ "normalized": false,
1018
+ "rstrip": false,
1019
+ "single_word": false,
1020
+ "special": true
1021
+ },
1022
+ "128127": {
1023
+ "content": "<|reserved_special_token_119|>",
1024
+ "lstrip": false,
1025
+ "normalized": false,
1026
+ "rstrip": false,
1027
+ "single_word": false,
1028
+ "special": true
1029
+ },
1030
+ "128128": {
1031
+ "content": "<|reserved_special_token_120|>",
1032
+ "lstrip": false,
1033
+ "normalized": false,
1034
+ "rstrip": false,
1035
+ "single_word": false,
1036
+ "special": true
1037
+ },
1038
+ "128129": {
1039
+ "content": "<|reserved_special_token_121|>",
1040
+ "lstrip": false,
1041
+ "normalized": false,
1042
+ "rstrip": false,
1043
+ "single_word": false,
1044
+ "special": true
1045
+ },
1046
+ "128130": {
1047
+ "content": "<|reserved_special_token_122|>",
1048
+ "lstrip": false,
1049
+ "normalized": false,
1050
+ "rstrip": false,
1051
+ "single_word": false,
1052
+ "special": true
1053
+ },
1054
+ "128131": {
1055
+ "content": "<|reserved_special_token_123|>",
1056
+ "lstrip": false,
1057
+ "normalized": false,
1058
+ "rstrip": false,
1059
+ "single_word": false,
1060
+ "special": true
1061
+ },
1062
+ "128132": {
1063
+ "content": "<|reserved_special_token_124|>",
1064
+ "lstrip": false,
1065
+ "normalized": false,
1066
+ "rstrip": false,
1067
+ "single_word": false,
1068
+ "special": true
1069
+ },
1070
+ "128133": {
1071
+ "content": "<|reserved_special_token_125|>",
1072
+ "lstrip": false,
1073
+ "normalized": false,
1074
+ "rstrip": false,
1075
+ "single_word": false,
1076
+ "special": true
1077
+ },
1078
+ "128134": {
1079
+ "content": "<|reserved_special_token_126|>",
1080
+ "lstrip": false,
1081
+ "normalized": false,
1082
+ "rstrip": false,
1083
+ "single_word": false,
1084
+ "special": true
1085
+ },
1086
+ "128135": {
1087
+ "content": "<|reserved_special_token_127|>",
1088
+ "lstrip": false,
1089
+ "normalized": false,
1090
+ "rstrip": false,
1091
+ "single_word": false,
1092
+ "special": true
1093
+ },
1094
+ "128136": {
1095
+ "content": "<|reserved_special_token_128|>",
1096
+ "lstrip": false,
1097
+ "normalized": false,
1098
+ "rstrip": false,
1099
+ "single_word": false,
1100
+ "special": true
1101
+ },
1102
+ "128137": {
1103
+ "content": "<|reserved_special_token_129|>",
1104
+ "lstrip": false,
1105
+ "normalized": false,
1106
+ "rstrip": false,
1107
+ "single_word": false,
1108
+ "special": true
1109
+ },
1110
+ "128138": {
1111
+ "content": "<|reserved_special_token_130|>",
1112
+ "lstrip": false,
1113
+ "normalized": false,
1114
+ "rstrip": false,
1115
+ "single_word": false,
1116
+ "special": true
1117
+ },
1118
+ "128139": {
1119
+ "content": "<|reserved_special_token_131|>",
1120
+ "lstrip": false,
1121
+ "normalized": false,
1122
+ "rstrip": false,
1123
+ "single_word": false,
1124
+ "special": true
1125
+ },
1126
+ "128140": {
1127
+ "content": "<|reserved_special_token_132|>",
1128
+ "lstrip": false,
1129
+ "normalized": false,
1130
+ "rstrip": false,
1131
+ "single_word": false,
1132
+ "special": true
1133
+ },
1134
+ "128141": {
1135
+ "content": "<|reserved_special_token_133|>",
1136
+ "lstrip": false,
1137
+ "normalized": false,
1138
+ "rstrip": false,
1139
+ "single_word": false,
1140
+ "special": true
1141
+ },
1142
+ "128142": {
1143
+ "content": "<|reserved_special_token_134|>",
1144
+ "lstrip": false,
1145
+ "normalized": false,
1146
+ "rstrip": false,
1147
+ "single_word": false,
1148
+ "special": true
1149
+ },
1150
+ "128143": {
1151
+ "content": "<|reserved_special_token_135|>",
1152
+ "lstrip": false,
1153
+ "normalized": false,
1154
+ "rstrip": false,
1155
+ "single_word": false,
1156
+ "special": true
1157
+ },
1158
+ "128144": {
1159
+ "content": "<|reserved_special_token_136|>",
1160
+ "lstrip": false,
1161
+ "normalized": false,
1162
+ "rstrip": false,
1163
+ "single_word": false,
1164
+ "special": true
1165
+ },
1166
+ "128145": {
1167
+ "content": "<|reserved_special_token_137|>",
1168
+ "lstrip": false,
1169
+ "normalized": false,
1170
+ "rstrip": false,
1171
+ "single_word": false,
1172
+ "special": true
1173
+ },
1174
+ "128146": {
1175
+ "content": "<|reserved_special_token_138|>",
1176
+ "lstrip": false,
1177
+ "normalized": false,
1178
+ "rstrip": false,
1179
+ "single_word": false,
1180
+ "special": true
1181
+ },
1182
+ "128147": {
1183
+ "content": "<|reserved_special_token_139|>",
1184
+ "lstrip": false,
1185
+ "normalized": false,
1186
+ "rstrip": false,
1187
+ "single_word": false,
1188
+ "special": true
1189
+ },
1190
+ "128148": {
1191
+ "content": "<|reserved_special_token_140|>",
1192
+ "lstrip": false,
1193
+ "normalized": false,
1194
+ "rstrip": false,
1195
+ "single_word": false,
1196
+ "special": true
1197
+ },
1198
+ "128149": {
1199
+ "content": "<|reserved_special_token_141|>",
1200
+ "lstrip": false,
1201
+ "normalized": false,
1202
+ "rstrip": false,
1203
+ "single_word": false,
1204
+ "special": true
1205
+ },
1206
+ "128150": {
1207
+ "content": "<|reserved_special_token_142|>",
1208
+ "lstrip": false,
1209
+ "normalized": false,
1210
+ "rstrip": false,
1211
+ "single_word": false,
1212
+ "special": true
1213
+ },
1214
+ "128151": {
1215
+ "content": "<|reserved_special_token_143|>",
1216
+ "lstrip": false,
1217
+ "normalized": false,
1218
+ "rstrip": false,
1219
+ "single_word": false,
1220
+ "special": true
1221
+ },
1222
+ "128152": {
1223
+ "content": "<|reserved_special_token_144|>",
1224
+ "lstrip": false,
1225
+ "normalized": false,
1226
+ "rstrip": false,
1227
+ "single_word": false,
1228
+ "special": true
1229
+ },
1230
+ "128153": {
1231
+ "content": "<|reserved_special_token_145|>",
1232
+ "lstrip": false,
1233
+ "normalized": false,
1234
+ "rstrip": false,
1235
+ "single_word": false,
1236
+ "special": true
1237
+ },
1238
+ "128154": {
1239
+ "content": "<|reserved_special_token_146|>",
1240
+ "lstrip": false,
1241
+ "normalized": false,
1242
+ "rstrip": false,
1243
+ "single_word": false,
1244
+ "special": true
1245
+ },
1246
+ "128155": {
1247
+ "content": "<|reserved_special_token_147|>",
1248
+ "lstrip": false,
1249
+ "normalized": false,
1250
+ "rstrip": false,
1251
+ "single_word": false,
1252
+ "special": true
1253
+ },
1254
+ "128156": {
1255
+ "content": "<|reserved_special_token_148|>",
1256
+ "lstrip": false,
1257
+ "normalized": false,
1258
+ "rstrip": false,
1259
+ "single_word": false,
1260
+ "special": true
1261
+ },
1262
+ "128157": {
1263
+ "content": "<|reserved_special_token_149|>",
1264
+ "lstrip": false,
1265
+ "normalized": false,
1266
+ "rstrip": false,
1267
+ "single_word": false,
1268
+ "special": true
1269
+ },
1270
+ "128158": {
1271
+ "content": "<|reserved_special_token_150|>",
1272
+ "lstrip": false,
1273
+ "normalized": false,
1274
+ "rstrip": false,
1275
+ "single_word": false,
1276
+ "special": true
1277
+ },
1278
+ "128159": {
1279
+ "content": "<|reserved_special_token_151|>",
1280
+ "lstrip": false,
1281
+ "normalized": false,
1282
+ "rstrip": false,
1283
+ "single_word": false,
1284
+ "special": true
1285
+ },
1286
+ "128160": {
1287
+ "content": "<|reserved_special_token_152|>",
1288
+ "lstrip": false,
1289
+ "normalized": false,
1290
+ "rstrip": false,
1291
+ "single_word": false,
1292
+ "special": true
1293
+ },
1294
+ "128161": {
1295
+ "content": "<|reserved_special_token_153|>",
1296
+ "lstrip": false,
1297
+ "normalized": false,
1298
+ "rstrip": false,
1299
+ "single_word": false,
1300
+ "special": true
1301
+ },
1302
+ "128162": {
1303
+ "content": "<|reserved_special_token_154|>",
1304
+ "lstrip": false,
1305
+ "normalized": false,
1306
+ "rstrip": false,
1307
+ "single_word": false,
1308
+ "special": true
1309
+ },
1310
+ "128163": {
1311
+ "content": "<|reserved_special_token_155|>",
1312
+ "lstrip": false,
1313
+ "normalized": false,
1314
+ "rstrip": false,
1315
+ "single_word": false,
1316
+ "special": true
1317
+ },
1318
+ "128164": {
1319
+ "content": "<|reserved_special_token_156|>",
1320
+ "lstrip": false,
1321
+ "normalized": false,
1322
+ "rstrip": false,
1323
+ "single_word": false,
1324
+ "special": true
1325
+ },
1326
+ "128165": {
1327
+ "content": "<|reserved_special_token_157|>",
1328
+ "lstrip": false,
1329
+ "normalized": false,
1330
+ "rstrip": false,
1331
+ "single_word": false,
1332
+ "special": true
1333
+ },
1334
+ "128166": {
1335
+ "content": "<|reserved_special_token_158|>",
1336
+ "lstrip": false,
1337
+ "normalized": false,
1338
+ "rstrip": false,
1339
+ "single_word": false,
1340
+ "special": true
1341
+ },
1342
+ "128167": {
1343
+ "content": "<|reserved_special_token_159|>",
1344
+ "lstrip": false,
1345
+ "normalized": false,
1346
+ "rstrip": false,
1347
+ "single_word": false,
1348
+ "special": true
1349
+ },
1350
+ "128168": {
1351
+ "content": "<|reserved_special_token_160|>",
1352
+ "lstrip": false,
1353
+ "normalized": false,
1354
+ "rstrip": false,
1355
+ "single_word": false,
1356
+ "special": true
1357
+ },
1358
+ "128169": {
1359
+ "content": "<|reserved_special_token_161|>",
1360
+ "lstrip": false,
1361
+ "normalized": false,
1362
+ "rstrip": false,
1363
+ "single_word": false,
1364
+ "special": true
1365
+ },
1366
+ "128170": {
1367
+ "content": "<|reserved_special_token_162|>",
1368
+ "lstrip": false,
1369
+ "normalized": false,
1370
+ "rstrip": false,
1371
+ "single_word": false,
1372
+ "special": true
1373
+ },
1374
+ "128171": {
1375
+ "content": "<|reserved_special_token_163|>",
1376
+ "lstrip": false,
1377
+ "normalized": false,
1378
+ "rstrip": false,
1379
+ "single_word": false,
1380
+ "special": true
1381
+ },
1382
+ "128172": {
1383
+ "content": "<|reserved_special_token_164|>",
1384
+ "lstrip": false,
1385
+ "normalized": false,
1386
+ "rstrip": false,
1387
+ "single_word": false,
1388
+ "special": true
1389
+ },
1390
+ "128173": {
1391
+ "content": "<|reserved_special_token_165|>",
1392
+ "lstrip": false,
1393
+ "normalized": false,
1394
+ "rstrip": false,
1395
+ "single_word": false,
1396
+ "special": true
1397
+ },
1398
+ "128174": {
1399
+ "content": "<|reserved_special_token_166|>",
1400
+ "lstrip": false,
1401
+ "normalized": false,
1402
+ "rstrip": false,
1403
+ "single_word": false,
1404
+ "special": true
1405
+ },
1406
+ "128175": {
1407
+ "content": "<|reserved_special_token_167|>",
1408
+ "lstrip": false,
1409
+ "normalized": false,
1410
+ "rstrip": false,
1411
+ "single_word": false,
1412
+ "special": true
1413
+ },
1414
+ "128176": {
1415
+ "content": "<|reserved_special_token_168|>",
1416
+ "lstrip": false,
1417
+ "normalized": false,
1418
+ "rstrip": false,
1419
+ "single_word": false,
1420
+ "special": true
1421
+ },
1422
+ "128177": {
1423
+ "content": "<|reserved_special_token_169|>",
1424
+ "lstrip": false,
1425
+ "normalized": false,
1426
+ "rstrip": false,
1427
+ "single_word": false,
1428
+ "special": true
1429
+ },
1430
+ "128178": {
1431
+ "content": "<|reserved_special_token_170|>",
1432
+ "lstrip": false,
1433
+ "normalized": false,
1434
+ "rstrip": false,
1435
+ "single_word": false,
1436
+ "special": true
1437
+ },
1438
+ "128179": {
1439
+ "content": "<|reserved_special_token_171|>",
1440
+ "lstrip": false,
1441
+ "normalized": false,
1442
+ "rstrip": false,
1443
+ "single_word": false,
1444
+ "special": true
1445
+ },
1446
+ "128180": {
1447
+ "content": "<|reserved_special_token_172|>",
1448
+ "lstrip": false,
1449
+ "normalized": false,
1450
+ "rstrip": false,
1451
+ "single_word": false,
1452
+ "special": true
1453
+ },
1454
+ "128181": {
1455
+ "content": "<|reserved_special_token_173|>",
1456
+ "lstrip": false,
1457
+ "normalized": false,
1458
+ "rstrip": false,
1459
+ "single_word": false,
1460
+ "special": true
1461
+ },
1462
+ "128182": {
1463
+ "content": "<|reserved_special_token_174|>",
1464
+ "lstrip": false,
1465
+ "normalized": false,
1466
+ "rstrip": false,
1467
+ "single_word": false,
1468
+ "special": true
1469
+ },
1470
+ "128183": {
1471
+ "content": "<|reserved_special_token_175|>",
1472
+ "lstrip": false,
1473
+ "normalized": false,
1474
+ "rstrip": false,
1475
+ "single_word": false,
1476
+ "special": true
1477
+ },
1478
+ "128184": {
1479
+ "content": "<|reserved_special_token_176|>",
1480
+ "lstrip": false,
1481
+ "normalized": false,
1482
+ "rstrip": false,
1483
+ "single_word": false,
1484
+ "special": true
1485
+ },
1486
+ "128185": {
1487
+ "content": "<|reserved_special_token_177|>",
1488
+ "lstrip": false,
1489
+ "normalized": false,
1490
+ "rstrip": false,
1491
+ "single_word": false,
1492
+ "special": true
1493
+ },
1494
+ "128186": {
1495
+ "content": "<|reserved_special_token_178|>",
1496
+ "lstrip": false,
1497
+ "normalized": false,
1498
+ "rstrip": false,
1499
+ "single_word": false,
1500
+ "special": true
1501
+ },
1502
+ "128187": {
1503
+ "content": "<|reserved_special_token_179|>",
1504
+ "lstrip": false,
1505
+ "normalized": false,
1506
+ "rstrip": false,
1507
+ "single_word": false,
1508
+ "special": true
1509
+ },
1510
+ "128188": {
1511
+ "content": "<|reserved_special_token_180|>",
1512
+ "lstrip": false,
1513
+ "normalized": false,
1514
+ "rstrip": false,
1515
+ "single_word": false,
1516
+ "special": true
1517
+ },
1518
+ "128189": {
1519
+ "content": "<|reserved_special_token_181|>",
1520
+ "lstrip": false,
1521
+ "normalized": false,
1522
+ "rstrip": false,
1523
+ "single_word": false,
1524
+ "special": true
1525
+ },
1526
+ "128190": {
1527
+ "content": "<|reserved_special_token_182|>",
1528
+ "lstrip": false,
1529
+ "normalized": false,
1530
+ "rstrip": false,
1531
+ "single_word": false,
1532
+ "special": true
1533
+ },
1534
+ "128191": {
1535
+ "content": "<|reserved_special_token_183|>",
1536
+ "lstrip": false,
1537
+ "normalized": false,
1538
+ "rstrip": false,
1539
+ "single_word": false,
1540
+ "special": true
1541
+ },
1542
+ "128192": {
1543
+ "content": "<|reserved_special_token_184|>",
1544
+ "lstrip": false,
1545
+ "normalized": false,
1546
+ "rstrip": false,
1547
+ "single_word": false,
1548
+ "special": true
1549
+ },
1550
+ "128193": {
1551
+ "content": "<|reserved_special_token_185|>",
1552
+ "lstrip": false,
1553
+ "normalized": false,
1554
+ "rstrip": false,
1555
+ "single_word": false,
1556
+ "special": true
1557
+ },
1558
+ "128194": {
1559
+ "content": "<|reserved_special_token_186|>",
1560
+ "lstrip": false,
1561
+ "normalized": false,
1562
+ "rstrip": false,
1563
+ "single_word": false,
1564
+ "special": true
1565
+ },
1566
+ "128195": {
1567
+ "content": "<|reserved_special_token_187|>",
1568
+ "lstrip": false,
1569
+ "normalized": false,
1570
+ "rstrip": false,
1571
+ "single_word": false,
1572
+ "special": true
1573
+ },
1574
+ "128196": {
1575
+ "content": "<|reserved_special_token_188|>",
1576
+ "lstrip": false,
1577
+ "normalized": false,
1578
+ "rstrip": false,
1579
+ "single_word": false,
1580
+ "special": true
1581
+ },
1582
+ "128197": {
1583
+ "content": "<|reserved_special_token_189|>",
1584
+ "lstrip": false,
1585
+ "normalized": false,
1586
+ "rstrip": false,
1587
+ "single_word": false,
1588
+ "special": true
1589
+ },
1590
+ "128198": {
1591
+ "content": "<|reserved_special_token_190|>",
1592
+ "lstrip": false,
1593
+ "normalized": false,
1594
+ "rstrip": false,
1595
+ "single_word": false,
1596
+ "special": true
1597
+ },
1598
+ "128199": {
1599
+ "content": "<|reserved_special_token_191|>",
1600
+ "lstrip": false,
1601
+ "normalized": false,
1602
+ "rstrip": false,
1603
+ "single_word": false,
1604
+ "special": true
1605
+ },
1606
+ "128200": {
1607
+ "content": "<|reserved_special_token_192|>",
1608
+ "lstrip": false,
1609
+ "normalized": false,
1610
+ "rstrip": false,
1611
+ "single_word": false,
1612
+ "special": true
1613
+ },
1614
+ "128201": {
1615
+ "content": "<|reserved_special_token_193|>",
1616
+ "lstrip": false,
1617
+ "normalized": false,
1618
+ "rstrip": false,
1619
+ "single_word": false,
1620
+ "special": true
1621
+ },
1622
+ "128202": {
1623
+ "content": "<|reserved_special_token_194|>",
1624
+ "lstrip": false,
1625
+ "normalized": false,
1626
+ "rstrip": false,
1627
+ "single_word": false,
1628
+ "special": true
1629
+ },
1630
+ "128203": {
1631
+ "content": "<|reserved_special_token_195|>",
1632
+ "lstrip": false,
1633
+ "normalized": false,
1634
+ "rstrip": false,
1635
+ "single_word": false,
1636
+ "special": true
1637
+ },
1638
+ "128204": {
1639
+ "content": "<|reserved_special_token_196|>",
1640
+ "lstrip": false,
1641
+ "normalized": false,
1642
+ "rstrip": false,
1643
+ "single_word": false,
1644
+ "special": true
1645
+ },
1646
+ "128205": {
1647
+ "content": "<|reserved_special_token_197|>",
1648
+ "lstrip": false,
1649
+ "normalized": false,
1650
+ "rstrip": false,
1651
+ "single_word": false,
1652
+ "special": true
1653
+ },
1654
+ "128206": {
1655
+ "content": "<|reserved_special_token_198|>",
1656
+ "lstrip": false,
1657
+ "normalized": false,
1658
+ "rstrip": false,
1659
+ "single_word": false,
1660
+ "special": true
1661
+ },
1662
+ "128207": {
1663
+ "content": "<|reserved_special_token_199|>",
1664
+ "lstrip": false,
1665
+ "normalized": false,
1666
+ "rstrip": false,
1667
+ "single_word": false,
1668
+ "special": true
1669
+ },
1670
+ "128208": {
1671
+ "content": "<|reserved_special_token_200|>",
1672
+ "lstrip": false,
1673
+ "normalized": false,
1674
+ "rstrip": false,
1675
+ "single_word": false,
1676
+ "special": true
1677
+ },
1678
+ "128209": {
1679
+ "content": "<|reserved_special_token_201|>",
1680
+ "lstrip": false,
1681
+ "normalized": false,
1682
+ "rstrip": false,
1683
+ "single_word": false,
1684
+ "special": true
1685
+ },
1686
+ "128210": {
1687
+ "content": "<|reserved_special_token_202|>",
1688
+ "lstrip": false,
1689
+ "normalized": false,
1690
+ "rstrip": false,
1691
+ "single_word": false,
1692
+ "special": true
1693
+ },
1694
+ "128211": {
1695
+ "content": "<|reserved_special_token_203|>",
1696
+ "lstrip": false,
1697
+ "normalized": false,
1698
+ "rstrip": false,
1699
+ "single_word": false,
1700
+ "special": true
1701
+ },
1702
+ "128212": {
1703
+ "content": "<|reserved_special_token_204|>",
1704
+ "lstrip": false,
1705
+ "normalized": false,
1706
+ "rstrip": false,
1707
+ "single_word": false,
1708
+ "special": true
1709
+ },
1710
+ "128213": {
1711
+ "content": "<|reserved_special_token_205|>",
1712
+ "lstrip": false,
1713
+ "normalized": false,
1714
+ "rstrip": false,
1715
+ "single_word": false,
1716
+ "special": true
1717
+ },
1718
+ "128214": {
1719
+ "content": "<|reserved_special_token_206|>",
1720
+ "lstrip": false,
1721
+ "normalized": false,
1722
+ "rstrip": false,
1723
+ "single_word": false,
1724
+ "special": true
1725
+ },
1726
+ "128215": {
1727
+ "content": "<|reserved_special_token_207|>",
1728
+ "lstrip": false,
1729
+ "normalized": false,
1730
+ "rstrip": false,
1731
+ "single_word": false,
1732
+ "special": true
1733
+ },
1734
+ "128216": {
1735
+ "content": "<|reserved_special_token_208|>",
1736
+ "lstrip": false,
1737
+ "normalized": false,
1738
+ "rstrip": false,
1739
+ "single_word": false,
1740
+ "special": true
1741
+ },
1742
+ "128217": {
1743
+ "content": "<|reserved_special_token_209|>",
1744
+ "lstrip": false,
1745
+ "normalized": false,
1746
+ "rstrip": false,
1747
+ "single_word": false,
1748
+ "special": true
1749
+ },
1750
+ "128218": {
1751
+ "content": "<|reserved_special_token_210|>",
1752
+ "lstrip": false,
1753
+ "normalized": false,
1754
+ "rstrip": false,
1755
+ "single_word": false,
1756
+ "special": true
1757
+ },
1758
+ "128219": {
1759
+ "content": "<|reserved_special_token_211|>",
1760
+ "lstrip": false,
1761
+ "normalized": false,
1762
+ "rstrip": false,
1763
+ "single_word": false,
1764
+ "special": true
1765
+ },
1766
+ "128220": {
1767
+ "content": "<|reserved_special_token_212|>",
1768
+ "lstrip": false,
1769
+ "normalized": false,
1770
+ "rstrip": false,
1771
+ "single_word": false,
1772
+ "special": true
1773
+ },
1774
+ "128221": {
1775
+ "content": "<|reserved_special_token_213|>",
1776
+ "lstrip": false,
1777
+ "normalized": false,
1778
+ "rstrip": false,
1779
+ "single_word": false,
1780
+ "special": true
1781
+ },
1782
+ "128222": {
1783
+ "content": "<|reserved_special_token_214|>",
1784
+ "lstrip": false,
1785
+ "normalized": false,
1786
+ "rstrip": false,
1787
+ "single_word": false,
1788
+ "special": true
1789
+ },
1790
+ "128223": {
1791
+ "content": "<|reserved_special_token_215|>",
1792
+ "lstrip": false,
1793
+ "normalized": false,
1794
+ "rstrip": false,
1795
+ "single_word": false,
1796
+ "special": true
1797
+ },
1798
+ "128224": {
1799
+ "content": "<|reserved_special_token_216|>",
1800
+ "lstrip": false,
1801
+ "normalized": false,
1802
+ "rstrip": false,
1803
+ "single_word": false,
1804
+ "special": true
1805
+ },
1806
+ "128225": {
1807
+ "content": "<|reserved_special_token_217|>",
1808
+ "lstrip": false,
1809
+ "normalized": false,
1810
+ "rstrip": false,
1811
+ "single_word": false,
1812
+ "special": true
1813
+ },
1814
+ "128226": {
1815
+ "content": "<|reserved_special_token_218|>",
1816
+ "lstrip": false,
1817
+ "normalized": false,
1818
+ "rstrip": false,
1819
+ "single_word": false,
1820
+ "special": true
1821
+ },
1822
+ "128227": {
1823
+ "content": "<|reserved_special_token_219|>",
1824
+ "lstrip": false,
1825
+ "normalized": false,
1826
+ "rstrip": false,
1827
+ "single_word": false,
1828
+ "special": true
1829
+ },
1830
+ "128228": {
1831
+ "content": "<|reserved_special_token_220|>",
1832
+ "lstrip": false,
1833
+ "normalized": false,
1834
+ "rstrip": false,
1835
+ "single_word": false,
1836
+ "special": true
1837
+ },
1838
+ "128229": {
1839
+ "content": "<|reserved_special_token_221|>",
1840
+ "lstrip": false,
1841
+ "normalized": false,
1842
+ "rstrip": false,
1843
+ "single_word": false,
1844
+ "special": true
1845
+ },
1846
+ "128230": {
1847
+ "content": "<|reserved_special_token_222|>",
1848
+ "lstrip": false,
1849
+ "normalized": false,
1850
+ "rstrip": false,
1851
+ "single_word": false,
1852
+ "special": true
1853
+ },
1854
+ "128231": {
1855
+ "content": "<|reserved_special_token_223|>",
1856
+ "lstrip": false,
1857
+ "normalized": false,
1858
+ "rstrip": false,
1859
+ "single_word": false,
1860
+ "special": true
1861
+ },
1862
+ "128232": {
1863
+ "content": "<|reserved_special_token_224|>",
1864
+ "lstrip": false,
1865
+ "normalized": false,
1866
+ "rstrip": false,
1867
+ "single_word": false,
1868
+ "special": true
1869
+ },
1870
+ "128233": {
1871
+ "content": "<|reserved_special_token_225|>",
1872
+ "lstrip": false,
1873
+ "normalized": false,
1874
+ "rstrip": false,
1875
+ "single_word": false,
1876
+ "special": true
1877
+ },
1878
+ "128234": {
1879
+ "content": "<|reserved_special_token_226|>",
1880
+ "lstrip": false,
1881
+ "normalized": false,
1882
+ "rstrip": false,
1883
+ "single_word": false,
1884
+ "special": true
1885
+ },
1886
+ "128235": {
1887
+ "content": "<|reserved_special_token_227|>",
1888
+ "lstrip": false,
1889
+ "normalized": false,
1890
+ "rstrip": false,
1891
+ "single_word": false,
1892
+ "special": true
1893
+ },
1894
+ "128236": {
1895
+ "content": "<|reserved_special_token_228|>",
1896
+ "lstrip": false,
1897
+ "normalized": false,
1898
+ "rstrip": false,
1899
+ "single_word": false,
1900
+ "special": true
1901
+ },
1902
+ "128237": {
1903
+ "content": "<|reserved_special_token_229|>",
1904
+ "lstrip": false,
1905
+ "normalized": false,
1906
+ "rstrip": false,
1907
+ "single_word": false,
1908
+ "special": true
1909
+ },
1910
+ "128238": {
1911
+ "content": "<|reserved_special_token_230|>",
1912
+ "lstrip": false,
1913
+ "normalized": false,
1914
+ "rstrip": false,
1915
+ "single_word": false,
1916
+ "special": true
1917
+ },
1918
+ "128239": {
1919
+ "content": "<|reserved_special_token_231|>",
1920
+ "lstrip": false,
1921
+ "normalized": false,
1922
+ "rstrip": false,
1923
+ "single_word": false,
1924
+ "special": true
1925
+ },
1926
+ "128240": {
1927
+ "content": "<|reserved_special_token_232|>",
1928
+ "lstrip": false,
1929
+ "normalized": false,
1930
+ "rstrip": false,
1931
+ "single_word": false,
1932
+ "special": true
1933
+ },
1934
+ "128241": {
1935
+ "content": "<|reserved_special_token_233|>",
1936
+ "lstrip": false,
1937
+ "normalized": false,
1938
+ "rstrip": false,
1939
+ "single_word": false,
1940
+ "special": true
1941
+ },
1942
+ "128242": {
1943
+ "content": "<|reserved_special_token_234|>",
1944
+ "lstrip": false,
1945
+ "normalized": false,
1946
+ "rstrip": false,
1947
+ "single_word": false,
1948
+ "special": true
1949
+ },
1950
+ "128243": {
1951
+ "content": "<|reserved_special_token_235|>",
1952
+ "lstrip": false,
1953
+ "normalized": false,
1954
+ "rstrip": false,
1955
+ "single_word": false,
1956
+ "special": true
1957
+ },
1958
+ "128244": {
1959
+ "content": "<|reserved_special_token_236|>",
1960
+ "lstrip": false,
1961
+ "normalized": false,
1962
+ "rstrip": false,
1963
+ "single_word": false,
1964
+ "special": true
1965
+ },
1966
+ "128245": {
1967
+ "content": "<|reserved_special_token_237|>",
1968
+ "lstrip": false,
1969
+ "normalized": false,
1970
+ "rstrip": false,
1971
+ "single_word": false,
1972
+ "special": true
1973
+ },
1974
+ "128246": {
1975
+ "content": "<|reserved_special_token_238|>",
1976
+ "lstrip": false,
1977
+ "normalized": false,
1978
+ "rstrip": false,
1979
+ "single_word": false,
1980
+ "special": true
1981
+ },
1982
+ "128247": {
1983
+ "content": "<|reserved_special_token_239|>",
1984
+ "lstrip": false,
1985
+ "normalized": false,
1986
+ "rstrip": false,
1987
+ "single_word": false,
1988
+ "special": true
1989
+ },
1990
+ "128248": {
1991
+ "content": "<|reserved_special_token_240|>",
1992
+ "lstrip": false,
1993
+ "normalized": false,
1994
+ "rstrip": false,
1995
+ "single_word": false,
1996
+ "special": true
1997
+ },
1998
+ "128249": {
1999
+ "content": "<|reserved_special_token_241|>",
2000
+ "lstrip": false,
2001
+ "normalized": false,
2002
+ "rstrip": false,
2003
+ "single_word": false,
2004
+ "special": true
2005
+ },
2006
+ "128250": {
2007
+ "content": "<|reserved_special_token_242|>",
2008
+ "lstrip": false,
2009
+ "normalized": false,
2010
+ "rstrip": false,
2011
+ "single_word": false,
2012
+ "special": true
2013
+ },
2014
+ "128251": {
2015
+ "content": "<|reserved_special_token_243|>",
2016
+ "lstrip": false,
2017
+ "normalized": false,
2018
+ "rstrip": false,
2019
+ "single_word": false,
2020
+ "special": true
2021
+ },
2022
+ "128252": {
2023
+ "content": "<|reserved_special_token_244|>",
2024
+ "lstrip": false,
2025
+ "normalized": false,
2026
+ "rstrip": false,
2027
+ "single_word": false,
2028
+ "special": true
2029
+ },
2030
+ "128253": {
2031
+ "content": "<|reserved_special_token_245|>",
2032
+ "lstrip": false,
2033
+ "normalized": false,
2034
+ "rstrip": false,
2035
+ "single_word": false,
2036
+ "special": true
2037
+ },
2038
+ "128254": {
2039
+ "content": "<|reserved_special_token_246|>",
2040
+ "lstrip": false,
2041
+ "normalized": false,
2042
+ "rstrip": false,
2043
+ "single_word": false,
2044
+ "special": true
2045
+ },
2046
+ "128255": {
2047
+ "content": "<|reserved_special_token_247|>",
2048
+ "lstrip": false,
2049
+ "normalized": false,
2050
+ "rstrip": false,
2051
+ "single_word": false,
2052
+ "special": true
2053
+ }
2054
+ },
2055
+ "bos_token": "<｜begin▁of▁sentence｜>",
2056
+ "chat_template": "{% if not add_generation_prompt is defined %}{% set add_generation_prompt = false %}{% endif %}{% set ns = namespace(is_first=false, is_tool=false, is_output_first=true, system_prompt='') %}{%- for message in messages %}{%- if message['role'] == 'system' %}{% set ns.system_prompt = message['content'] %}{%- endif %}{%- endfor %}{{bos_token}}{{ns.system_prompt}}{%- for message in messages %}{%- if message['role'] == 'user' %}{%- set ns.is_tool = false -%}{{'<｜User｜>' + message['content']}}{%- endif %}{%- if message['role'] == 'assistant' and message['content'] is none %}{%- set ns.is_tool = false -%}{%- for tool in message['tool_calls']%}{%- if not ns.is_first %}{{'<｜Assistant｜><｜tool▁calls▁begin｜><｜tool▁call▁begin｜>' + tool['type'] + '<｜tool▁sep｜>' + tool['function']['name'] + '\\n' + '```json' + '\\n' + tool['function']['arguments'] + '\\n' + '```' + '<｜tool▁call▁end｜>'}}{%- set ns.is_first = true -%}{%- else %}{{'\\n' + '<｜tool▁call▁begin｜>' + tool['type'] + '<｜tool▁sep｜>' + tool['function']['name'] + '\\n' + '```json' + '\\n' + tool['function']['arguments'] + '\\n' + '```' + '<｜tool▁call▁end｜>'}}{{'<｜tool▁calls▁end｜><｜end▁of▁sentence｜>'}}{%- endif %}{%- endfor %}{%- endif %}{%- if message['role'] == 'assistant' and message['content'] is not none %}{%- if ns.is_tool %}{{'<｜tool▁outputs▁end｜>' + message['content'] + '<｜end▁of▁sentence｜>'}}{%- set ns.is_tool = false -%}{%- else %}{% set content = message['content'] %}{% if '</think>' in content %}{% set content = content.split('</think>')[-1] %}{% endif %}{{'<｜Assistant｜>' + content + '<｜end▁of▁sentence｜>'}}{%- endif %}{%- endif %}{%- if message['role'] == 'tool' %}{%- set ns.is_tool = true -%}{%- if ns.is_output_first %}{{'<｜tool▁outputs▁begin｜><｜tool▁output▁begin｜>' + message['content'] + '<｜tool▁output▁end｜>'}}{%- set ns.is_output_first = false %}{%- else %}{{'\\n<｜tool▁output▁begin｜>' + message['content'] + '<｜tool▁output▁end｜>'}}{%- endif %}{%- endif %}{%- endfor -%}{% if ns.is_tool %}{{'<｜tool▁outputs▁end｜>'}}{% endif %}{% if add_generation_prompt and not ns.is_tool %}{{'<｜Assistant｜><think>\\n'}}{% endif %}",
2057
+ "clean_up_tokenization_spaces": false,
2058
+ "eos_token": "<｜end▁of▁sentence｜>",
2059
+ "extra_special_tokens": {},
2060
+ "legacy": true,
2061
+ "model_max_length": 16384,
2062
+ "pad_token": "<｜end▁of▁sentence｜>",
2063
+ "sp_model_kwargs": {},
2064
+ "tokenizer_class": "LlamaTokenizerFast",
2065
+ "unk_token": null,
2066
+ "use_default_system_prompt": false
2067
+ }
checkpoint-140/trainer_state.json ADDED
@@ -0,0 +1,1013 @@
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
1
+ {
2
+ "best_metric": null,
3
+ "best_model_checkpoint": null,
4
+ "epoch": 4.837209302325581,
5
+ "eval_steps": 500,
6
+ "global_step": 140,
7
+ "is_hyper_param_search": false,
8
+ "is_local_process_zero": true,
9
+ "is_world_process_zero": true,
10
+ "log_history": [
11
+ {
12
+ "epoch": 0.03488372093023256,
13
+ "grad_norm": 43.01280212402344,
14
+ "learning_rate": 5.0000000000000004e-08,
15
+ "loss": 2.7496,
16
+ "step": 1
17
+ },
18
+ {
19
+ "epoch": 0.06976744186046512,
20
+ "grad_norm": 44.429630279541016,
21
+ "learning_rate": 1.0000000000000001e-07,
22
+ "loss": 2.7809,
23
+ "step": 2
24
+ },
25
+ {
26
+ "epoch": 0.10465116279069768,
27
+ "grad_norm": 42.754234313964844,
28
+ "learning_rate": 1.5000000000000002e-07,
29
+ "loss": 2.6968,
30
+ "step": 3
31
+ },
32
+ {
33
+ "epoch": 0.13953488372093023,
34
+ "grad_norm": 45.27891159057617,
35
+ "learning_rate": 2.0000000000000002e-07,
36
+ "loss": 2.8076,
37
+ "step": 4
38
+ },
39
+ {
40
+ "epoch": 0.1744186046511628,
41
+ "grad_norm": 43.920860290527344,
42
+ "learning_rate": 2.5000000000000004e-07,
43
+ "loss": 2.7255,
44
+ "step": 5
45
+ },
46
+ {
47
+ "epoch": 0.20930232558139536,
48
+ "grad_norm": 44.61790466308594,
49
+ "learning_rate": 3.0000000000000004e-07,
50
+ "loss": 2.7178,
51
+ "step": 6
52
+ },
53
+ {
54
+ "epoch": 0.2441860465116279,
55
+ "grad_norm": 44.28205871582031,
56
+ "learning_rate": 3.5000000000000004e-07,
57
+ "loss": 2.7193,
58
+ "step": 7
59
+ },
60
+ {
61
+ "epoch": 0.27906976744186046,
62
+ "grad_norm": 45.1516227722168,
63
+ "learning_rate": 4.0000000000000003e-07,
64
+ "loss": 2.6846,
65
+ "step": 8
66
+ },
67
+ {
68
+ "epoch": 0.313953488372093,
69
+ "grad_norm": 46.8723258972168,
70
+ "learning_rate": 4.5000000000000003e-07,
71
+ "loss": 2.67,
72
+ "step": 9
73
+ },
74
+ {
75
+ "epoch": 0.3488372093023256,
76
+ "grad_norm": 48.33848571777344,
77
+ "learning_rate": 5.000000000000001e-07,
78
+ "loss": 2.6368,
79
+ "step": 10
80
+ },
81
+ {
82
+ "epoch": 0.38372093023255816,
83
+ "grad_norm": 46.772193908691406,
84
+ "learning_rate": 5.5e-07,
85
+ "loss": 2.5562,
86
+ "step": 11
87
+ },
88
+ {
89
+ "epoch": 0.4186046511627907,
90
+ "grad_norm": 48.12417984008789,
91
+ "learning_rate": 6.000000000000001e-07,
92
+ "loss": 2.5362,
93
+ "step": 12
94
+ },
95
+ {
96
+ "epoch": 0.45348837209302323,
97
+ "grad_norm": 49.0787239074707,
98
+ "learning_rate": 6.5e-07,
99
+ "loss": 2.3981,
100
+ "step": 13
101
+ },
102
+ {
103
+ "epoch": 0.4883720930232558,
104
+ "grad_norm": 52.367183685302734,
105
+ "learning_rate": 7.000000000000001e-07,
106
+ "loss": 2.3837,
107
+ "step": 14
108
+ },
109
+ {
110
+ "epoch": 0.5232558139534884,
111
+ "grad_norm": 53.57649230957031,
112
+ "learning_rate": 7.5e-07,
113
+ "loss": 2.2885,
114
+ "step": 15
115
+ },
116
+ {
117
+ "epoch": 0.5581395348837209,
118
+ "grad_norm": 56.58591079711914,
119
+ "learning_rate": 8.000000000000001e-07,
120
+ "loss": 2.1447,
121
+ "step": 16
122
+ },
123
+ {
124
+ "epoch": 0.5930232558139535,
125
+ "grad_norm": 55.064735412597656,
126
+ "learning_rate": 8.500000000000001e-07,
127
+ "loss": 2.0242,
128
+ "step": 17
129
+ },
130
+ {
131
+ "epoch": 0.627906976744186,
132
+ "grad_norm": 52.61149597167969,
133
+ "learning_rate": 9.000000000000001e-07,
134
+ "loss": 1.9263,
135
+ "step": 18
136
+ },
137
+ {
138
+ "epoch": 0.6627906976744186,
139
+ "grad_norm": 44.34925079345703,
140
+ "learning_rate": 9.500000000000001e-07,
141
+ "loss": 1.7466,
142
+ "step": 19
143
+ },
144
+ {
145
+ "epoch": 0.6976744186046512,
146
+ "grad_norm": 36.627296447753906,
147
+ "learning_rate": 1.0000000000000002e-06,
148
+ "loss": 1.6084,
149
+ "step": 20
150
+ },
151
+ {
152
+ "epoch": 0.7325581395348837,
153
+ "grad_norm": 30.89563751220703,
154
+ "learning_rate": 1.0500000000000001e-06,
155
+ "loss": 1.4701,
156
+ "step": 21
157
+ },
158
+ {
159
+ "epoch": 0.7674418604651163,
160
+ "grad_norm": 29.17167091369629,
161
+ "learning_rate": 1.1e-06,
162
+ "loss": 1.4218,
163
+ "step": 22
164
+ },
165
+ {
166
+ "epoch": 0.8023255813953488,
167
+ "grad_norm": 28.237022399902344,
168
+ "learning_rate": 1.1500000000000002e-06,
169
+ "loss": 1.3014,
170
+ "step": 23
171
+ },
172
+ {
173
+ "epoch": 0.8372093023255814,
174
+ "grad_norm": 28.778654098510742,
175
+ "learning_rate": 1.2000000000000002e-06,
176
+ "loss": 1.1857,
177
+ "step": 24
178
+ },
179
+ {
180
+ "epoch": 0.872093023255814,
181
+ "grad_norm": 29.368289947509766,
182
+ "learning_rate": 1.25e-06,
183
+ "loss": 1.082,
184
+ "step": 25
185
+ },
186
+ {
187
+ "epoch": 0.9069767441860465,
188
+ "grad_norm": 28.46448516845703,
189
+ "learning_rate": 1.3e-06,
190
+ "loss": 0.8716,
191
+ "step": 26
192
+ },
193
+ {
194
+ "epoch": 0.9418604651162791,
195
+ "grad_norm": 24.584318161010742,
196
+ "learning_rate": 1.3500000000000002e-06,
197
+ "loss": 0.6848,
198
+ "step": 27
199
+ },
200
+ {
201
+ "epoch": 0.9767441860465116,
202
+ "grad_norm": 21.166847229003906,
203
+ "learning_rate": 1.4000000000000001e-06,
204
+ "loss": 0.5438,
205
+ "step": 28
206
+ },
207
+ {
208
+ "epoch": 1.0,
209
+ "grad_norm": 21.166847229003906,
210
+ "learning_rate": 1.45e-06,
211
+ "loss": 0.4084,
212
+ "step": 29
213
+ },
214
+ {
215
+ "epoch": 1.0348837209302326,
216
+ "grad_norm": 24.098440170288086,
217
+ "learning_rate": 1.5e-06,
218
+ "loss": 0.3545,
219
+ "step": 30
220
+ },
221
+ {
222
+ "epoch": 1.069767441860465,
223
+ "grad_norm": 12.6813325881958,
224
+ "learning_rate": 1.5500000000000002e-06,
225
+ "loss": 0.2673,
226
+ "step": 31
227
+ },
228
+ {
229
+ "epoch": 1.1046511627906976,
230
+ "grad_norm": 7.6524128913879395,
231
+ "learning_rate": 1.6000000000000001e-06,
232
+ "loss": 0.2033,
233
+ "step": 32
234
+ },
235
+ {
236
+ "epoch": 1.1395348837209303,
237
+ "grad_norm": 5.23082160949707,
238
+ "learning_rate": 1.6500000000000003e-06,
239
+ "loss": 0.1887,
240
+ "step": 33
241
+ },
242
+ {
243
+ "epoch": 1.1744186046511629,
244
+ "grad_norm": 4.729929447174072,
245
+ "learning_rate": 1.7000000000000002e-06,
246
+ "loss": 0.1683,
247
+ "step": 34
248
+ },
249
+ {
250
+ "epoch": 1.2093023255813953,
251
+ "grad_norm": 5.831579208374023,
252
+ "learning_rate": 1.75e-06,
253
+ "loss": 0.1643,
254
+ "step": 35
255
+ },
256
+ {
257
+ "epoch": 1.244186046511628,
258
+ "grad_norm": 4.030057430267334,
259
+ "learning_rate": 1.8000000000000001e-06,
260
+ "loss": 0.1528,
261
+ "step": 36
262
+ },
263
+ {
264
+ "epoch": 1.2790697674418605,
265
+ "grad_norm": 3.560523509979248,
266
+ "learning_rate": 1.85e-06,
267
+ "loss": 0.1422,
268
+ "step": 37
269
+ },
270
+ {
271
+ "epoch": 1.3139534883720931,
272
+ "grad_norm": 3.3749780654907227,
273
+ "learning_rate": 1.9000000000000002e-06,
274
+ "loss": 0.1327,
275
+ "step": 38
276
+ },
277
+ {
278
+ "epoch": 1.3488372093023255,
279
+ "grad_norm": 2.7184131145477295,
280
+ "learning_rate": 1.9500000000000004e-06,
281
+ "loss": 0.1199,
282
+ "step": 39
283
+ },
284
+ {
285
+ "epoch": 1.3837209302325582,
286
+ "grad_norm": 2.8681583404541016,
287
+ "learning_rate": 2.0000000000000003e-06,
288
+ "loss": 0.1107,
289
+ "step": 40
290
+ },
291
+ {
292
+ "epoch": 1.4186046511627908,
293
+ "grad_norm": 2.8731987476348877,
294
+ "learning_rate": 2.05e-06,
295
+ "loss": 0.1063,
296
+ "step": 41
297
+ },
298
+ {
299
+ "epoch": 1.4534883720930232,
300
+ "grad_norm": 2.9522054195404053,
301
+ "learning_rate": 2.1000000000000002e-06,
302
+ "loss": 0.0989,
303
+ "step": 42
304
+ },
305
+ {
306
+ "epoch": 1.4883720930232558,
307
+ "grad_norm": 2.689038038253784,
308
+ "learning_rate": 2.15e-06,
309
+ "loss": 0.0949,
310
+ "step": 43
311
+ },
312
+ {
313
+ "epoch": 1.5232558139534884,
314
+ "grad_norm": 2.6328952312469482,
315
+ "learning_rate": 2.2e-06,
316
+ "loss": 0.1006,
317
+ "step": 44
318
+ },
319
+ {
320
+ "epoch": 1.558139534883721,
321
+ "grad_norm": 2.7056033611297607,
322
+ "learning_rate": 2.25e-06,
323
+ "loss": 0.0929,
324
+ "step": 45
325
+ },
326
+ {
327
+ "epoch": 1.5930232558139537,
328
+ "grad_norm": 2.568206310272217,
329
+ "learning_rate": 2.3000000000000004e-06,
330
+ "loss": 0.0881,
331
+ "step": 46
332
+ },
333
+ {
334
+ "epoch": 1.627906976744186,
335
+ "grad_norm": 2.714211940765381,
336
+ "learning_rate": 2.35e-06,
337
+ "loss": 0.096,
338
+ "step": 47
339
+ },
340
+ {
341
+ "epoch": 1.6627906976744184,
342
+ "grad_norm": 2.3650710582733154,
343
+ "learning_rate": 2.4000000000000003e-06,
344
+ "loss": 0.086,
345
+ "step": 48
346
+ },
347
+ {
348
+ "epoch": 1.697674418604651,
349
+ "grad_norm": 2.1763498783111572,
350
+ "learning_rate": 2.4500000000000003e-06,
351
+ "loss": 0.0858,
352
+ "step": 49
353
+ },
354
+ {
355
+ "epoch": 1.7325581395348837,
356
+ "grad_norm": 2.124727487564087,
357
+ "learning_rate": 2.5e-06,
358
+ "loss": 0.0843,
359
+ "step": 50
360
+ },
361
+ {
362
+ "epoch": 1.7674418604651163,
363
+ "grad_norm": 1.641269564628601,
364
+ "learning_rate": 2.55e-06,
365
+ "loss": 0.0803,
366
+ "step": 51
367
+ },
368
+ {
369
+ "epoch": 1.802325581395349,
370
+ "grad_norm": 1.5644842386245728,
371
+ "learning_rate": 2.6e-06,
372
+ "loss": 0.0779,
373
+ "step": 52
374
+ },
375
+ {
376
+ "epoch": 1.8372093023255816,
377
+ "grad_norm": 1.1404681205749512,
378
+ "learning_rate": 2.6500000000000005e-06,
379
+ "loss": 0.0765,
380
+ "step": 53
381
+ },
382
+ {
383
+ "epoch": 1.872093023255814,
384
+ "grad_norm": 0.8386123776435852,
385
+ "learning_rate": 2.7000000000000004e-06,
386
+ "loss": 0.0749,
387
+ "step": 54
388
+ },
389
+ {
390
+ "epoch": 1.9069767441860463,
391
+ "grad_norm": 1.8363338708877563,
392
+ "learning_rate": 2.7500000000000004e-06,
393
+ "loss": 0.0737,
394
+ "step": 55
395
+ },
396
+ {
397
+ "epoch": 1.941860465116279,
398
+ "grad_norm": 1.2105377912521362,
399
+ "learning_rate": 2.8000000000000003e-06,
400
+ "loss": 0.0762,
401
+ "step": 56
402
+ },
403
+ {
404
+ "epoch": 1.9767441860465116,
405
+ "grad_norm": 1.5502218008041382,
406
+ "learning_rate": 2.85e-06,
407
+ "loss": 0.07,
408
+ "step": 57
409
+ },
410
+ {
411
+ "epoch": 2.0,
412
+ "grad_norm": 1.0641449689865112,
413
+ "learning_rate": 2.9e-06,
414
+ "loss": 0.0679,
415
+ "step": 58
416
+ },
417
+ {
418
+ "epoch": 2.0348837209302326,
419
+ "grad_norm": 0.9201306104660034,
420
+ "learning_rate": 2.95e-06,
421
+ "loss": 0.0677,
422
+ "step": 59
423
+ },
424
+ {
425
+ "epoch": 2.0697674418604652,
426
+ "grad_norm": 0.5951386094093323,
427
+ "learning_rate": 3e-06,
428
+ "loss": 0.0654,
429
+ "step": 60
430
+ },
431
+ {
432
+ "epoch": 2.104651162790698,
433
+ "grad_norm": 0.8307608962059021,
434
+ "learning_rate": 3.05e-06,
435
+ "loss": 0.0706,
436
+ "step": 61
437
+ },
438
+ {
439
+ "epoch": 2.13953488372093,
440
+ "grad_norm": 0.7110892534255981,
441
+ "learning_rate": 3.1000000000000004e-06,
442
+ "loss": 0.0665,
443
+ "step": 62
444
+ },
445
+ {
446
+ "epoch": 2.1744186046511627,
447
+ "grad_norm": 0.6766234040260315,
448
+ "learning_rate": 3.1500000000000003e-06,
449
+ "loss": 0.0618,
450
+ "step": 63
451
+ },
452
+ {
453
+ "epoch": 2.2093023255813953,
454
+ "grad_norm": 0.3967410922050476,
455
+ "learning_rate": 3.2000000000000003e-06,
456
+ "loss": 0.0644,
457
+ "step": 64
458
+ },
459
+ {
460
+ "epoch": 2.244186046511628,
461
+ "grad_norm": 0.3713420331478119,
462
+ "learning_rate": 3.2500000000000002e-06,
463
+ "loss": 0.0607,
464
+ "step": 65
465
+ },
466
+ {
467
+ "epoch": 2.2790697674418605,
468
+ "grad_norm": 0.5613359212875366,
469
+ "learning_rate": 3.3000000000000006e-06,
470
+ "loss": 0.0623,
471
+ "step": 66
472
+ },
473
+ {
474
+ "epoch": 2.313953488372093,
475
+ "grad_norm": 0.5458635687828064,
476
+ "learning_rate": 3.3500000000000005e-06,
477
+ "loss": 0.0563,
478
+ "step": 67
479
+ },
480
+ {
481
+ "epoch": 2.3488372093023258,
482
+ "grad_norm": 0.4196176528930664,
483
+ "learning_rate": 3.4000000000000005e-06,
484
+ "loss": 0.0547,
485
+ "step": 68
486
+ },
487
+ {
488
+ "epoch": 2.383720930232558,
489
+ "grad_norm": 0.5757117867469788,
490
+ "learning_rate": 3.45e-06,
491
+ "loss": 0.0551,
492
+ "step": 69
493
+ },
494
+ {
495
+ "epoch": 2.4186046511627906,
496
+ "grad_norm": 0.45777687430381775,
497
+ "learning_rate": 3.5e-06,
498
+ "loss": 0.0575,
499
+ "step": 70
500
+ },
501
+ {
502
+ "epoch": 2.453488372093023,
503
+ "grad_norm": 0.5204553008079529,
504
+ "learning_rate": 3.5500000000000003e-06,
505
+ "loss": 0.057,
506
+ "step": 71
507
+ },
508
+ {
509
+ "epoch": 2.488372093023256,
510
+ "grad_norm": 0.6110821962356567,
511
+ "learning_rate": 3.6000000000000003e-06,
512
+ "loss": 0.0556,
513
+ "step": 72
514
+ },
515
+ {
516
+ "epoch": 2.5232558139534884,
517
+ "grad_norm": 0.45246991515159607,
518
+ "learning_rate": 3.65e-06,
519
+ "loss": 0.0564,
520
+ "step": 73
521
+ },
522
+ {
523
+ "epoch": 2.558139534883721,
524
+ "grad_norm": 0.441976398229599,
525
+ "learning_rate": 3.7e-06,
526
+ "loss": 0.0534,
527
+ "step": 74
528
+ },
529
+ {
530
+ "epoch": 2.5930232558139537,
531
+ "grad_norm": 0.46637651324272156,
532
+ "learning_rate": 3.7500000000000005e-06,
533
+ "loss": 0.0498,
534
+ "step": 75
535
+ },
536
+ {
537
+ "epoch": 2.6279069767441863,
538
+ "grad_norm": 0.482038289308548,
539
+ "learning_rate": 3.8000000000000005e-06,
540
+ "loss": 0.0521,
541
+ "step": 76
542
+ },
543
+ {
544
+ "epoch": 2.6627906976744184,
545
+ "grad_norm": 0.5914385318756104,
546
+ "learning_rate": 3.85e-06,
547
+ "loss": 0.0511,
548
+ "step": 77
549
+ },
550
+ {
551
+ "epoch": 2.697674418604651,
552
+ "grad_norm": 0.34532907605171204,
553
+ "learning_rate": 3.900000000000001e-06,
554
+ "loss": 0.0527,
555
+ "step": 78
556
+ },
557
+ {
558
+ "epoch": 2.7325581395348837,
559
+ "grad_norm": 0.35315006971359253,
560
+ "learning_rate": 3.95e-06,
561
+ "loss": 0.0486,
562
+ "step": 79
563
+ },
564
+ {
565
+ "epoch": 2.7674418604651163,
566
+ "grad_norm": 0.4521324634552002,
567
+ "learning_rate": 4.000000000000001e-06,
568
+ "loss": 0.0485,
569
+ "step": 80
570
+ },
571
+ {
572
+ "epoch": 2.802325581395349,
573
+ "grad_norm": 0.49457868933677673,
574
+ "learning_rate": 4.05e-06,
575
+ "loss": 0.0525,
576
+ "step": 81
577
+ },
578
+ {
579
+ "epoch": 2.8372093023255816,
580
+ "grad_norm": 0.7616601586341858,
581
+ "learning_rate": 4.1e-06,
582
+ "loss": 0.0528,
583
+ "step": 82
584
+ },
585
+ {
586
+ "epoch": 2.8720930232558137,
587
+ "grad_norm": 0.4791123569011688,
588
+ "learning_rate": 4.15e-06,
589
+ "loss": 0.0512,
590
+ "step": 83
591
+ },
592
+ {
593
+ "epoch": 2.9069767441860463,
594
+ "grad_norm": 0.38186997175216675,
595
+ "learning_rate": 4.2000000000000004e-06,
596
+ "loss": 0.0442,
597
+ "step": 84
598
+ },
599
+ {
600
+ "epoch": 2.941860465116279,
601
+ "grad_norm": 0.32648414373397827,
602
+ "learning_rate": 4.25e-06,
603
+ "loss": 0.0432,
604
+ "step": 85
605
+ },
606
+ {
607
+ "epoch": 2.9767441860465116,
608
+ "grad_norm": 0.410888671875,
609
+ "learning_rate": 4.3e-06,
610
+ "loss": 0.042,
611
+ "step": 86
612
+ },
613
+ {
614
+ "epoch": 3.0,
615
+ "grad_norm": 0.4732547104358673,
616
+ "learning_rate": 4.350000000000001e-06,
617
+ "loss": 0.0483,
618
+ "step": 87
619
+ },
620
+ {
621
+ "epoch": 3.0348837209302326,
622
+ "grad_norm": 0.6372231841087341,
623
+ "learning_rate": 4.4e-06,
624
+ "loss": 0.0382,
625
+ "step": 88
626
+ },
627
+ {
628
+ "epoch": 3.0697674418604652,
629
+ "grad_norm": 0.393078088760376,
630
+ "learning_rate": 4.450000000000001e-06,
631
+ "loss": 0.0403,
632
+ "step": 89
633
+ },
634
+ {
635
+ "epoch": 3.104651162790698,
636
+ "grad_norm": 0.4366248846054077,
637
+ "learning_rate": 4.5e-06,
638
+ "loss": 0.0367,
639
+ "step": 90
640
+ },
641
+ {
642
+ "epoch": 3.13953488372093,
643
+ "grad_norm": 0.301724374294281,
644
+ "learning_rate": 4.5500000000000005e-06,
645
+ "loss": 0.0351,
646
+ "step": 91
647
+ },
648
+ {
649
+ "epoch": 3.1744186046511627,
650
+ "grad_norm": 0.37665656208992004,
651
+ "learning_rate": 4.600000000000001e-06,
652
+ "loss": 0.03,
653
+ "step": 92
654
+ },
655
+ {
656
+ "epoch": 3.2093023255813953,
657
+ "grad_norm": 0.581331729888916,
658
+ "learning_rate": 4.65e-06,
659
+ "loss": 0.0316,
660
+ "step": 93
661
+ },
662
+ {
663
+ "epoch": 3.244186046511628,
664
+ "grad_norm": 0.44136878848075867,
665
+ "learning_rate": 4.7e-06,
666
+ "loss": 0.0312,
667
+ "step": 94
668
+ },
669
+ {
670
+ "epoch": 3.2790697674418605,
671
+ "grad_norm": 0.7624006271362305,
672
+ "learning_rate": 4.75e-06,
673
+ "loss": 0.0323,
674
+ "step": 95
675
+ },
676
+ {
677
+ "epoch": 3.313953488372093,
678
+ "grad_norm": 0.35294386744499207,
679
+ "learning_rate": 4.800000000000001e-06,
680
+ "loss": 0.0286,
681
+ "step": 96
682
+ },
683
+ {
684
+ "epoch": 3.3488372093023258,
685
+ "grad_norm": 0.6240035891532898,
686
+ "learning_rate": 4.85e-06,
687
+ "loss": 0.0289,
688
+ "step": 97
689
+ },
690
+ {
691
+ "epoch": 3.383720930232558,
692
+ "grad_norm": 0.40580570697784424,
693
+ "learning_rate": 4.9000000000000005e-06,
694
+ "loss": 0.0259,
695
+ "step": 98
696
+ },
697
+ {
698
+ "epoch": 3.4186046511627906,
699
+ "grad_norm": 0.2971636652946472,
700
+ "learning_rate": 4.95e-06,
701
+ "loss": 0.0251,
702
+ "step": 99
703
+ },
704
+ {
705
+ "epoch": 3.453488372093023,
706
+ "grad_norm": 0.3758476972579956,
707
+ "learning_rate": 5e-06,
708
+ "loss": 0.0244,
709
+ "step": 100
710
+ },
711
+ {
712
+ "epoch": 3.488372093023256,
713
+ "grad_norm": 0.3845921754837036,
714
+ "learning_rate": 4.997332437005932e-06,
715
+ "loss": 0.0242,
716
+ "step": 101
717
+ },
718
+ {
719
+ "epoch": 3.5232558139534884,
720
+ "grad_norm": 0.48894616961479187,
721
+ "learning_rate": 4.989335440737587e-06,
722
+ "loss": 0.0252,
723
+ "step": 102
724
+ },
725
+ {
726
+ "epoch": 3.558139534883721,
727
+ "grad_norm": 0.36844325065612793,
728
+ "learning_rate": 4.976026077188013e-06,
729
+ "loss": 0.0197,
730
+ "step": 103
731
+ },
732
+ {
733
+ "epoch": 3.5930232558139537,
734
+ "grad_norm": 0.2800253629684448,
735
+ "learning_rate": 4.957432749209755e-06,
736
+ "loss": 0.0187,
737
+ "step": 104
738
+ },
739
+ {
740
+ "epoch": 3.6279069767441863,
741
+ "grad_norm": 0.34042924642562866,
742
+ "learning_rate": 4.933595135901733e-06,
743
+ "loss": 0.0182,
744
+ "step": 105
745
+ },
746
+ {
747
+ "epoch": 3.6627906976744184,
748
+ "grad_norm": 0.33217477798461914,
749
+ "learning_rate": 4.904564107932048e-06,
750
+ "loss": 0.0166,
751
+ "step": 106
752
+ },
753
+ {
754
+ "epoch": 3.697674418604651,
755
+ "grad_norm": 0.32867398858070374,
756
+ "learning_rate": 4.870401618977415e-06,
757
+ "loss": 0.017,
758
+ "step": 107
759
+ },
760
+ {
761
+ "epoch": 3.7325581395348837,
762
+ "grad_norm": 0.24199356138706207,
763
+ "learning_rate": 4.83118057351089e-06,
764
+ "loss": 0.0138,
765
+ "step": 108
766
+ },
767
+ {
768
+ "epoch": 3.7674418604651163,
769
+ "grad_norm": 0.3216392695903778,
770
+ "learning_rate": 4.786984671220053e-06,
771
+ "loss": 0.0171,
772
+ "step": 109
773
+ },
774
+ {
775
+ "epoch": 3.802325581395349,
776
+ "grad_norm": 0.3574189841747284,
777
+ "learning_rate": 4.737908228387656e-06,
778
+ "loss": 0.0146,
779
+ "step": 110
780
+ },
781
+ {
782
+ "epoch": 3.8372093023255816,
783
+ "grad_norm": 0.30931738018989563,
784
+ "learning_rate": 4.684055976615924e-06,
785
+ "loss": 0.0128,
786
+ "step": 111
787
+ },
788
+ {
789
+ "epoch": 3.8720930232558137,
790
+ "grad_norm": 0.2074785828590393,
791
+ "learning_rate": 4.625542839324036e-06,
792
+ "loss": 0.0089,
793
+ "step": 112
794
+ },
795
+ {
796
+ "epoch": 3.9069767441860463,
797
+ "grad_norm": 0.28993067145347595,
798
+ "learning_rate": 4.562493686495756e-06,
799
+ "loss": 0.0088,
800
+ "step": 113
801
+ },
802
+ {
803
+ "epoch": 3.941860465116279,
804
+ "grad_norm": 0.21493716537952423,
805
+ "learning_rate": 4.4950430682005995e-06,
806
+ "loss": 0.0092,
807
+ "step": 114
808
+ },
809
+ {
810
+ "epoch": 3.9767441860465116,
811
+ "grad_norm": 0.4238521158695221,
812
+ "learning_rate": 4.423334927457198e-06,
813
+ "loss": 0.0096,
814
+ "step": 115
815
+ },
816
+ {
817
+ "epoch": 4.0,
818
+ "grad_norm": 0.4238521158695221,
819
+ "learning_rate": 4.3475222930516484e-06,
820
+ "loss": 0.0092,
821
+ "step": 116
822
+ },
823
+ {
824
+ "epoch": 4.034883720930233,
825
+ "grad_norm": 0.31130537390708923,
826
+ "learning_rate": 4.267766952966369e-06,
827
+ "loss": 0.0052,
828
+ "step": 117
829
+ },
830
+ {
831
+ "epoch": 4.069767441860465,
832
+ "grad_norm": 0.19592078030109406,
833
+ "learning_rate": 4.184239109116393e-06,
834
+ "loss": 0.0052,
835
+ "step": 118
836
+ },
837
+ {
838
+ "epoch": 4.104651162790698,
839
+ "grad_norm": 0.1717974841594696,
840
+ "learning_rate": 4.097117014129903e-06,
841
+ "loss": 0.0041,
842
+ "step": 119
843
+ },
844
+ {
845
+ "epoch": 4.1395348837209305,
846
+ "grad_norm": 0.15367873013019562,
847
+ "learning_rate": 4.006586590948141e-06,
848
+ "loss": 0.0061,
849
+ "step": 120
850
+ },
851
+ {
852
+ "epoch": 4.174418604651163,
853
+ "grad_norm": 0.19444844126701355,
854
+ "learning_rate": 3.91284103605648e-06,
855
+ "loss": 0.0072,
856
+ "step": 121
857
+ },
858
+ {
859
+ "epoch": 4.209302325581396,
860
+ "grad_norm": 0.22090725600719452,
861
+ "learning_rate": 3.81608040719339e-06,
862
+ "loss": 0.0041,
863
+ "step": 122
864
+ },
865
+ {
866
+ "epoch": 4.2441860465116275,
867
+ "grad_norm": 0.13613103330135345,
868
+ "learning_rate": 3.7165111964171407e-06,
869
+ "loss": 0.0042,
870
+ "step": 123
871
+ },
872
+ {
873
+ "epoch": 4.27906976744186,
874
+ "grad_norm": 0.5199090838432312,
875
+ "learning_rate": 3.6143458894413463e-06,
876
+ "loss": 0.0056,
877
+ "step": 124
878
+ },
879
+ {
880
+ "epoch": 4.313953488372093,
881
+ "grad_norm": 0.2258583903312683,
882
+ "learning_rate": 3.5098025121797375e-06,
883
+ "loss": 0.004,
884
+ "step": 125
885
+ },
886
+ {
887
+ "epoch": 4.348837209302325,
888
+ "grad_norm": 0.2459285408258438,
889
+ "learning_rate": 3.403104165467883e-06,
890
+ "loss": 0.0045,
891
+ "step": 126
892
+ },
893
+ {
894
+ "epoch": 4.383720930232558,
895
+ "grad_norm": 0.18142816424369812,
896
+ "learning_rate": 3.2944785489547544e-06,
897
+ "loss": 0.0043,
898
+ "step": 127
899
+ },
900
+ {
901
+ "epoch": 4.4186046511627906,
902
+ "grad_norm": 0.14586737751960754,
903
+ "learning_rate": 3.184157475180208e-06,
904
+ "loss": 0.0027,
905
+ "step": 128
906
+ },
907
+ {
908
+ "epoch": 4.453488372093023,
909
+ "grad_norm": 0.17918141186237335,
910
+ "learning_rate": 3.0723763748753354e-06,
911
+ "loss": 0.0048,
912
+ "step": 129
913
+ },
914
+ {
915
+ "epoch": 4.488372093023256,
916
+ "grad_norm": 0.23981280624866486,
917
+ "learning_rate": 2.9593737945414264e-06,
918
+ "loss": 0.0048,
919
+ "step": 130
920
+ },
921
+ {
922
+ "epoch": 4.523255813953488,
923
+ "grad_norm": 0.20855571329593658,
924
+ "learning_rate": 2.845390887379706e-06,
925
+ "loss": 0.002,
926
+ "step": 131
927
+ },
928
+ {
929
+ "epoch": 4.558139534883721,
930
+ "grad_norm": 0.15569192171096802,
931
+ "learning_rate": 2.730670898658255e-06,
932
+ "loss": 0.0053,
933
+ "step": 132
934
+ },
935
+ {
936
+ "epoch": 4.593023255813954,
937
+ "grad_norm": 0.17529034614562988,
938
+ "learning_rate": 2.6154586466143495e-06,
939
+ "loss": 0.0037,
940
+ "step": 133
941
+ },
942
+ {
943
+ "epoch": 4.627906976744186,
944
+ "grad_norm": 0.2721562683582306,
945
+ "learning_rate": 2.5e-06,
946
+ "loss": 0.0047,
947
+ "step": 134
948
+ },
949
+ {
950
+ "epoch": 4.662790697674419,
951
+ "grad_norm": 0.15811768174171448,
952
+ "learning_rate": 2.3845413533856517e-06,
953
+ "loss": 0.0031,
954
+ "step": 135
955
+ },
956
+ {
957
+ "epoch": 4.6976744186046515,
958
+ "grad_norm": 0.3956407904624939,
959
+ "learning_rate": 2.269329101341745e-06,
960
+ "loss": 0.0048,
961
+ "step": 136
962
+ },
963
+ {
964
+ "epoch": 4.732558139534884,
965
+ "grad_norm": 1.2278478145599365,
966
+ "learning_rate": 2.1546091126202955e-06,
967
+ "loss": 0.0036,
968
+ "step": 137
969
+ },
970
+ {
971
+ "epoch": 4.767441860465116,
972
+ "grad_norm": 0.16240213811397552,
973
+ "learning_rate": 2.040626205458574e-06,
974
+ "loss": 0.0032,
975
+ "step": 138
976
+ },
977
+ {
978
+ "epoch": 4.8023255813953485,
979
+ "grad_norm": 0.15741980075836182,
980
+ "learning_rate": 1.9276236251246655e-06,
981
+ "loss": 0.0027,
982
+ "step": 139
983
+ },
984
+ {
985
+ "epoch": 4.837209302325581,
986
+ "grad_norm": 0.15352095663547516,
987
+ "learning_rate": 1.8158425248197931e-06,
988
+ "loss": 0.0021,
989
+ "step": 140
990
+ }
991
+ ],
992
+ "logging_steps": 1,
993
+ "max_steps": 168,
994
+ "num_input_tokens_seen": 0,
995
+ "num_train_epochs": 6,
996
+ "save_steps": 28,
997
+ "stateful_callbacks": {
998
+ "TrainerControl": {
999
+ "args": {
1000
+ "should_epoch_stop": false,
1001
+ "should_evaluate": false,
1002
+ "should_log": false,
1003
+ "should_save": true,
1004
+ "should_training_stop": false
1005
+ },
1006
+ "attributes": {}
1007
+ }
1008
+ },
1009
+ "total_flos": 1.1894241464697225e+19,
1010
+ "train_batch_size": 4,
1011
+ "trial_name": null,
1012
+ "trial_params": null
1013
+ }
checkpoint-140/training_args.bin ADDED
@@ -0,0 +1,3 @@
 
 
 
 
1
+ version https://git-lfs.github.com/spec/v1
2
+ oid sha256:cb267fa439e36c44dd8202fc6ee1621eee6c4bd83bf7c45ed193d0b07b59aaca
3
+ size 8056
checkpoint-140/zero_to_fp32.py ADDED
@@ -0,0 +1,604 @@
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
1
+ #!/usr/bin/env python
2
+
3
+ # Copyright (c) Microsoft Corporation.
4
+ # SPDX-License-Identifier: Apache-2.0
5
+
6
+ # DeepSpeed Team
7
+
8
+ # This script extracts fp32 consolidated weights from a zero 1, 2 and 3 DeepSpeed checkpoints. It gets
9
+ # copied into the top level checkpoint dir, so the user can easily do the conversion at any point in
10
+ # the future. Once extracted, the weights don't require DeepSpeed and can be used in any
11
+ # application.
12
+ #
13
+ # example: python zero_to_fp32.py . pytorch_model.bin
14
+
15
+ import argparse
16
+ import torch
17
+ import glob
18
+ import math
19
+ import os
20
+ import re
21
+ from collections import OrderedDict
22
+ from dataclasses import dataclass
23
+
24
+ # while this script doesn't use deepspeed to recover data, since the checkpoints are pickled with
25
+ # DeepSpeed data structures it has to be available in the current python environment.
26
+ from deepspeed.utils import logger
27
+ from deepspeed.checkpoint.constants import (DS_VERSION, OPTIMIZER_STATE_DICT, SINGLE_PARTITION_OF_FP32_GROUPS,
28
+ FP32_FLAT_GROUPS, ZERO_STAGE, PARTITION_COUNT, PARAM_SHAPES, BUFFER_NAMES,
29
+ FROZEN_PARAM_SHAPES, FROZEN_PARAM_FRAGMENTS)
30
+
31
+
32
+ @dataclass
33
+ class zero_model_state:
34
+ buffers: dict()
35
+ param_shapes: dict()
36
+ shared_params: list
37
+ ds_version: int
38
+ frozen_param_shapes: dict()
39
+ frozen_param_fragments: dict()
40
+
41
+
42
+ debug = 0
43
+
44
+ # load to cpu
45
+ device = torch.device('cpu')
46
+
47
+
48
+ def atoi(text):
49
+ return int(text) if text.isdigit() else text
50
+
51
+
52
+ def natural_keys(text):
53
+ '''
54
+ alist.sort(key=natural_keys) sorts in human order
55
+ http://nedbatchelder.com/blog/200712/human_sorting.html
56
+ (See Toothy's implementation in the comments)
57
+ '''
58
+ return [atoi(c) for c in re.split(r'(\d+)', text)]
59
+
60
+
61
+ def get_model_state_file(checkpoint_dir, zero_stage):
62
+ if not os.path.isdir(checkpoint_dir):
63
+ raise FileNotFoundError(f"Directory '{checkpoint_dir}' doesn't exist")
64
+
65
+ # there should be only one file
66
+ if zero_stage <= 2:
67
+ file = os.path.join(checkpoint_dir, "mp_rank_00_model_states.pt")
68
+ elif zero_stage == 3:
69
+ file = os.path.join(checkpoint_dir, "zero_pp_rank_0_mp_rank_00_model_states.pt")
70
+
71
+ if not os.path.exists(file):
72
+ raise FileNotFoundError(f"can't find model states file at '{file}'")
73
+
74
+ return file
75
+
76
+
77
+ def get_checkpoint_files(checkpoint_dir, glob_pattern):
78
+ # XXX: need to test that this simple glob rule works for multi-node setup too
79
+ ckpt_files = sorted(glob.glob(os.path.join(checkpoint_dir, glob_pattern)), key=natural_keys)
80
+
81
+ if len(ckpt_files) == 0:
82
+ raise FileNotFoundError(f"can't find {glob_pattern} files in directory '{checkpoint_dir}'")
83
+
84
+ return ckpt_files
85
+
86
+
87
+ def get_optim_files(checkpoint_dir):
88
+ return get_checkpoint_files(checkpoint_dir, "*_optim_states.pt")
89
+
90
+
91
+ def get_model_state_files(checkpoint_dir):
92
+ return get_checkpoint_files(checkpoint_dir, "*_model_states.pt")
93
+
94
+
95
+ def parse_model_states(files):
96
+ zero_model_states = []
97
+ for file in files:
98
+ state_dict = torch.load(file, map_location=device)
99
+
100
+ if BUFFER_NAMES not in state_dict:
101
+ raise ValueError(f"{file} is not a model state checkpoint")
102
+ buffer_names = state_dict[BUFFER_NAMES]
103
+ if debug:
104
+ print("Found buffers:", buffer_names)
105
+
106
+ # recover just the buffers while restoring them to fp32 if they were saved in fp16
107
+ buffers = {k: v.float() for k, v in state_dict["module"].items() if k in buffer_names}
108
+ param_shapes = state_dict[PARAM_SHAPES]
109
+
110
+ # collect parameters that are included in param_shapes
111
+ param_names = []
112
+ for s in param_shapes:
113
+ for name in s.keys():
114
+ param_names.append(name)
115
+
116
+ # update with frozen parameters
117
+ frozen_param_shapes = state_dict.get(FROZEN_PARAM_SHAPES, None)
118
+ if frozen_param_shapes is not None:
119
+ if debug:
120
+ print(f"Found frozen_param_shapes: {frozen_param_shapes}")
121
+ param_names += list(frozen_param_shapes.keys())
122
+
123
+ # handle shared params
124
+ shared_params = [[k, v] for k, v in state_dict["shared_params"].items()]
125
+
126
+ ds_version = state_dict.get(DS_VERSION, None)
127
+
128
+ frozen_param_fragments = state_dict.get(FROZEN_PARAM_FRAGMENTS, None)
129
+
130
+ z_model_state = zero_model_state(buffers=buffers,
131
+ param_shapes=param_shapes,
132
+ shared_params=shared_params,
133
+ ds_version=ds_version,
134
+ frozen_param_shapes=frozen_param_shapes,
135
+ frozen_param_fragments=frozen_param_fragments)
136
+ zero_model_states.append(z_model_state)
137
+
138
+ return zero_model_states
139
+
140
+
141
+ def parse_optim_states(files, ds_checkpoint_dir):
142
+
143
+ total_files = len(files)
144
+ state_dicts = []
145
+ for f in files:
146
+ state_dict = torch.load(f, map_location=device)
147
+ # immediately discard the potentially huge 2 optimizer states as we only care for fp32 master weights
148
+ # and also handle the case where it was already removed by another helper script
149
+ state_dict["optimizer_state_dict"].pop("optimizer_state_dict", None)
150
+ state_dicts.append(state_dict)
151
+
152
+ if not ZERO_STAGE in state_dicts[0][OPTIMIZER_STATE_DICT]:
153
+ raise ValueError(f"{files[0]} is not a zero checkpoint")
154
+ zero_stage = state_dicts[0][OPTIMIZER_STATE_DICT][ZERO_STAGE]
155
+ world_size = state_dicts[0][OPTIMIZER_STATE_DICT][PARTITION_COUNT]
156
+
157
+ # For ZeRO-2 each param group can have different partition_count as data parallelism for expert
158
+ # parameters can be different from data parallelism for non-expert parameters. So we can just
159
+ # use the max of the partition_count to get the dp world_size.
160
+
161
+ if type(world_size) is list:
162
+ world_size = max(world_size)
163
+
164
+ if world_size != total_files:
165
+ raise ValueError(
166
+ f"Expected {world_size} of '*_optim_states.pt' under '{ds_checkpoint_dir}' but found {total_files} files. "
167
+ "Possibly due to an overwrite of an old checkpoint, or a checkpoint didn't get saved by one or more processes."
168
+ )
169
+
170
+ # the groups are named differently in each stage
171
+ if zero_stage <= 2:
172
+ fp32_groups_key = SINGLE_PARTITION_OF_FP32_GROUPS
173
+ elif zero_stage == 3:
174
+ fp32_groups_key = FP32_FLAT_GROUPS
175
+ else:
176
+ raise ValueError(f"unknown zero stage {zero_stage}")
177
+
178
+ if zero_stage <= 2:
179
+ fp32_flat_groups = [state_dicts[i][OPTIMIZER_STATE_DICT][fp32_groups_key] for i in range(len(state_dicts))]
180
+ elif zero_stage == 3:
181
+ # if there is more than one param group, there will be multiple flattened tensors - one
182
+ # flattened tensor per group - for simplicity merge them into a single tensor
183
+ #
184
+ # XXX: could make the script more memory efficient for when there are multiple groups - it
185
+ # will require matching the sub-lists of param_shapes for each param group flattened tensor
186
+
187
+ fp32_flat_groups = [
188
+ torch.cat(state_dicts[i][OPTIMIZER_STATE_DICT][fp32_groups_key], 0) for i in range(len(state_dicts))
189
+ ]
190
+
191
+ return zero_stage, world_size, fp32_flat_groups
192
+
193
+
194
+ def _get_fp32_state_dict_from_zero_checkpoint(ds_checkpoint_dir, exclude_frozen_parameters):
195
+ """
196
+ Returns fp32 state_dict reconstructed from ds checkpoint
197
+
198
+ Args:
199
+ - ``ds_checkpoint_dir``: path to the deepspeed checkpoint folder (where the optimizer files are)
200
+
201
+ """
202
+ print(f"Processing zero checkpoint '{ds_checkpoint_dir}'")
203
+
204
+ optim_files = get_optim_files(ds_checkpoint_dir)
205
+ zero_stage, world_size, fp32_flat_groups = parse_optim_states(optim_files, ds_checkpoint_dir)
206
+ print(f"Detected checkpoint of type zero stage {zero_stage}, world_size: {world_size}")
207
+
208
+ model_files = get_model_state_files(ds_checkpoint_dir)
209
+
210
+ zero_model_states = parse_model_states(model_files)
211
+ print(f'Parsing checkpoint created by deepspeed=={zero_model_states[0].ds_version}')
212
+
213
+ if zero_stage <= 2:
214
+ return _get_fp32_state_dict_from_zero2_checkpoint(world_size, fp32_flat_groups, zero_model_states,
215
+ exclude_frozen_parameters)
216
+ elif zero_stage == 3:
217
+ return _get_fp32_state_dict_from_zero3_checkpoint(world_size, fp32_flat_groups, zero_model_states,
218
+ exclude_frozen_parameters)
219
+
220
+
221
+ def _zero2_merge_frozen_params(state_dict, zero_model_states):
222
+ if zero_model_states[0].frozen_param_shapes is None or len(zero_model_states[0].frozen_param_shapes) == 0:
223
+ return
224
+
225
+ frozen_param_shapes = zero_model_states[0].frozen_param_shapes
226
+ frozen_param_fragments = zero_model_states[0].frozen_param_fragments
227
+
228
+ if debug:
229
+ num_elem = sum(s.numel() for s in frozen_param_shapes.values())
230
+ print(f'rank 0: {FROZEN_PARAM_SHAPES}.numel = {num_elem}')
231
+
232
+ wanted_params = len(frozen_param_shapes)
233
+ wanted_numel = sum(s.numel() for s in frozen_param_shapes.values())
234
+ avail_numel = sum([p.numel() for p in frozen_param_fragments.values()])
235
+ print(f'Frozen params: Have {avail_numel} numels to process.')
236
+ print(f'Frozen params: Need {wanted_numel} numels in {wanted_params} params')
237
+
238
+ total_params = 0
239
+ total_numel = 0
240
+ for name, shape in frozen_param_shapes.items():
241
+ total_params += 1
242
+ unpartitioned_numel = shape.numel()
243
+ total_numel += unpartitioned_numel
244
+
245
+ state_dict[name] = frozen_param_fragments[name]
246
+
247
+ if debug:
248
+ print(f"{name} full shape: {shape} unpartitioned numel {unpartitioned_numel} ")
249
+
250
+ print(f"Reconstructed Frozen fp32 state dict with {total_params} params {total_numel} elements")
251
+
252
+
253
+ def _has_callable(obj, fn):
254
+ attr = getattr(obj, fn, None)
255
+ return callable(attr)
256
+
257
+
258
+ def _zero2_merge_trainable_params(state_dict, world_size, fp32_flat_groups, zero_model_states):
259
+ param_shapes = zero_model_states[0].param_shapes
260
+
261
+ # Reconstruction protocol:
262
+ #
263
+ # XXX: document this
264
+
265
+ if debug:
266
+ for i in range(world_size):
267
+ for j in range(len(fp32_flat_groups[0])):
268
+ print(f"{FP32_FLAT_GROUPS}[{i}][{j}].shape={fp32_flat_groups[i][j].shape}")
269
+
270
+ # XXX: memory usage doubles here (zero2)
271
+ num_param_groups = len(fp32_flat_groups[0])
272
+ merged_single_partition_of_fp32_groups = []
273
+ for i in range(num_param_groups):
274
+ merged_partitions = [sd[i] for sd in fp32_flat_groups]
275
+ full_single_fp32_vector = torch.cat(merged_partitions, 0)
276
+ merged_single_partition_of_fp32_groups.append(full_single_fp32_vector)
277
+ avail_numel = sum(
278
+ [full_single_fp32_vector.numel() for full_single_fp32_vector in merged_single_partition_of_fp32_groups])
279
+
280
+ if debug:
281
+ wanted_params = sum([len(shapes) for shapes in param_shapes])
282
+ wanted_numel = sum([sum(shape.numel() for shape in shapes.values()) for shapes in param_shapes])
283
+ # not asserting if there is a mismatch due to possible padding
284
+ print(f"Have {avail_numel} numels to process.")
285
+ print(f"Need {wanted_numel} numels in {wanted_params} params.")
286
+
287
+ # params
288
+ # XXX: for huge models that can't fit into the host's RAM we will have to recode this to support
289
+ # out-of-core computing solution
290
+ total_numel = 0
291
+ total_params = 0
292
+ for shapes, full_single_fp32_vector in zip(param_shapes, merged_single_partition_of_fp32_groups):
293
+ offset = 0
294
+ avail_numel = full_single_fp32_vector.numel()
295
+ for name, shape in shapes.items():
296
+
297
+ unpartitioned_numel = shape.numel() if _has_callable(shape, 'numel') else math.prod(shape)
298
+ total_numel += unpartitioned_numel
299
+ total_params += 1
300
+
301
+ if debug:
302
+ print(f"{name} full shape: {shape} unpartitioned numel {unpartitioned_numel} ")
303
+ state_dict[name] = full_single_fp32_vector.narrow(0, offset, unpartitioned_numel).view(shape)
304
+ offset += unpartitioned_numel
305
+
306
+ # Z2 started to align to 2*world_size to improve nccl performance. Therefore both offset and
307
+ # avail_numel can differ by anywhere between 0..2*world_size. Due to two unrelated complex
308
+ # paddings performed in the code it's almost impossible to predict the exact numbers w/o the
309
+ # live optimizer object, so we are checking that the numbers are within the right range
310
+ align_to = 2 * world_size
311
+
312
+ def zero2_align(x):
313
+ return align_to * math.ceil(x / align_to)
314
+
315
+ if debug:
316
+ print(f"original offset={offset}, avail_numel={avail_numel}")
317
+
318
+ offset = zero2_align(offset)
319
+ avail_numel = zero2_align(avail_numel)
320
+
321
+ if debug:
322
+ print(f"aligned offset={offset}, avail_numel={avail_numel}")
323
+
324
+ # Sanity check
325
+ if offset != avail_numel:
326
+ raise ValueError(f"consumed {offset} numels out of {avail_numel} - something is wrong")
327
+
328
+ print(f"Reconstructed fp32 state dict with {total_params} params {total_numel} elements")
329
+
330
+
331
+ def _get_fp32_state_dict_from_zero2_checkpoint(world_size, fp32_flat_groups, zero_model_states,
332
+ exclude_frozen_parameters):
333
+ state_dict = OrderedDict()
334
+
335
+ # buffers
336
+ buffers = zero_model_states[0].buffers
337
+ state_dict.update(buffers)
338
+ if debug:
339
+ print(f"added {len(buffers)} buffers")
340
+
341
+ if not exclude_frozen_parameters:
342
+ _zero2_merge_frozen_params(state_dict, zero_model_states)
343
+
344
+ _zero2_merge_trainable_params(state_dict, world_size, fp32_flat_groups, zero_model_states)
345
+
346
+ # recover shared parameters
347
+ for pair in zero_model_states[0].shared_params:
348
+ if pair[1] in state_dict:
349
+ state_dict[pair[0]] = state_dict[pair[1]]
350
+
351
+ return state_dict
352
+
353
+
354
+ def zero3_partitioned_param_info(unpartitioned_numel, world_size):
355
+ remainder = unpartitioned_numel % world_size
356
+ padding_numel = (world_size - remainder) if remainder else 0
357
+ partitioned_numel = math.ceil(unpartitioned_numel / world_size)
358
+ return partitioned_numel, padding_numel
359
+
360
+
361
+ def _zero3_merge_frozen_params(state_dict, world_size, zero_model_states):
362
+ if zero_model_states[0].frozen_param_shapes is None or len(zero_model_states[0].frozen_param_shapes) == 0:
363
+ return
364
+
365
+ if debug:
366
+ for i in range(world_size):
367
+ num_elem = sum(s.numel() for s in zero_model_states[i].frozen_param_fragments.values())
368
+ print(f'rank {i}: {FROZEN_PARAM_SHAPES}.numel = {num_elem}')
369
+
370
+ frozen_param_shapes = zero_model_states[0].frozen_param_shapes
371
+ wanted_params = len(frozen_param_shapes)
372
+ wanted_numel = sum(s.numel() for s in frozen_param_shapes.values())
373
+ avail_numel = sum([p.numel() for p in zero_model_states[0].frozen_param_fragments.values()]) * world_size
374
+ print(f'Frozen params: Have {avail_numel} numels to process.')
375
+ print(f'Frozen params: Need {wanted_numel} numels in {wanted_params} params')
376
+
377
+ total_params = 0
378
+ total_numel = 0
379
+ for name, shape in zero_model_states[0].frozen_param_shapes.items():
380
+ total_params += 1
381
+ unpartitioned_numel = shape.numel()
382
+ total_numel += unpartitioned_numel
383
+
384
+ param_frags = tuple(model_state.frozen_param_fragments[name] for model_state in zero_model_states)
385
+ state_dict[name] = torch.cat(param_frags, 0).narrow(0, 0, unpartitioned_numel).view(shape)
386
+
387
+ partitioned_numel, partitioned_padding_numel = zero3_partitioned_param_info(unpartitioned_numel, world_size)
388
+
389
+ if debug:
390
+ print(
391
+ f"Frozen params: {total_params} {name} full shape: {shape} partition0 numel={partitioned_numel} partitioned_padding_numel={partitioned_padding_numel}"
392
+ )
393
+
394
+ print(f"Reconstructed Frozen fp32 state dict with {total_params} params {total_numel} elements")
395
+
396
+
397
+ def _zero3_merge_trainable_params(state_dict, world_size, fp32_flat_groups, zero_model_states):
398
+ param_shapes = zero_model_states[0].param_shapes
399
+ avail_numel = fp32_flat_groups[0].numel() * world_size
400
+ # Reconstruction protocol: For zero3 we need to zip the partitions together at boundary of each
401
+ # param, re-consolidating each param, while dealing with padding if any
402
+
403
+ # merge list of dicts, preserving order
404
+ param_shapes = {k: v for d in param_shapes for k, v in d.items()}
405
+
406
+ if debug:
407
+ for i in range(world_size):
408
+ print(f"{FP32_FLAT_GROUPS}[{i}].shape={fp32_flat_groups[i].shape}")
409
+
410
+ wanted_params = len(param_shapes)
411
+ wanted_numel = sum(shape.numel() for shape in param_shapes.values())
412
+ # not asserting if there is a mismatch due to possible padding
413
+ avail_numel = fp32_flat_groups[0].numel() * world_size
414
+ print(f"Trainable params: Have {avail_numel} numels to process.")
415
+ print(f"Trainable params: Need {wanted_numel} numels in {wanted_params} params.")
416
+
417
+ # params
418
+ # XXX: for huge models that can't fit into the host's RAM we will have to recode this to support
419
+ # out-of-core computing solution
420
+ offset = 0
421
+ total_numel = 0
422
+ total_params = 0
423
+ for name, shape in param_shapes.items():
424
+
425
+ unpartitioned_numel = shape.numel()
426
+ total_numel += unpartitioned_numel
427
+ total_params += 1
428
+
429
+ partitioned_numel, partitioned_padding_numel = zero3_partitioned_param_info(unpartitioned_numel, world_size)
430
+
431
+ if debug:
432
+ print(
433
+ f"Trainable params: {total_params} {name} full shape: {shape} partition0 numel={partitioned_numel} partitioned_padding_numel={partitioned_padding_numel}"
434
+ )
435
+
436
+ # XXX: memory usage doubles here
437
+ state_dict[name] = torch.cat(
438
+ tuple(fp32_flat_groups[i].narrow(0, offset, partitioned_numel) for i in range(world_size)),
439
+ 0).narrow(0, 0, unpartitioned_numel).view(shape)
440
+ offset += partitioned_numel
441
+
442
+ offset *= world_size
443
+
444
+ # Sanity check
445
+ if offset != avail_numel:
446
+ raise ValueError(f"consumed {offset} numels out of {avail_numel} - something is wrong")
447
+
448
+ print(f"Reconstructed Trainable fp32 state dict with {total_params} params {total_numel} elements")
449
+
450
+
451
+ def _get_fp32_state_dict_from_zero3_checkpoint(world_size, fp32_flat_groups, zero_model_states,
452
+ exclude_frozen_parameters):
453
+ state_dict = OrderedDict()
454
+
455
+ # buffers
456
+ buffers = zero_model_states[0].buffers
457
+ state_dict.update(buffers)
458
+ if debug:
459
+ print(f"added {len(buffers)} buffers")
460
+
461
+ if not exclude_frozen_parameters:
462
+ _zero3_merge_frozen_params(state_dict, world_size, zero_model_states)
463
+
464
+ _zero3_merge_trainable_params(state_dict, world_size, fp32_flat_groups, zero_model_states)
465
+
466
+ # recover shared parameters
467
+ for pair in zero_model_states[0].shared_params:
468
+ if pair[1] in state_dict:
469
+ state_dict[pair[0]] = state_dict[pair[1]]
470
+
471
+ return state_dict
472
+
473
+
474
+ def get_fp32_state_dict_from_zero_checkpoint(checkpoint_dir, tag=None, exclude_frozen_parameters=False):
475
+ """
476
+ Convert ZeRO 2 or 3 checkpoint into a single fp32 consolidated state_dict that can be loaded with
477
+ ``load_state_dict()`` and used for training without DeepSpeed or shared with others, for example
478
+ via a model hub.
479
+
480
+ Args:
481
+ - ``checkpoint_dir``: path to the desired checkpoint folder
482
+ - ``tag``: checkpoint tag used as a unique identifier for checkpoint. If not provided will attempt to load tag in 'latest' file. e.g., ``global_step14``
483
+ - ``exclude_frozen_parameters``: exclude frozen parameters
484
+
485
+ Returns:
486
+ - pytorch ``state_dict``
487
+
488
+ Note: this approach may not work if your application doesn't have sufficient free CPU memory and
489
+ you may need to use the offline approach using the ``zero_to_fp32.py`` script that is saved with
490
+ the checkpoint.
491
+
492
+ A typical usage might be ::
493
+
494
+ from deepspeed.utils.zero_to_fp32 import get_fp32_state_dict_from_zero_checkpoint
495
+ # do the training and checkpoint saving
496
+ state_dict = get_fp32_state_dict_from_zero_checkpoint(checkpoint_dir) # already on cpu
497
+ model = model.cpu() # move to cpu
498
+ model.load_state_dict(state_dict)
499
+ # submit to model hub or save the model to share with others
500
+
501
+ In this example the ``model`` will no longer be usable in the deepspeed context of the same
502
+ application. i.e. you will need to re-initialize the deepspeed engine, since
503
+ ``model.load_state_dict(state_dict)`` will remove all the deepspeed magic from it.
504
+
505
+ If you want it all done for you, use ``load_state_dict_from_zero_checkpoint`` instead.
506
+
507
+ """
508
+ if tag is None:
509
+ latest_path = os.path.join(checkpoint_dir, 'latest')
510
+ if os.path.isfile(latest_path):
511
+ with open(latest_path, 'r') as fd:
512
+ tag = fd.read().strip()
513
+ else:
514
+ raise ValueError(f"Unable to find 'latest' file at {latest_path}")
515
+
516
+ ds_checkpoint_dir = os.path.join(checkpoint_dir, tag)
517
+
518
+ if not os.path.isdir(ds_checkpoint_dir):
519
+ raise FileNotFoundError(f"Directory '{ds_checkpoint_dir}' doesn't exist")
520
+
521
+ return _get_fp32_state_dict_from_zero_checkpoint(ds_checkpoint_dir, exclude_frozen_parameters)
522
+
523
+
524
+ def convert_zero_checkpoint_to_fp32_state_dict(checkpoint_dir, output_file, tag=None, exclude_frozen_parameters=False):
525
+ """
526
+ Convert ZeRO 2 or 3 checkpoint into a single fp32 consolidated ``state_dict`` file that can be
527
+ loaded with ``torch.load(file)`` + ``load_state_dict()`` and used for training without DeepSpeed.
528
+
529
+ Args:
530
+ - ``checkpoint_dir``: path to the desired checkpoint folder. (one that contains the tag-folder, like ``global_step14``)
531
+ - ``output_file``: path to the pytorch fp32 state_dict output file (e.g. path/pytorch_model.bin)
532
+ - ``tag``: checkpoint tag used as a unique identifier for checkpoint. If not provided will attempt to load tag in the file named ``latest`` in the checkpoint folder, e.g., ``global_step14``
533
+ - ``exclude_frozen_parameters``: exclude frozen parameters
534
+ """
535
+
536
+ state_dict = get_fp32_state_dict_from_zero_checkpoint(checkpoint_dir, tag, exclude_frozen_parameters)
537
+ print(f"Saving fp32 state dict to {output_file}")
538
+ torch.save(state_dict, output_file)
539
+
540
+
541
+ def load_state_dict_from_zero_checkpoint(model, checkpoint_dir, tag=None):
542
+ """
543
+ 1. Put the provided model to cpu
544
+ 2. Convert ZeRO 2 or 3 checkpoint into a single fp32 consolidated ``state_dict``
545
+ 3. Load it into the provided model
546
+
547
+ Args:
548
+ - ``model``: the model object to update
549
+ - ``checkpoint_dir``: path to the desired checkpoint folder. (one that contains the tag-folder, like ``global_step14``)
550
+ - ``tag``: checkpoint tag used as a unique identifier for checkpoint. If not provided will attempt to load tag in the file named ``latest`` in the checkpoint folder, e.g., ``global_step14``
551
+
552
+ Returns:
553
+ - ``model`: modified model
554
+
555
+ Make sure you have plenty of CPU memory available before you call this function. If you don't
556
+ have enough use the ``zero_to_fp32.py`` utility to do the conversion. You will find it
557
+ conveniently placed for you in the checkpoint folder.
558
+
559
+ A typical usage might be ::
560
+
561
+ from deepspeed.utils.zero_to_fp32 import load_state_dict_from_zero_checkpoint
562
+ model = load_state_dict_from_zero_checkpoint(trainer.model, checkpoint_dir)
563
+ # submit to model hub or save the model to share with others
564
+
565
+ Note, that once this was run, the ``model`` will no longer be usable in the deepspeed context
566
+ of the same application. i.e. you will need to re-initialize the deepspeed engine, since
567
+ ``model.load_state_dict(state_dict)`` will remove all the deepspeed magic from it.
568
+
569
+ """
570
+ logger.info(f"Extracting fp32 weights")
571
+ state_dict = get_fp32_state_dict_from_zero_checkpoint(checkpoint_dir, tag)
572
+
573
+ logger.info(f"Overwriting model with fp32 weights")
574
+ model = model.cpu()
575
+ model.load_state_dict(state_dict, strict=False)
576
+
577
+ return model
578
+
579
+
580
+ if __name__ == "__main__":
581
+
582
+ parser = argparse.ArgumentParser()
583
+ parser.add_argument("checkpoint_dir",
584
+ type=str,
585
+ help="path to the desired checkpoint folder, e.g., path/checkpoint-12")
586
+ parser.add_argument(
587
+ "output_file",
588
+ type=str,
589
+ help="path to the pytorch fp32 state_dict output file (e.g. path/checkpoint-12/pytorch_model.bin)")
590
+ parser.add_argument("-t",
591
+ "--tag",
592
+ type=str,
593
+ default=None,
594
+ help="checkpoint tag used as a unique identifier for checkpoint. e.g., global_step1")
595
+ parser.add_argument("--exclude_frozen_parameters", action='store_true', help="exclude frozen parameters")
596
+ parser.add_argument("-d", "--debug", action='store_true', help="enable debug")
597
+ args = parser.parse_args()
598
+
599
+ debug = args.debug
600
+
601
+ convert_zero_checkpoint_to_fp32_state_dict(args.checkpoint_dir,
602
+ args.output_file,
603
+ tag=args.tag,
604
+ exclude_frozen_parameters=args.exclude_frozen_parameters)
checkpoint-168/README.md ADDED
@@ -0,0 +1,202 @@
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
1
+ ---
2
+ base_model: deepseek-ai/DeepSeek-R1-Distill-Llama-70B
3
+ library_name: peft
4
+ ---
5
+
6
+ # Model Card for Model ID
7
+
8
+ <!-- Provide a quick summary of what the model is/does. -->
9
+
10
+
11
+
12
+ ## Model Details
13
+
14
+ ### Model Description
15
+
16
+ <!-- Provide a longer summary of what this model is. -->
17
+
18
+
19
+
20
+ - **Developed by:** [More Information Needed]
21
+ - **Funded by [optional]:** [More Information Needed]
22
+ - **Shared by [optional]:** [More Information Needed]
23
+ - **Model type:** [More Information Needed]
24
+ - **Language(s) (NLP):** [More Information Needed]
25
+ - **License:** [More Information Needed]
26
+ - **Finetuned from model [optional]:** [More Information Needed]
27
+
28
+ ### Model Sources [optional]
29
+
30
+ <!-- Provide the basic links for the model. -->
31
+
32
+ - **Repository:** [More Information Needed]
33
+ - **Paper [optional]:** [More Information Needed]
34
+ - **Demo [optional]:** [More Information Needed]
35
+
36
+ ## Uses
37
+
38
+ <!-- Address questions around how the model is intended to be used, including the foreseeable users of the model and those affected by the model. -->
39
+
40
+ ### Direct Use
41
+
42
+ <!-- This section is for the model use without fine-tuning or plugging into a larger ecosystem/app. -->
43
+
44
+ [More Information Needed]
45
+
46
+ ### Downstream Use [optional]
47
+
48
+ <!-- This section is for the model use when fine-tuned for a task, or when plugged into a larger ecosystem/app -->
49
+
50
+ [More Information Needed]
51
+
52
+ ### Out-of-Scope Use
53
+
54
+ <!-- This section addresses misuse, malicious use, and uses that the model will not work well for. -->
55
+
56
+ [More Information Needed]
57
+
58
+ ## Bias, Risks, and Limitations
59
+
60
+ <!-- This section is meant to convey both technical and sociotechnical limitations. -->
61
+
62
+ [More Information Needed]
63
+
64
+ ### Recommendations
65
+
66
+ <!-- This section is meant to convey recommendations with respect to the bias, risk, and technical limitations. -->
67
+
68
+ Users (both direct and downstream) should be made aware of the risks, biases and limitations of the model. More information needed for further recommendations.
69
+
70
+ ## How to Get Started with the Model
71
+
72
+ Use the code below to get started with the model.
73
+
74
+ [More Information Needed]
75
+
76
+ ## Training Details
77
+
78
+ ### Training Data
79
+
80
+ <!-- This should link to a Dataset Card, perhaps with a short stub of information on what the training data is all about as well as documentation related to data pre-processing or additional filtering. -->
81
+
82
+ [More Information Needed]
83
+
84
+ ### Training Procedure
85
+
86
+ <!-- This relates heavily to the Technical Specifications. Content here should link to that section when it is relevant to the training procedure. -->
87
+
88
+ #### Preprocessing [optional]
89
+
90
+ [More Information Needed]
91
+
92
+
93
+ #### Training Hyperparameters
94
+
95
+ - **Training regime:** [More Information Needed] <!--fp32, fp16 mixed precision, bf16 mixed precision, bf16 non-mixed precision, fp16 non-mixed precision, fp8 mixed precision -->
96
+
97
+ #### Speeds, Sizes, Times [optional]
98
+
99
+ <!-- This section provides information about throughput, start/end time, checkpoint size if relevant, etc. -->
100
+
101
+ [More Information Needed]
102
+
103
+ ## Evaluation
104
+
105
+ <!-- This section describes the evaluation protocols and provides the results. -->
106
+
107
+ ### Testing Data, Factors & Metrics
108
+
109
+ #### Testing Data
110
+
111
+ <!-- This should link to a Dataset Card if possible. -->
112
+
113
+ [More Information Needed]
114
+
115
+ #### Factors
116
+
117
+ <!-- These are the things the evaluation is disaggregating by, e.g., subpopulations or domains. -->
118
+
119
+ [More Information Needed]
120
+
121
+ #### Metrics
122
+
123
+ <!-- These are the evaluation metrics being used, ideally with a description of why. -->
124
+
125
+ [More Information Needed]
126
+
127
+ ### Results
128
+
129
+ [More Information Needed]
130
+
131
+ #### Summary
132
+
133
+
134
+
135
+ ## Model Examination [optional]
136
+
137
+ <!-- Relevant interpretability work for the model goes here -->
138
+
139
+ [More Information Needed]
140
+
141
+ ## Environmental Impact
142
+
143
+ <!-- Total emissions (in grams of CO2eq) and additional considerations, such as electricity usage, go here. Edit the suggested text below accordingly -->
144
+
145
+ Carbon emissions can be estimated using the [Machine Learning Impact calculator](https://mlco2.github.io/impact#compute) presented in [Lacoste et al. (2019)](https://arxiv.org/abs/1910.09700).
146
+
147
+ - **Hardware Type:** [More Information Needed]
148
+ - **Hours used:** [More Information Needed]
149
+ - **Cloud Provider:** [More Information Needed]
150
+ - **Compute Region:** [More Information Needed]
151
+ - **Carbon Emitted:** [More Information Needed]
152
+
153
+ ## Technical Specifications [optional]
154
+
155
+ ### Model Architecture and Objective
156
+
157
+ [More Information Needed]
158
+
159
+ ### Compute Infrastructure
160
+
161
+ [More Information Needed]
162
+
163
+ #### Hardware
164
+
165
+ [More Information Needed]
166
+
167
+ #### Software
168
+
169
+ [More Information Needed]
170
+
171
+ ## Citation [optional]
172
+
173
+ <!-- If there is a paper or blog post introducing the model, the APA and Bibtex information for that should go in this section. -->
174
+
175
+ **BibTeX:**
176
+
177
+ [More Information Needed]
178
+
179
+ **APA:**
180
+
181
+ [More Information Needed]
182
+
183
+ ## Glossary [optional]
184
+
185
+ <!-- If relevant, include terms and calculations in this section that can help readers understand the model or model card. -->
186
+
187
+ [More Information Needed]
188
+
189
+ ## More Information [optional]
190
+
191
+ [More Information Needed]
192
+
193
+ ## Model Card Authors [optional]
194
+
195
+ [More Information Needed]
196
+
197
+ ## Model Card Contact
198
+
199
+ [More Information Needed]
200
+ ### Framework versions
201
+
202
+ - PEFT 0.14.0
checkpoint-168/adapter_config.json ADDED
@@ -0,0 +1,40 @@
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
1
+ {
2
+ "alpha_pattern": {},
3
+ "auto_mapping": null,
4
+ "base_model_name_or_path": "deepseek-ai/DeepSeek-R1-Distill-Llama-70B",
5
+ "bias": "none",
6
+ "eva_config": null,
7
+ "exclude_modules": null,
8
+ "fan_in_fan_out": null,
9
+ "inference_mode": true,
10
+ "init_lora_weights": true,
11
+ "layer_replication": null,
12
+ "layers_pattern": null,
13
+ "layers_to_transform": null,
14
+ "loftq_config": {},
15
+ "lora_alpha": 512,
16
+ "lora_bias": false,
17
+ "lora_dropout": 0.05,
18
+ "megatron_config": null,
19
+ "megatron_core": "megatron.core",
20
+ "modules_to_save": [
21
+ "embed_tokens",
22
+ "lm_head"
23
+ ],
24
+ "peft_type": "LORA",
25
+ "r": 256,
26
+ "rank_pattern": {},
27
+ "revision": null,
28
+ "target_modules": [
29
+ "q_proj",
30
+ "o_proj",
31
+ "down_proj",
32
+ "gate_proj",
33
+ "v_proj",
34
+ "k_proj",
35
+ "up_proj"
36
+ ],
37
+ "task_type": "CAUSAL_LM",
38
+ "use_dora": false,
39
+ "use_rslora": false
40
+ }
checkpoint-168/adapter_model.safetensors ADDED
@@ -0,0 +1,3 @@
 
 
 
 
1
+ version https://git-lfs.github.com/spec/v1
2
+ oid sha256:d2b845ca0fbcec9230e875bb8f2fe063ff8b87bbe037863318ec4c7f8ffc431c
3
+ size 10829849744
checkpoint-168/global_step166/bf16_zero_pp_rank_0_mp_rank_00_optim_states.pt ADDED
@@ -0,0 +1,3 @@
 
 
 
 
1
+ version https://git-lfs.github.com/spec/v1
2
+ oid sha256:fed1c50cc7cda2f426adeb1eeb2b219808d1a1b856b39db5ee2416079aeb3d2e
3
+ size 16244554780
checkpoint-168/global_step166/bf16_zero_pp_rank_1_mp_rank_00_optim_states.pt ADDED
@@ -0,0 +1,3 @@
 
 
 
 
1
+ version https://git-lfs.github.com/spec/v1
2
+ oid sha256:e938466de9d05656b9d538b02e249a38dc83cca5c2af5ffdf440b5fb2b3a6129
3
+ size 16244593500
checkpoint-168/global_step166/bf16_zero_pp_rank_2_mp_rank_00_optim_states.pt ADDED
@@ -0,0 +1,3 @@
 
 
 
 
1
+ version https://git-lfs.github.com/spec/v1
2
+ oid sha256:de9fdb8257d88981aca3412f7d6c8b9c66907c4ba2578234ed788c0ed6a92073
3
+ size 16244594204