Hugging Face's logo

kurtpayne
/
skillscan-deberta-adapter

Text Classification
PEFT
ONNX
Safetensors
English
deberta-v2
deberta
prompt-injection
security
lora
skill-security
agent-security
Eval Results (legacy)
Model card Files
xet
 Community
kurtpayne commited on
Commit
acc7b75
·
verified ·
1 Parent(s): f9b2f5a

Auto fine-tune: 11648 examples (LoRA adapter)

Browse files
This view is limited to 50 files because it contains too many changes.   See raw diff
Files changed (50) hide show
  1. adapter_model.safetensors +1 -1
  2. checkpoint-1431/README.md +206 -0
  3. checkpoint-1431/adapter_config.json +44 -0
  4. checkpoint-1431/adapter_model.safetensors +3 -0
  5. checkpoint-1431/optimizer.pt +3 -0
  6. checkpoint-1431/rng_state.pth +3 -0
  7. checkpoint-1431/scaler.pt +3 -0
  8. checkpoint-1431/scheduler.pt +3 -0
  9. checkpoint-1431/special_tokens_map.json +51 -0
  10. checkpoint-1431/tokenizer.json +0 -0
  11. checkpoint-1431/tokenizer_config.json +63 -0
  12. checkpoint-1431/trainer_state.json +1035 -0
  13. checkpoint-1431/training_args.bin +3 -0
  14. checkpoint-2862/README.md +206 -0
  15. checkpoint-2862/adapter_config.json +44 -0
  16. checkpoint-2862/adapter_model.safetensors +3 -0
  17. checkpoint-2862/optimizer.pt +3 -0
  18. checkpoint-2862/rng_state.pth +3 -0
  19. checkpoint-2862/scaler.pt +3 -0
  20. checkpoint-2862/scheduler.pt +3 -0
  21. checkpoint-2862/special_tokens_map.json +51 -0
  22. checkpoint-2862/tokenizer.json +0 -0
  23. checkpoint-2862/tokenizer_config.json +63 -0
  24. checkpoint-2862/trainer_state.json +2036 -0
  25. checkpoint-2862/training_args.bin +3 -0
  26. checkpoint-4293/README.md +206 -0
  27. checkpoint-4293/adapter_config.json +44 -0
  28. checkpoint-4293/adapter_model.safetensors +3 -0
  29. checkpoint-4293/optimizer.pt +3 -0
  30. checkpoint-4293/rng_state.pth +3 -0
  31. checkpoint-4293/scaler.pt +3 -0
  32. checkpoint-4293/scheduler.pt +3 -0
  33. checkpoint-4293/special_tokens_map.json +51 -0
  34. checkpoint-4293/tokenizer.json +0 -0
  35. checkpoint-4293/tokenizer_config.json +63 -0
  36. checkpoint-4293/trainer_state.json +3037 -0
  37. checkpoint-4293/training_args.bin +3 -0
  38. checkpoint-5724/README.md +206 -0
  39. checkpoint-5724/adapter_config.json +44 -0
  40. checkpoint-5724/adapter_model.safetensors +3 -0
  41. checkpoint-5724/optimizer.pt +3 -0
  42. checkpoint-5724/rng_state.pth +3 -0
  43. checkpoint-5724/scaler.pt +3 -0
  44. checkpoint-5724/scheduler.pt +3 -0
  45. checkpoint-5724/special_tokens_map.json +51 -0
  46. checkpoint-5724/tokenizer.json +0 -0
  47. checkpoint-5724/tokenizer_config.json +63 -0
  48. checkpoint-5724/trainer_state.json +0 -0
  49. checkpoint-5724/training_args.bin +3 -0
  50. checkpoint-7155/README.md +206 -0
adapter_model.safetensors CHANGED
@@ -1,3 +1,3 @@
1
  version https://git-lfs.github.com/spec/v1
2
- oid sha256:5853bb105504b2220440be1c0c83605da0dc7b367499fb093c88b9c79b10dc96
3
  size 9450704
 
1
  version https://git-lfs.github.com/spec/v1
2
+ oid sha256:b7e770b0faa195df5977e8d302a1747068350d573934a792741fdc6ad99e7a1e
3
  size 9450704
checkpoint-1431/README.md ADDED
@@ -0,0 +1,206 @@
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
1
+ ---
2
+ base_model: protectai/deberta-v3-base-prompt-injection-v2
3
+ library_name: peft
4
+ tags:
5
+ - base_model:adapter:protectai/deberta-v3-base-prompt-injection-v2
6
+ - lora
7
+ - transformers
8
+ ---
9
+
10
+ # Model Card for Model ID
11
+
12
+ <!-- Provide a quick summary of what the model is/does. -->
13
+
14
+
15
+
16
+ ## Model Details
17
+
18
+ ### Model Description
19
+
20
+ <!-- Provide a longer summary of what this model is. -->
21
+
22
+
23
+
24
+ - **Developed by:** [More Information Needed]
25
+ - **Funded by [optional]:** [More Information Needed]
26
+ - **Shared by [optional]:** [More Information Needed]
27
+ - **Model type:** [More Information Needed]
28
+ - **Language(s) (NLP):** [More Information Needed]
29
+ - **License:** [More Information Needed]
30
+ - **Finetuned from model [optional]:** [More Information Needed]
31
+
32
+ ### Model Sources [optional]
33
+
34
+ <!-- Provide the basic links for the model. -->
35
+
36
+ - **Repository:** [More Information Needed]
37
+ - **Paper [optional]:** [More Information Needed]
38
+ - **Demo [optional]:** [More Information Needed]
39
+
40
+ ## Uses
41
+
42
+ <!-- Address questions around how the model is intended to be used, including the foreseeable users of the model and those affected by the model. -->
43
+
44
+ ### Direct Use
45
+
46
+ <!-- This section is for the model use without fine-tuning or plugging into a larger ecosystem/app. -->
47
+
48
+ [More Information Needed]
49
+
50
+ ### Downstream Use [optional]
51
+
52
+ <!-- This section is for the model use when fine-tuned for a task, or when plugged into a larger ecosystem/app -->
53
+
54
+ [More Information Needed]
55
+
56
+ ### Out-of-Scope Use
57
+
58
+ <!-- This section addresses misuse, malicious use, and uses that the model will not work well for. -->
59
+
60
+ [More Information Needed]
61
+
62
+ ## Bias, Risks, and Limitations
63
+
64
+ <!-- This section is meant to convey both technical and sociotechnical limitations. -->
65
+
66
+ [More Information Needed]
67
+
68
+ ### Recommendations
69
+
70
+ <!-- This section is meant to convey recommendations with respect to the bias, risk, and technical limitations. -->
71
+
72
+ Users (both direct and downstream) should be made aware of the risks, biases and limitations of the model. More information needed for further recommendations.
73
+
74
+ ## How to Get Started with the Model
75
+
76
+ Use the code below to get started with the model.
77
+
78
+ [More Information Needed]
79
+
80
+ ## Training Details
81
+
82
+ ### Training Data
83
+
84
+ <!-- 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. -->
85
+
86
+ [More Information Needed]
87
+
88
+ ### Training Procedure
89
+
90
+ <!-- This relates heavily to the Technical Specifications. Content here should link to that section when it is relevant to the training procedure. -->
91
+
92
+ #### Preprocessing [optional]
93
+
94
+ [More Information Needed]
95
+
96
+
97
+ #### Training Hyperparameters
98
+
99
+ - **Training regime:** [More Information Needed] <!--fp32, fp16 mixed precision, bf16 mixed precision, bf16 non-mixed precision, fp16 non-mixed precision, fp8 mixed precision -->
100
+
101
+ #### Speeds, Sizes, Times [optional]
102
+
103
+ <!-- This section provides information about throughput, start/end time, checkpoint size if relevant, etc. -->
104
+
105
+ [More Information Needed]
106
+
107
+ ## Evaluation
108
+
109
+ <!-- This section describes the evaluation protocols and provides the results. -->
110
+
111
+ ### Testing Data, Factors & Metrics
112
+
113
+ #### Testing Data
114
+
115
+ <!-- This should link to a Dataset Card if possible. -->
116
+
117
+ [More Information Needed]
118
+
119
+ #### Factors
120
+
121
+ <!-- These are the things the evaluation is disaggregating by, e.g., subpopulations or domains. -->
122
+
123
+ [More Information Needed]
124
+
125
+ #### Metrics
126
+
127
+ <!-- These are the evaluation metrics being used, ideally with a description of why. -->
128
+
129
+ [More Information Needed]
130
+
131
+ ### Results
132
+
133
+ [More Information Needed]
134
+
135
+ #### Summary
136
+
137
+
138
+
139
+ ## Model Examination [optional]
140
+
141
+ <!-- Relevant interpretability work for the model goes here -->
142
+
143
+ [More Information Needed]
144
+
145
+ ## Environmental Impact
146
+
147
+ <!-- Total emissions (in grams of CO2eq) and additional considerations, such as electricity usage, go here. Edit the suggested text below accordingly -->
148
+
149
+ 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).
150
+
151
+ - **Hardware Type:** [More Information Needed]
152
+ - **Hours used:** [More Information Needed]
153
+ - **Cloud Provider:** [More Information Needed]
154
+ - **Compute Region:** [More Information Needed]
155
+ - **Carbon Emitted:** [More Information Needed]
156
+
157
+ ## Technical Specifications [optional]
158
+
159
+ ### Model Architecture and Objective
160
+
161
+ [More Information Needed]
162
+
163
+ ### Compute Infrastructure
164
+
165
+ [More Information Needed]
166
+
167
+ #### Hardware
168
+
169
+ [More Information Needed]
170
+
171
+ #### Software
172
+
173
+ [More Information Needed]
174
+
175
+ ## Citation [optional]
176
+
177
+ <!-- If there is a paper or blog post introducing the model, the APA and Bibtex information for that should go in this section. -->
178
+
179
+ **BibTeX:**
180
+
181
+ [More Information Needed]
182
+
183
+ **APA:**
184
+
185
+ [More Information Needed]
186
+
187
+ ## Glossary [optional]
188
+
189
+ <!-- If relevant, include terms and calculations in this section that can help readers understand the model or model card. -->
190
+
191
+ [More Information Needed]
192
+
193
+ ## More Information [optional]
194
+
195
+ [More Information Needed]
196
+
197
+ ## Model Card Authors [optional]
198
+
199
+ [More Information Needed]
200
+
201
+ ## Model Card Contact
202
+
203
+ [More Information Needed]
204
+ ### Framework versions
205
+
206
+ - PEFT 0.18.1
checkpoint-1431/adapter_config.json ADDED
@@ -0,0 +1,44 @@
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
1
+ {
2
+ "alora_invocation_tokens": null,
3
+ "alpha_pattern": {},
4
+ "arrow_config": null,
5
+ "auto_mapping": null,
6
+ "base_model_name_or_path": "protectai/deberta-v3-base-prompt-injection-v2",
7
+ "bias": "none",
8
+ "corda_config": null,
9
+ "ensure_weight_tying": false,
10
+ "eva_config": null,
11
+ "exclude_modules": null,
12
+ "fan_in_fan_out": false,
13
+ "inference_mode": true,
14
+ "init_lora_weights": true,
15
+ "layer_replication": null,
16
+ "layers_pattern": null,
17
+ "layers_to_transform": null,
18
+ "loftq_config": {},
19
+ "lora_alpha": 64,
20
+ "lora_bias": false,
21
+ "lora_dropout": 0.05,
22
+ "megatron_config": null,
23
+ "megatron_core": "megatron.core",
24
+ "modules_to_save": [
25
+ "classifier",
26
+ "score"
27
+ ],
28
+ "peft_type": "LORA",
29
+ "peft_version": "0.18.1",
30
+ "qalora_group_size": 16,
31
+ "r": 64,
32
+ "rank_pattern": {},
33
+ "revision": null,
34
+ "target_modules": [
35
+ "query_proj",
36
+ "value_proj"
37
+ ],
38
+ "target_parameters": null,
39
+ "task_type": "SEQ_CLS",
40
+ "trainable_token_indices": null,
41
+ "use_dora": false,
42
+ "use_qalora": false,
43
+ "use_rslora": false
44
+ }
checkpoint-1431/adapter_model.safetensors ADDED
@@ -0,0 +1,3 @@
 
 
 
 
1
+ version https://git-lfs.github.com/spec/v1
2
+ oid sha256:f64a53199e886595c41ae3edbb80f9fc06e70db576f6d2c2173cc2bc05c58d66
3
+ size 9450704
checkpoint-1431/optimizer.pt ADDED
@@ -0,0 +1,3 @@
 
 
 
 
1
+ version https://git-lfs.github.com/spec/v1
2
+ oid sha256:2f065d62f10aa9c7d9478e9af4cc0b77b52bd0959fa9de7df234624bac926382
3
+ size 18930123
checkpoint-1431/rng_state.pth ADDED
@@ -0,0 +1,3 @@
 
 
 
 
1
+ version https://git-lfs.github.com/spec/v1
2
+ oid sha256:5a843f7c19de0c0e5fa0160e092dfb7623ce309c626eb2544c48885613c82362
3
+ size 14645
checkpoint-1431/scaler.pt ADDED
@@ -0,0 +1,3 @@
 
 
 
 
1
+ version https://git-lfs.github.com/spec/v1
2
+ oid sha256:1c3685736d6ab59ba2a977295663d0f082a48711bb529296d33ab712923f0652
3
+ size 1383
checkpoint-1431/scheduler.pt ADDED
@@ -0,0 +1,3 @@
 
 
 
 
1
+ version https://git-lfs.github.com/spec/v1
2
+ oid sha256:19044e65fdc56737d3cb91212f33c7b7f23f4c38440ca075595890b8fe1143ad
3
+ size 1465
checkpoint-1431/special_tokens_map.json ADDED
@@ -0,0 +1,51 @@
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
1
+ {
2
+ "bos_token": {
3
+ "content": "[CLS]",
4
+ "lstrip": false,
5
+ "normalized": false,
6
+ "rstrip": false,
7
+ "single_word": false
8
+ },
9
+ "cls_token": {
10
+ "content": "[CLS]",
11
+ "lstrip": false,
12
+ "normalized": false,
13
+ "rstrip": false,
14
+ "single_word": false
15
+ },
16
+ "eos_token": {
17
+ "content": "[SEP]",
18
+ "lstrip": false,
19
+ "normalized": false,
20
+ "rstrip": false,
21
+ "single_word": false
22
+ },
23
+ "mask_token": {
24
+ "content": "[MASK]",
25
+ "lstrip": false,
26
+ "normalized": false,
27
+ "rstrip": false,
28
+ "single_word": false
29
+ },
30
+ "pad_token": {
31
+ "content": "[PAD]",
32
+ "lstrip": false,
33
+ "normalized": false,
34
+ "rstrip": false,
35
+ "single_word": false
36
+ },
37
+ "sep_token": {
38
+ "content": "[SEP]",
39
+ "lstrip": false,
40
+ "normalized": false,
41
+ "rstrip": false,
42
+ "single_word": false
43
+ },
44
+ "unk_token": {
45
+ "content": "[UNK]",
46
+ "lstrip": false,
47
+ "normalized": true,
48
+ "rstrip": false,
49
+ "single_word": false
50
+ }
51
+ }
checkpoint-1431/tokenizer.json ADDED
The diff for this file is too large to render. See raw diff
 
checkpoint-1431/tokenizer_config.json ADDED
@@ -0,0 +1,63 @@
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
1
+ {
2
+ "added_tokens_decoder": {
3
+ "0": {
4
+ "content": "[PAD]",
5
+ "lstrip": false,
6
+ "normalized": false,
7
+ "rstrip": false,
8
+ "single_word": false,
9
+ "special": true
10
+ },
11
+ "1": {
12
+ "content": "[CLS]",
13
+ "lstrip": false,
14
+ "normalized": false,
15
+ "rstrip": false,
16
+ "single_word": false,
17
+ "special": true
18
+ },
19
+ "2": {
20
+ "content": "[SEP]",
21
+ "lstrip": false,
22
+ "normalized": false,
23
+ "rstrip": false,
24
+ "single_word": false,
25
+ "special": true
26
+ },
27
+ "3": {
28
+ "content": "[UNK]",
29
+ "lstrip": false,
30
+ "normalized": true,
31
+ "rstrip": false,
32
+ "single_word": false,
33
+ "special": true
34
+ },
35
+ "128000": {
36
+ "content": "[MASK]",
37
+ "lstrip": false,
38
+ "normalized": false,
39
+ "rstrip": false,
40
+ "single_word": false,
41
+ "special": true
42
+ }
43
+ },
44
+ "bos_token": "[CLS]",
45
+ "clean_up_tokenization_spaces": true,
46
+ "cls_token": "[CLS]",
47
+ "do_lower_case": false,
48
+ "eos_token": "[SEP]",
49
+ "extra_special_tokens": {},
50
+ "mask_token": "[MASK]",
51
+ "max_length": 512,
52
+ "model_max_length": 1000000000000000019884624838656,
53
+ "pad_token": "[PAD]",
54
+ "sep_token": "[SEP]",
55
+ "sp_model_kwargs": {},
56
+ "split_by_punct": false,
57
+ "stride": 0,
58
+ "tokenizer_class": "DebertaV2Tokenizer",
59
+ "truncation_side": "right",
60
+ "truncation_strategy": "longest_first",
61
+ "unk_token": "[UNK]",
62
+ "vocab_type": "spm"
63
+ }
checkpoint-1431/trainer_state.json ADDED
@@ -0,0 +1,1035 @@
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
1
+ {
2
+ "best_global_step": null,
3
+ "best_metric": null,
4
+ "best_model_checkpoint": null,
5
+ "epoch": 1.0,
6
+ "eval_steps": 500,
7
+ "global_step": 1431,
8
+ "is_hyper_param_search": false,
9
+ "is_local_process_zero": true,
10
+ "is_world_process_zero": true,
11
+ "log_history": [
12
+ {
13
+ "epoch": 0.0069881201956673656,
14
+ "grad_norm": 14.528067588806152,
15
+ "learning_rate": 1.99748427672956e-05,
16
+ "loss": 4.6827,
17
+ "step": 10
18
+ },
19
+ {
20
+ "epoch": 0.013976240391334731,
21
+ "grad_norm": 13.050820350646973,
22
+ "learning_rate": 1.994689028651293e-05,
23
+ "loss": 4.529,
24
+ "step": 20
25
+ },
26
+ {
27
+ "epoch": 0.020964360587002098,
28
+ "grad_norm": 2.765507698059082,
29
+ "learning_rate": 1.991893780573026e-05,
30
+ "loss": 3.4294,
31
+ "step": 30
32
+ },
33
+ {
34
+ "epoch": 0.027952480782669462,
35
+ "grad_norm": 16.95716667175293,
36
+ "learning_rate": 1.989098532494759e-05,
37
+ "loss": 3.352,
38
+ "step": 40
39
+ },
40
+ {
41
+ "epoch": 0.03494060097833683,
42
+ "grad_norm": 5.339346408843994,
43
+ "learning_rate": 1.986303284416492e-05,
44
+ "loss": 2.2654,
45
+ "step": 50
46
+ },
47
+ {
48
+ "epoch": 0.041928721174004195,
49
+ "grad_norm": 7.168309688568115,
50
+ "learning_rate": 1.983508036338225e-05,
51
+ "loss": 2.0076,
52
+ "step": 60
53
+ },
54
+ {
55
+ "epoch": 0.04891684136967156,
56
+ "grad_norm": 5.793034076690674,
57
+ "learning_rate": 1.9807127882599582e-05,
58
+ "loss": 1.6779,
59
+ "step": 70
60
+ },
61
+ {
62
+ "epoch": 0.055904961565338925,
63
+ "grad_norm": 6.093694686889648,
64
+ "learning_rate": 1.9779175401816913e-05,
65
+ "loss": 1.2564,
66
+ "step": 80
67
+ },
68
+ {
69
+ "epoch": 0.06289308176100629,
70
+ "grad_norm": 3.321807622909546,
71
+ "learning_rate": 1.9751222921034243e-05,
72
+ "loss": 1.4143,
73
+ "step": 90
74
+ },
75
+ {
76
+ "epoch": 0.06988120195667366,
77
+ "grad_norm": 2.9472098350524902,
78
+ "learning_rate": 1.9723270440251574e-05,
79
+ "loss": 1.0533,
80
+ "step": 100
81
+ },
82
+ {
83
+ "epoch": 0.07686932215234102,
84
+ "grad_norm": 8.515447616577148,
85
+ "learning_rate": 1.9695317959468904e-05,
86
+ "loss": 1.2895,
87
+ "step": 110
88
+ },
89
+ {
90
+ "epoch": 0.08385744234800839,
91
+ "grad_norm": 3.560924530029297,
92
+ "learning_rate": 1.9667365478686235e-05,
93
+ "loss": 1.097,
94
+ "step": 120
95
+ },
96
+ {
97
+ "epoch": 0.09084556254367575,
98
+ "grad_norm": 4.986349582672119,
99
+ "learning_rate": 1.9639412997903566e-05,
100
+ "loss": 0.9556,
101
+ "step": 130
102
+ },
103
+ {
104
+ "epoch": 0.09783368273934312,
105
+ "grad_norm": 1.27532160282135,
106
+ "learning_rate": 1.9611460517120896e-05,
107
+ "loss": 0.8306,
108
+ "step": 140
109
+ },
110
+ {
111
+ "epoch": 0.10482180293501048,
112
+ "grad_norm": 2.6796844005584717,
113
+ "learning_rate": 1.9583508036338227e-05,
114
+ "loss": 0.7747,
115
+ "step": 150
116
+ },
117
+ {
118
+ "epoch": 0.11180992313067785,
119
+ "grad_norm": 3.4592413902282715,
120
+ "learning_rate": 1.9555555555555557e-05,
121
+ "loss": 0.6187,
122
+ "step": 160
123
+ },
124
+ {
125
+ "epoch": 0.1187980433263452,
126
+ "grad_norm": 4.833241939544678,
127
+ "learning_rate": 1.9527603074772888e-05,
128
+ "loss": 0.6831,
129
+ "step": 170
130
+ },
131
+ {
132
+ "epoch": 0.12578616352201258,
133
+ "grad_norm": 1.7203434705734253,
134
+ "learning_rate": 1.949965059399022e-05,
135
+ "loss": 0.586,
136
+ "step": 180
137
+ },
138
+ {
139
+ "epoch": 0.13277428371767994,
140
+ "grad_norm": 1.333950161933899,
141
+ "learning_rate": 1.947169811320755e-05,
142
+ "loss": 0.6214,
143
+ "step": 190
144
+ },
145
+ {
146
+ "epoch": 0.13976240391334732,
147
+ "grad_norm": 1.5078257322311401,
148
+ "learning_rate": 1.944374563242488e-05,
149
+ "loss": 0.5652,
150
+ "step": 200
151
+ },
152
+ {
153
+ "epoch": 0.14675052410901468,
154
+ "grad_norm": 3.7645132541656494,
155
+ "learning_rate": 1.941579315164221e-05,
156
+ "loss": 0.5229,
157
+ "step": 210
158
+ },
159
+ {
160
+ "epoch": 0.15373864430468204,
161
+ "grad_norm": 1.2441000938415527,
162
+ "learning_rate": 1.938784067085954e-05,
163
+ "loss": 0.3669,
164
+ "step": 220
165
+ },
166
+ {
167
+ "epoch": 0.1607267645003494,
168
+ "grad_norm": 2.2875142097473145,
169
+ "learning_rate": 1.935988819007687e-05,
170
+ "loss": 0.4162,
171
+ "step": 230
172
+ },
173
+ {
174
+ "epoch": 0.16771488469601678,
175
+ "grad_norm": 2.1816585063934326,
176
+ "learning_rate": 1.9331935709294202e-05,
177
+ "loss": 0.3836,
178
+ "step": 240
179
+ },
180
+ {
181
+ "epoch": 0.17470300489168414,
182
+ "grad_norm": 3.4903249740600586,
183
+ "learning_rate": 1.9303983228511532e-05,
184
+ "loss": 0.4131,
185
+ "step": 250
186
+ },
187
+ {
188
+ "epoch": 0.1816911250873515,
189
+ "grad_norm": 2.1905159950256348,
190
+ "learning_rate": 1.9276030747728863e-05,
191
+ "loss": 0.3725,
192
+ "step": 260
193
+ },
194
+ {
195
+ "epoch": 0.18867924528301888,
196
+ "grad_norm": 3.6003665924072266,
197
+ "learning_rate": 1.9248078266946194e-05,
198
+ "loss": 0.4131,
199
+ "step": 270
200
+ },
201
+ {
202
+ "epoch": 0.19566736547868624,
203
+ "grad_norm": 3.012568235397339,
204
+ "learning_rate": 1.9220125786163524e-05,
205
+ "loss": 0.3991,
206
+ "step": 280
207
+ },
208
+ {
209
+ "epoch": 0.2026554856743536,
210
+ "grad_norm": 1.2049503326416016,
211
+ "learning_rate": 1.9192173305380855e-05,
212
+ "loss": 0.2732,
213
+ "step": 290
214
+ },
215
+ {
216
+ "epoch": 0.20964360587002095,
217
+ "grad_norm": 2.0943682193756104,
218
+ "learning_rate": 1.9164220824598185e-05,
219
+ "loss": 0.3052,
220
+ "step": 300
221
+ },
222
+ {
223
+ "epoch": 0.21663172606568834,
224
+ "grad_norm": 4.1970086097717285,
225
+ "learning_rate": 1.9136268343815516e-05,
226
+ "loss": 0.2544,
227
+ "step": 310
228
+ },
229
+ {
230
+ "epoch": 0.2236198462613557,
231
+ "grad_norm": 2.6826064586639404,
232
+ "learning_rate": 1.9108315863032846e-05,
233
+ "loss": 0.1921,
234
+ "step": 320
235
+ },
236
+ {
237
+ "epoch": 0.23060796645702306,
238
+ "grad_norm": 6.013642311096191,
239
+ "learning_rate": 1.9080363382250177e-05,
240
+ "loss": 0.3014,
241
+ "step": 330
242
+ },
243
+ {
244
+ "epoch": 0.2375960866526904,
245
+ "grad_norm": 4.2075629234313965,
246
+ "learning_rate": 1.9052410901467508e-05,
247
+ "loss": 0.2569,
248
+ "step": 340
249
+ },
250
+ {
251
+ "epoch": 0.2445842068483578,
252
+ "grad_norm": 0.5800023078918457,
253
+ "learning_rate": 1.9024458420684838e-05,
254
+ "loss": 0.1486,
255
+ "step": 350
256
+ },
257
+ {
258
+ "epoch": 0.25157232704402516,
259
+ "grad_norm": 8.123499870300293,
260
+ "learning_rate": 1.899650593990217e-05,
261
+ "loss": 0.2215,
262
+ "step": 360
263
+ },
264
+ {
265
+ "epoch": 0.2585604472396925,
266
+ "grad_norm": 4.439255237579346,
267
+ "learning_rate": 1.89685534591195e-05,
268
+ "loss": 0.1692,
269
+ "step": 370
270
+ },
271
+ {
272
+ "epoch": 0.2655485674353599,
273
+ "grad_norm": 4.177631855010986,
274
+ "learning_rate": 1.894060097833683e-05,
275
+ "loss": 0.1176,
276
+ "step": 380
277
+ },
278
+ {
279
+ "epoch": 0.27253668763102723,
280
+ "grad_norm": 0.5754140019416809,
281
+ "learning_rate": 1.891264849755416e-05,
282
+ "loss": 0.1538,
283
+ "step": 390
284
+ },
285
+ {
286
+ "epoch": 0.27952480782669464,
287
+ "grad_norm": 0.5612320303916931,
288
+ "learning_rate": 1.888469601677149e-05,
289
+ "loss": 0.1054,
290
+ "step": 400
291
+ },
292
+ {
293
+ "epoch": 0.286512928022362,
294
+ "grad_norm": 1.949729561805725,
295
+ "learning_rate": 1.8856743535988818e-05,
296
+ "loss": 0.0824,
297
+ "step": 410
298
+ },
299
+ {
300
+ "epoch": 0.29350104821802936,
301
+ "grad_norm": 3.153822898864746,
302
+ "learning_rate": 1.8828791055206152e-05,
303
+ "loss": 0.1174,
304
+ "step": 420
305
+ },
306
+ {
307
+ "epoch": 0.3004891684136967,
308
+ "grad_norm": 0.6545250415802002,
309
+ "learning_rate": 1.8800838574423483e-05,
310
+ "loss": 0.1305,
311
+ "step": 430
312
+ },
313
+ {
314
+ "epoch": 0.3074772886093641,
315
+ "grad_norm": 1.2621614933013916,
316
+ "learning_rate": 1.8772886093640813e-05,
317
+ "loss": 0.0684,
318
+ "step": 440
319
+ },
320
+ {
321
+ "epoch": 0.31446540880503143,
322
+ "grad_norm": 0.3028416335582733,
323
+ "learning_rate": 1.8744933612858144e-05,
324
+ "loss": 0.0863,
325
+ "step": 450
326
+ },
327
+ {
328
+ "epoch": 0.3214535290006988,
329
+ "grad_norm": 0.4452091455459595,
330
+ "learning_rate": 1.8716981132075474e-05,
331
+ "loss": 0.1016,
332
+ "step": 460
333
+ },
334
+ {
335
+ "epoch": 0.3284416491963662,
336
+ "grad_norm": 0.4046468436717987,
337
+ "learning_rate": 1.8689028651292805e-05,
338
+ "loss": 0.2317,
339
+ "step": 470
340
+ },
341
+ {
342
+ "epoch": 0.33542976939203356,
343
+ "grad_norm": 0.5121694803237915,
344
+ "learning_rate": 1.8661076170510135e-05,
345
+ "loss": 0.1965,
346
+ "step": 480
347
+ },
348
+ {
349
+ "epoch": 0.3424178895877009,
350
+ "grad_norm": 2.7892794609069824,
351
+ "learning_rate": 1.8633123689727466e-05,
352
+ "loss": 0.2082,
353
+ "step": 490
354
+ },
355
+ {
356
+ "epoch": 0.3494060097833683,
357
+ "grad_norm": 0.29012227058410645,
358
+ "learning_rate": 1.8605171208944793e-05,
359
+ "loss": 0.1393,
360
+ "step": 500
361
+ },
362
+ {
363
+ "epoch": 0.35639412997903563,
364
+ "grad_norm": 0.14969919621944427,
365
+ "learning_rate": 1.8577218728162124e-05,
366
+ "loss": 0.1917,
367
+ "step": 510
368
+ },
369
+ {
370
+ "epoch": 0.363382250174703,
371
+ "grad_norm": 2.5301618576049805,
372
+ "learning_rate": 1.8549266247379458e-05,
373
+ "loss": 0.0939,
374
+ "step": 520
375
+ },
376
+ {
377
+ "epoch": 0.37037037037037035,
378
+ "grad_norm": 0.5171930193901062,
379
+ "learning_rate": 1.8521313766596788e-05,
380
+ "loss": 0.0643,
381
+ "step": 530
382
+ },
383
+ {
384
+ "epoch": 0.37735849056603776,
385
+ "grad_norm": 0.23746764659881592,
386
+ "learning_rate": 1.849336128581412e-05,
387
+ "loss": 0.2094,
388
+ "step": 540
389
+ },
390
+ {
391
+ "epoch": 0.3843466107617051,
392
+ "grad_norm": 1.2422730922698975,
393
+ "learning_rate": 1.846540880503145e-05,
394
+ "loss": 0.109,
395
+ "step": 550
396
+ },
397
+ {
398
+ "epoch": 0.3913347309573725,
399
+ "grad_norm": 0.8466539978981018,
400
+ "learning_rate": 1.843745632424878e-05,
401
+ "loss": 0.1514,
402
+ "step": 560
403
+ },
404
+ {
405
+ "epoch": 0.39832285115303984,
406
+ "grad_norm": 0.201154425740242,
407
+ "learning_rate": 1.840950384346611e-05,
408
+ "loss": 0.1597,
409
+ "step": 570
410
+ },
411
+ {
412
+ "epoch": 0.4053109713487072,
413
+ "grad_norm": 0.19054199755191803,
414
+ "learning_rate": 1.838155136268344e-05,
415
+ "loss": 0.0955,
416
+ "step": 580
417
+ },
418
+ {
419
+ "epoch": 0.41229909154437455,
420
+ "grad_norm": 1.8036576509475708,
421
+ "learning_rate": 1.8353598881900768e-05,
422
+ "loss": 0.1369,
423
+ "step": 590
424
+ },
425
+ {
426
+ "epoch": 0.4192872117400419,
427
+ "grad_norm": 0.3122437596321106,
428
+ "learning_rate": 1.83256464011181e-05,
429
+ "loss": 0.1439,
430
+ "step": 600
431
+ },
432
+ {
433
+ "epoch": 0.42627533193570927,
434
+ "grad_norm": 2.6353561878204346,
435
+ "learning_rate": 1.829769392033543e-05,
436
+ "loss": 0.1566,
437
+ "step": 610
438
+ },
439
+ {
440
+ "epoch": 0.4332634521313767,
441
+ "grad_norm": 3.645707368850708,
442
+ "learning_rate": 1.8269741439552763e-05,
443
+ "loss": 0.2505,
444
+ "step": 620
445
+ },
446
+ {
447
+ "epoch": 0.44025157232704404,
448
+ "grad_norm": 0.20553793013095856,
449
+ "learning_rate": 1.8241788958770094e-05,
450
+ "loss": 0.0509,
451
+ "step": 630
452
+ },
453
+ {
454
+ "epoch": 0.4472396925227114,
455
+ "grad_norm": 3.8460299968719482,
456
+ "learning_rate": 1.8213836477987425e-05,
457
+ "loss": 0.118,
458
+ "step": 640
459
+ },
460
+ {
461
+ "epoch": 0.45422781271837875,
462
+ "grad_norm": 0.4825759828090668,
463
+ "learning_rate": 1.8185883997204755e-05,
464
+ "loss": 0.0671,
465
+ "step": 650
466
+ },
467
+ {
468
+ "epoch": 0.4612159329140461,
469
+ "grad_norm": 2.5147883892059326,
470
+ "learning_rate": 1.8157931516422086e-05,
471
+ "loss": 0.1381,
472
+ "step": 660
473
+ },
474
+ {
475
+ "epoch": 0.46820405310971347,
476
+ "grad_norm": 4.72428035736084,
477
+ "learning_rate": 1.8129979035639413e-05,
478
+ "loss": 0.1093,
479
+ "step": 670
480
+ },
481
+ {
482
+ "epoch": 0.4751921733053808,
483
+ "grad_norm": 3.091625928878784,
484
+ "learning_rate": 1.8102026554856743e-05,
485
+ "loss": 0.2363,
486
+ "step": 680
487
+ },
488
+ {
489
+ "epoch": 0.48218029350104824,
490
+ "grad_norm": 0.10304142534732819,
491
+ "learning_rate": 1.8074074074074074e-05,
492
+ "loss": 0.1347,
493
+ "step": 690
494
+ },
495
+ {
496
+ "epoch": 0.4891684136967156,
497
+ "grad_norm": 0.4455307424068451,
498
+ "learning_rate": 1.8046121593291405e-05,
499
+ "loss": 0.1064,
500
+ "step": 700
501
+ },
502
+ {
503
+ "epoch": 0.49615653389238296,
504
+ "grad_norm": 0.11594414710998535,
505
+ "learning_rate": 1.8018169112508735e-05,
506
+ "loss": 0.123,
507
+ "step": 710
508
+ },
509
+ {
510
+ "epoch": 0.5031446540880503,
511
+ "grad_norm": 0.1224653497338295,
512
+ "learning_rate": 1.799021663172607e-05,
513
+ "loss": 0.1263,
514
+ "step": 720
515
+ },
516
+ {
517
+ "epoch": 0.5101327742837177,
518
+ "grad_norm": 4.951889991760254,
519
+ "learning_rate": 1.79622641509434e-05,
520
+ "loss": 0.2111,
521
+ "step": 730
522
+ },
523
+ {
524
+ "epoch": 0.517120894479385,
525
+ "grad_norm": 0.16316205263137817,
526
+ "learning_rate": 1.793431167016073e-05,
527
+ "loss": 0.0366,
528
+ "step": 740
529
+ },
530
+ {
531
+ "epoch": 0.5241090146750524,
532
+ "grad_norm": 0.1896398663520813,
533
+ "learning_rate": 1.790635918937806e-05,
534
+ "loss": 0.1656,
535
+ "step": 750
536
+ },
537
+ {
538
+ "epoch": 0.5310971348707197,
539
+ "grad_norm": 0.3920069932937622,
540
+ "learning_rate": 1.7878406708595388e-05,
541
+ "loss": 0.2804,
542
+ "step": 760
543
+ },
544
+ {
545
+ "epoch": 0.5380852550663872,
546
+ "grad_norm": 0.12546010315418243,
547
+ "learning_rate": 1.785045422781272e-05,
548
+ "loss": 0.1404,
549
+ "step": 770
550
+ },
551
+ {
552
+ "epoch": 0.5450733752620545,
553
+ "grad_norm": 5.382510185241699,
554
+ "learning_rate": 1.782250174703005e-05,
555
+ "loss": 0.1194,
556
+ "step": 780
557
+ },
558
+ {
559
+ "epoch": 0.5520614954577219,
560
+ "grad_norm": 0.3077394366264343,
561
+ "learning_rate": 1.779454926624738e-05,
562
+ "loss": 0.1744,
563
+ "step": 790
564
+ },
565
+ {
566
+ "epoch": 0.5590496156533893,
567
+ "grad_norm": 0.41757112741470337,
568
+ "learning_rate": 1.776659678546471e-05,
569
+ "loss": 0.0729,
570
+ "step": 800
571
+ },
572
+ {
573
+ "epoch": 0.5660377358490566,
574
+ "grad_norm": 0.17300930619239807,
575
+ "learning_rate": 1.773864430468204e-05,
576
+ "loss": 0.1443,
577
+ "step": 810
578
+ },
579
+ {
580
+ "epoch": 0.573025856044724,
581
+ "grad_norm": 0.09419082850217819,
582
+ "learning_rate": 1.7710691823899375e-05,
583
+ "loss": 0.1587,
584
+ "step": 820
585
+ },
586
+ {
587
+ "epoch": 0.5800139762403913,
588
+ "grad_norm": 0.14113612473011017,
589
+ "learning_rate": 1.7682739343116705e-05,
590
+ "loss": 0.1196,
591
+ "step": 830
592
+ },
593
+ {
594
+ "epoch": 0.5870020964360587,
595
+ "grad_norm": 0.6690120697021484,
596
+ "learning_rate": 1.7654786862334036e-05,
597
+ "loss": 0.0619,
598
+ "step": 840
599
+ },
600
+ {
601
+ "epoch": 0.593990216631726,
602
+ "grad_norm": 0.5076020956039429,
603
+ "learning_rate": 1.7626834381551363e-05,
604
+ "loss": 0.2922,
605
+ "step": 850
606
+ },
607
+ {
608
+ "epoch": 0.6009783368273934,
609
+ "grad_norm": 0.12461690604686737,
610
+ "learning_rate": 1.7598881900768694e-05,
611
+ "loss": 0.0351,
612
+ "step": 860
613
+ },
614
+ {
615
+ "epoch": 0.6079664570230608,
616
+ "grad_norm": 5.236762523651123,
617
+ "learning_rate": 1.7570929419986024e-05,
618
+ "loss": 0.1571,
619
+ "step": 870
620
+ },
621
+ {
622
+ "epoch": 0.6149545772187281,
623
+ "grad_norm": 0.12950463593006134,
624
+ "learning_rate": 1.7542976939203355e-05,
625
+ "loss": 0.0204,
626
+ "step": 880
627
+ },
628
+ {
629
+ "epoch": 0.6219426974143956,
630
+ "grad_norm": 0.08026222884654999,
631
+ "learning_rate": 1.7515024458420685e-05,
632
+ "loss": 0.1185,
633
+ "step": 890
634
+ },
635
+ {
636
+ "epoch": 0.6289308176100629,
637
+ "grad_norm": 4.237085342407227,
638
+ "learning_rate": 1.7487071977638016e-05,
639
+ "loss": 0.0876,
640
+ "step": 900
641
+ },
642
+ {
643
+ "epoch": 0.6359189378057303,
644
+ "grad_norm": 0.916748046875,
645
+ "learning_rate": 1.7459119496855346e-05,
646
+ "loss": 0.0619,
647
+ "step": 910
648
+ },
649
+ {
650
+ "epoch": 0.6429070580013976,
651
+ "grad_norm": 4.51674747467041,
652
+ "learning_rate": 1.743116701607268e-05,
653
+ "loss": 0.1116,
654
+ "step": 920
655
+ },
656
+ {
657
+ "epoch": 0.649895178197065,
658
+ "grad_norm": 1.1469999551773071,
659
+ "learning_rate": 1.740321453529001e-05,
660
+ "loss": 0.0147,
661
+ "step": 930
662
+ },
663
+ {
664
+ "epoch": 0.6568832983927324,
665
+ "grad_norm": 2.3627333641052246,
666
+ "learning_rate": 1.7375262054507338e-05,
667
+ "loss": 0.1485,
668
+ "step": 940
669
+ },
670
+ {
671
+ "epoch": 0.6638714185883997,
672
+ "grad_norm": 8.216973304748535,
673
+ "learning_rate": 1.734730957372467e-05,
674
+ "loss": 0.0472,
675
+ "step": 950
676
+ },
677
+ {
678
+ "epoch": 0.6708595387840671,
679
+ "grad_norm": 3.5901684761047363,
680
+ "learning_rate": 1.7319357092942e-05,
681
+ "loss": 0.0979,
682
+ "step": 960
683
+ },
684
+ {
685
+ "epoch": 0.6778476589797344,
686
+ "grad_norm": 7.015125274658203,
687
+ "learning_rate": 1.729140461215933e-05,
688
+ "loss": 0.0537,
689
+ "step": 970
690
+ },
691
+ {
692
+ "epoch": 0.6848357791754018,
693
+ "grad_norm": 4.372157096862793,
694
+ "learning_rate": 1.726345213137666e-05,
695
+ "loss": 0.1152,
696
+ "step": 980
697
+ },
698
+ {
699
+ "epoch": 0.6918238993710691,
700
+ "grad_norm": 4.862729072570801,
701
+ "learning_rate": 1.723549965059399e-05,
702
+ "loss": 0.1131,
703
+ "step": 990
704
+ },
705
+ {
706
+ "epoch": 0.6988120195667366,
707
+ "grad_norm": 0.07732150703668594,
708
+ "learning_rate": 1.720754716981132e-05,
709
+ "loss": 0.0707,
710
+ "step": 1000
711
+ },
712
+ {
713
+ "epoch": 0.705800139762404,
714
+ "grad_norm": 0.19970574975013733,
715
+ "learning_rate": 1.7179594689028652e-05,
716
+ "loss": 0.0738,
717
+ "step": 1010
718
+ },
719
+ {
720
+ "epoch": 0.7127882599580713,
721
+ "grad_norm": 0.21677608788013458,
722
+ "learning_rate": 1.7151642208245983e-05,
723
+ "loss": 0.0992,
724
+ "step": 1020
725
+ },
726
+ {
727
+ "epoch": 0.7197763801537387,
728
+ "grad_norm": 0.10738188028335571,
729
+ "learning_rate": 1.7123689727463313e-05,
730
+ "loss": 0.0103,
731
+ "step": 1030
732
+ },
733
+ {
734
+ "epoch": 0.726764500349406,
735
+ "grad_norm": 3.7014143466949463,
736
+ "learning_rate": 1.7095737246680644e-05,
737
+ "loss": 0.0803,
738
+ "step": 1040
739
+ },
740
+ {
741
+ "epoch": 0.7337526205450734,
742
+ "grad_norm": 0.04619055986404419,
743
+ "learning_rate": 1.7067784765897974e-05,
744
+ "loss": 0.049,
745
+ "step": 1050
746
+ },
747
+ {
748
+ "epoch": 0.7407407407407407,
749
+ "grad_norm": 0.2753339409828186,
750
+ "learning_rate": 1.7039832285115305e-05,
751
+ "loss": 0.144,
752
+ "step": 1060
753
+ },
754
+ {
755
+ "epoch": 0.7477288609364081,
756
+ "grad_norm": 2.8835108280181885,
757
+ "learning_rate": 1.7011879804332635e-05,
758
+ "loss": 0.0885,
759
+ "step": 1070
760
+ },
761
+ {
762
+ "epoch": 0.7547169811320755,
763
+ "grad_norm": 0.22654394805431366,
764
+ "learning_rate": 1.6983927323549966e-05,
765
+ "loss": 0.0512,
766
+ "step": 1080
767
+ },
768
+ {
769
+ "epoch": 0.7617051013277428,
770
+ "grad_norm": 2.421400547027588,
771
+ "learning_rate": 1.6955974842767297e-05,
772
+ "loss": 0.1544,
773
+ "step": 1090
774
+ },
775
+ {
776
+ "epoch": 0.7686932215234102,
777
+ "grad_norm": 0.03806561604142189,
778
+ "learning_rate": 1.6928022361984627e-05,
779
+ "loss": 0.0225,
780
+ "step": 1100
781
+ },
782
+ {
783
+ "epoch": 0.7756813417190775,
784
+ "grad_norm": 0.08347579091787338,
785
+ "learning_rate": 1.6900069881201958e-05,
786
+ "loss": 0.1386,
787
+ "step": 1110
788
+ },
789
+ {
790
+ "epoch": 0.782669461914745,
791
+ "grad_norm": 0.2059401273727417,
792
+ "learning_rate": 1.687211740041929e-05,
793
+ "loss": 0.2406,
794
+ "step": 1120
795
+ },
796
+ {
797
+ "epoch": 0.7896575821104123,
798
+ "grad_norm": 0.060870036482810974,
799
+ "learning_rate": 1.684416491963662e-05,
800
+ "loss": 0.1158,
801
+ "step": 1130
802
+ },
803
+ {
804
+ "epoch": 0.7966457023060797,
805
+ "grad_norm": 0.14614495635032654,
806
+ "learning_rate": 1.681621243885395e-05,
807
+ "loss": 0.1938,
808
+ "step": 1140
809
+ },
810
+ {
811
+ "epoch": 0.803633822501747,
812
+ "grad_norm": 0.09593985229730606,
813
+ "learning_rate": 1.678825995807128e-05,
814
+ "loss": 0.0627,
815
+ "step": 1150
816
+ },
817
+ {
818
+ "epoch": 0.8106219426974144,
819
+ "grad_norm": 9.543107032775879,
820
+ "learning_rate": 1.676030747728861e-05,
821
+ "loss": 0.1365,
822
+ "step": 1160
823
+ },
824
+ {
825
+ "epoch": 0.8176100628930818,
826
+ "grad_norm": 0.09678485244512558,
827
+ "learning_rate": 1.673235499650594e-05,
828
+ "loss": 0.1438,
829
+ "step": 1170
830
+ },
831
+ {
832
+ "epoch": 0.8245981830887491,
833
+ "grad_norm": 0.24556872248649597,
834
+ "learning_rate": 1.6704402515723272e-05,
835
+ "loss": 0.2233,
836
+ "step": 1180
837
+ },
838
+ {
839
+ "epoch": 0.8315863032844165,
840
+ "grad_norm": 0.06892254948616028,
841
+ "learning_rate": 1.6676450034940602e-05,
842
+ "loss": 0.0601,
843
+ "step": 1190
844
+ },
845
+ {
846
+ "epoch": 0.8385744234800838,
847
+ "grad_norm": 0.6429911255836487,
848
+ "learning_rate": 1.6648497554157933e-05,
849
+ "loss": 0.132,
850
+ "step": 1200
851
+ },
852
+ {
853
+ "epoch": 0.8455625436757512,
854
+ "grad_norm": 0.03480253368616104,
855
+ "learning_rate": 1.6620545073375263e-05,
856
+ "loss": 0.0472,
857
+ "step": 1210
858
+ },
859
+ {
860
+ "epoch": 0.8525506638714185,
861
+ "grad_norm": 0.25101321935653687,
862
+ "learning_rate": 1.6592592592592594e-05,
863
+ "loss": 0.0485,
864
+ "step": 1220
865
+ },
866
+ {
867
+ "epoch": 0.859538784067086,
868
+ "grad_norm": 0.0525231771171093,
869
+ "learning_rate": 1.6564640111809925e-05,
870
+ "loss": 0.1824,
871
+ "step": 1230
872
+ },
873
+ {
874
+ "epoch": 0.8665269042627534,
875
+ "grad_norm": 4.694910526275635,
876
+ "learning_rate": 1.6536687631027255e-05,
877
+ "loss": 0.1666,
878
+ "step": 1240
879
+ },
880
+ {
881
+ "epoch": 0.8735150244584207,
882
+ "grad_norm": 0.09604217112064362,
883
+ "learning_rate": 1.6508735150244586e-05,
884
+ "loss": 0.0638,
885
+ "step": 1250
886
+ },
887
+ {
888
+ "epoch": 0.8805031446540881,
889
+ "grad_norm": 9.125068664550781,
890
+ "learning_rate": 1.6480782669461916e-05,
891
+ "loss": 0.0744,
892
+ "step": 1260
893
+ },
894
+ {
895
+ "epoch": 0.8874912648497554,
896
+ "grad_norm": 0.5258679986000061,
897
+ "learning_rate": 1.6452830188679247e-05,
898
+ "loss": 0.019,
899
+ "step": 1270
900
+ },
901
+ {
902
+ "epoch": 0.8944793850454228,
903
+ "grad_norm": 3.6146976947784424,
904
+ "learning_rate": 1.6424877707896577e-05,
905
+ "loss": 0.0141,
906
+ "step": 1280
907
+ },
908
+ {
909
+ "epoch": 0.9014675052410901,
910
+ "grad_norm": 0.05512338504195213,
911
+ "learning_rate": 1.6396925227113908e-05,
912
+ "loss": 0.0138,
913
+ "step": 1290
914
+ },
915
+ {
916
+ "epoch": 0.9084556254367575,
917
+ "grad_norm": 0.028468603268265724,
918
+ "learning_rate": 1.636897274633124e-05,
919
+ "loss": 0.0815,
920
+ "step": 1300
921
+ },
922
+ {
923
+ "epoch": 0.9154437456324249,
924
+ "grad_norm": 0.7173348665237427,
925
+ "learning_rate": 1.634102026554857e-05,
926
+ "loss": 0.0919,
927
+ "step": 1310
928
+ },
929
+ {
930
+ "epoch": 0.9224318658280922,
931
+ "grad_norm": 0.06138516217470169,
932
+ "learning_rate": 1.63130677847659e-05,
933
+ "loss": 0.1383,
934
+ "step": 1320
935
+ },
936
+ {
937
+ "epoch": 0.9294199860237596,
938
+ "grad_norm": 10.367135047912598,
939
+ "learning_rate": 1.628511530398323e-05,
940
+ "loss": 0.1265,
941
+ "step": 1330
942
+ },
943
+ {
944
+ "epoch": 0.9364081062194269,
945
+ "grad_norm": 0.3321629762649536,
946
+ "learning_rate": 1.625716282320056e-05,
947
+ "loss": 0.1149,
948
+ "step": 1340
949
+ },
950
+ {
951
+ "epoch": 0.9433962264150944,
952
+ "grad_norm": 0.21935391426086426,
953
+ "learning_rate": 1.622921034241789e-05,
954
+ "loss": 0.1147,
955
+ "step": 1350
956
+ },
957
+ {
958
+ "epoch": 0.9503843466107617,
959
+ "grad_norm": 4.630390644073486,
960
+ "learning_rate": 1.6201257861635222e-05,
961
+ "loss": 0.1159,
962
+ "step": 1360
963
+ },
964
+ {
965
+ "epoch": 0.9573724668064291,
966
+ "grad_norm": 0.03748277202248573,
967
+ "learning_rate": 1.6173305380852552e-05,
968
+ "loss": 0.0029,
969
+ "step": 1370
970
+ },
971
+ {
972
+ "epoch": 0.9643605870020965,
973
+ "grad_norm": 0.09677435457706451,
974
+ "learning_rate": 1.6145352900069883e-05,
975
+ "loss": 0.179,
976
+ "step": 1380
977
+ },
978
+ {
979
+ "epoch": 0.9713487071977638,
980
+ "grad_norm": 0.022979794070124626,
981
+ "learning_rate": 1.6117400419287214e-05,
982
+ "loss": 0.0034,
983
+ "step": 1390
984
+ },
985
+ {
986
+ "epoch": 0.9783368273934312,
987
+ "grad_norm": 3.3972229957580566,
988
+ "learning_rate": 1.6089447938504544e-05,
989
+ "loss": 0.1633,
990
+ "step": 1400
991
+ },
992
+ {
993
+ "epoch": 0.9853249475890985,
994
+ "grad_norm": 0.0410487987101078,
995
+ "learning_rate": 1.6061495457721875e-05,
996
+ "loss": 0.0758,
997
+ "step": 1410
998
+ },
999
+ {
1000
+ "epoch": 0.9923130677847659,
1001
+ "grad_norm": 0.1519143134355545,
1002
+ "learning_rate": 1.6033542976939205e-05,
1003
+ "loss": 0.0967,
1004
+ "step": 1420
1005
+ },
1006
+ {
1007
+ "epoch": 0.9993011879804332,
1008
+ "grad_norm": 0.11000066250562668,
1009
+ "learning_rate": 1.6005590496156536e-05,
1010
+ "loss": 0.1054,
1011
+ "step": 1430
1012
+ }
1013
+ ],
1014
+ "logging_steps": 10,
1015
+ "max_steps": 7155,
1016
+ "num_input_tokens_seen": 0,
1017
+ "num_train_epochs": 5,
1018
+ "save_steps": 500,
1019
+ "stateful_callbacks": {
1020
+ "TrainerControl": {
1021
+ "args": {
1022
+ "should_epoch_stop": false,
1023
+ "should_evaluate": false,
1024
+ "should_log": false,
1025
+ "should_save": true,
1026
+ "should_training_stop": false
1027
+ },
1028
+ "attributes": {}
1029
+ }
1030
+ },
1031
+ "total_flos": 1546911976704000.0,
1032
+ "train_batch_size": 8,
1033
+ "trial_name": null,
1034
+ "trial_params": null
1035
+ }
checkpoint-1431/training_args.bin ADDED
@@ -0,0 +1,3 @@
 
 
 
 
1
+ version https://git-lfs.github.com/spec/v1
2
+ oid sha256:444dc926ee87ace38749c0cb481134ef6ff50da3debedc51baf9fb097d31d73d
3
+ size 5841
checkpoint-2862/README.md ADDED
@@ -0,0 +1,206 @@
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
1
+ ---
2
+ base_model: protectai/deberta-v3-base-prompt-injection-v2
3
+ library_name: peft
4
+ tags:
5
+ - base_model:adapter:protectai/deberta-v3-base-prompt-injection-v2
6
+ - lora
7
+ - transformers
8
+ ---
9
+
10
+ # Model Card for Model ID
11
+
12
+ <!-- Provide a quick summary of what the model is/does. -->
13
+
14
+
15
+
16
+ ## Model Details
17
+
18
+ ### Model Description
19
+
20
+ <!-- Provide a longer summary of what this model is. -->
21
+
22
+
23
+
24
+ - **Developed by:** [More Information Needed]
25
+ - **Funded by [optional]:** [More Information Needed]
26
+ - **Shared by [optional]:** [More Information Needed]
27
+ - **Model type:** [More Information Needed]
28
+ - **Language(s) (NLP):** [More Information Needed]
29
+ - **License:** [More Information Needed]
30
+ - **Finetuned from model [optional]:** [More Information Needed]
31
+
32
+ ### Model Sources [optional]
33
+
34
+ <!-- Provide the basic links for the model. -->
35
+
36
+ - **Repository:** [More Information Needed]
37
+ - **Paper [optional]:** [More Information Needed]
38
+ - **Demo [optional]:** [More Information Needed]
39
+
40
+ ## Uses
41
+
42
+ <!-- Address questions around how the model is intended to be used, including the foreseeable users of the model and those affected by the model. -->
43
+
44
+ ### Direct Use
45
+
46
+ <!-- This section is for the model use without fine-tuning or plugging into a larger ecosystem/app. -->
47
+
48
+ [More Information Needed]
49
+
50
+ ### Downstream Use [optional]
51
+
52
+ <!-- This section is for the model use when fine-tuned for a task, or when plugged into a larger ecosystem/app -->
53
+
54
+ [More Information Needed]
55
+
56
+ ### Out-of-Scope Use
57
+
58
+ <!-- This section addresses misuse, malicious use, and uses that the model will not work well for. -->
59
+
60
+ [More Information Needed]
61
+
62
+ ## Bias, Risks, and Limitations
63
+
64
+ <!-- This section is meant to convey both technical and sociotechnical limitations. -->
65
+
66
+ [More Information Needed]
67
+
68
+ ### Recommendations
69
+
70
+ <!-- This section is meant to convey recommendations with respect to the bias, risk, and technical limitations. -->
71
+
72
+ Users (both direct and downstream) should be made aware of the risks, biases and limitations of the model. More information needed for further recommendations.
73
+
74
+ ## How to Get Started with the Model
75
+
76
+ Use the code below to get started with the model.
77
+
78
+ [More Information Needed]
79
+
80
+ ## Training Details
81
+
82
+ ### Training Data
83
+
84
+ <!-- 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. -->
85
+
86
+ [More Information Needed]
87
+
88
+ ### Training Procedure
89
+
90
+ <!-- This relates heavily to the Technical Specifications. Content here should link to that section when it is relevant to the training procedure. -->
91
+
92
+ #### Preprocessing [optional]
93
+
94
+ [More Information Needed]
95
+
96
+
97
+ #### Training Hyperparameters
98
+
99
+ - **Training regime:** [More Information Needed] <!--fp32, fp16 mixed precision, bf16 mixed precision, bf16 non-mixed precision, fp16 non-mixed precision, fp8 mixed precision -->
100
+
101
+ #### Speeds, Sizes, Times [optional]
102
+
103
+ <!-- This section provides information about throughput, start/end time, checkpoint size if relevant, etc. -->
104
+
105
+ [More Information Needed]
106
+
107
+ ## Evaluation
108
+
109
+ <!-- This section describes the evaluation protocols and provides the results. -->
110
+
111
+ ### Testing Data, Factors & Metrics
112
+
113
+ #### Testing Data
114
+
115
+ <!-- This should link to a Dataset Card if possible. -->
116
+
117
+ [More Information Needed]
118
+
119
+ #### Factors
120
+
121
+ <!-- These are the things the evaluation is disaggregating by, e.g., subpopulations or domains. -->
122
+
123
+ [More Information Needed]
124
+
125
+ #### Metrics
126
+
127
+ <!-- These are the evaluation metrics being used, ideally with a description of why. -->
128
+
129
+ [More Information Needed]
130
+
131
+ ### Results
132
+
133
+ [More Information Needed]
134
+
135
+ #### Summary
136
+
137
+
138
+
139
+ ## Model Examination [optional]
140
+
141
+ <!-- Relevant interpretability work for the model goes here -->
142
+
143
+ [More Information Needed]
144
+
145
+ ## Environmental Impact
146
+
147
+ <!-- Total emissions (in grams of CO2eq) and additional considerations, such as electricity usage, go here. Edit the suggested text below accordingly -->
148
+
149
+ 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).
150
+
151
+ - **Hardware Type:** [More Information Needed]
152
+ - **Hours used:** [More Information Needed]
153
+ - **Cloud Provider:** [More Information Needed]
154
+ - **Compute Region:** [More Information Needed]
155
+ - **Carbon Emitted:** [More Information Needed]
156
+
157
+ ## Technical Specifications [optional]
158
+
159
+ ### Model Architecture and Objective
160
+
161
+ [More Information Needed]
162
+
163
+ ### Compute Infrastructure
164
+
165
+ [More Information Needed]
166
+
167
+ #### Hardware
168
+
169
+ [More Information Needed]
170
+
171
+ #### Software
172
+
173
+ [More Information Needed]
174
+
175
+ ## Citation [optional]
176
+
177
+ <!-- If there is a paper or blog post introducing the model, the APA and Bibtex information for that should go in this section. -->
178
+
179
+ **BibTeX:**
180
+
181
+ [More Information Needed]
182
+
183
+ **APA:**
184
+
185
+ [More Information Needed]
186
+
187
+ ## Glossary [optional]
188
+
189
+ <!-- If relevant, include terms and calculations in this section that can help readers understand the model or model card. -->
190
+
191
+ [More Information Needed]
192
+
193
+ ## More Information [optional]
194
+
195
+ [More Information Needed]
196
+
197
+ ## Model Card Authors [optional]
198
+
199
+ [More Information Needed]
200
+
201
+ ## Model Card Contact
202
+
203
+ [More Information Needed]
204
+ ### Framework versions
205
+
206
+ - PEFT 0.18.1
checkpoint-2862/adapter_config.json ADDED
@@ -0,0 +1,44 @@
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
1
+ {
2
+ "alora_invocation_tokens": null,
3
+ "alpha_pattern": {},
4
+ "arrow_config": null,
5
+ "auto_mapping": null,
6
+ "base_model_name_or_path": "protectai/deberta-v3-base-prompt-injection-v2",
7
+ "bias": "none",
8
+ "corda_config": null,
9
+ "ensure_weight_tying": false,
10
+ "eva_config": null,
11
+ "exclude_modules": null,
12
+ "fan_in_fan_out": false,
13
+ "inference_mode": true,
14
+ "init_lora_weights": true,
15
+ "layer_replication": null,
16
+ "layers_pattern": null,
17
+ "layers_to_transform": null,
18
+ "loftq_config": {},
19
+ "lora_alpha": 64,
20
+ "lora_bias": false,
21
+ "lora_dropout": 0.05,
22
+ "megatron_config": null,
23
+ "megatron_core": "megatron.core",
24
+ "modules_to_save": [
25
+ "classifier",
26
+ "score"
27
+ ],
28
+ "peft_type": "LORA",
29
+ "peft_version": "0.18.1",
30
+ "qalora_group_size": 16,
31
+ "r": 64,
32
+ "rank_pattern": {},
33
+ "revision": null,
34
+ "target_modules": [
35
+ "query_proj",
36
+ "value_proj"
37
+ ],
38
+ "target_parameters": null,
39
+ "task_type": "SEQ_CLS",
40
+ "trainable_token_indices": null,
41
+ "use_dora": false,
42
+ "use_qalora": false,
43
+ "use_rslora": false
44
+ }
checkpoint-2862/adapter_model.safetensors ADDED
@@ -0,0 +1,3 @@
 
 
 
 
1
+ version https://git-lfs.github.com/spec/v1
2
+ oid sha256:5ba27a81e489d1c9f7fc94ca2fae77756942bb45fa988e57ef96e06e59778f5b
3
+ size 9450704
checkpoint-2862/optimizer.pt ADDED
@@ -0,0 +1,3 @@
 
 
 
 
1
+ version https://git-lfs.github.com/spec/v1
2
+ oid sha256:bfec53197742965e80a639cbb49e2928fe5ed7de959595e776c99aee63fc4605
3
+ size 18930123
checkpoint-2862/rng_state.pth ADDED
@@ -0,0 +1,3 @@
 
 
 
 
1
+ version https://git-lfs.github.com/spec/v1
2
+ oid sha256:7fe361375480dcf5d55962cc7085bbdaedbd2df0bbc42f8b0a89251baf711010
3
+ size 14645
checkpoint-2862/scaler.pt ADDED
@@ -0,0 +1,3 @@
 
 
 
 
1
+ version https://git-lfs.github.com/spec/v1
2
+ oid sha256:58a6bd3b4955b3049cd29c9d4331ee092784f606065ffda56f824b622f03aa91
3
+ size 1383
checkpoint-2862/scheduler.pt ADDED
@@ -0,0 +1,3 @@
 
 
 
 
1
+ version https://git-lfs.github.com/spec/v1
2
+ oid sha256:6c1b8a4b1b027b2ca73fdf5d8f75852777b59c27edd2de44bfb85fe1d2bb6a5e
3
+ size 1465
checkpoint-2862/special_tokens_map.json ADDED
@@ -0,0 +1,51 @@
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
1
+ {
2
+ "bos_token": {
3
+ "content": "[CLS]",
4
+ "lstrip": false,
5
+ "normalized": false,
6
+ "rstrip": false,
7
+ "single_word": false
8
+ },
9
+ "cls_token": {
10
+ "content": "[CLS]",
11
+ "lstrip": false,
12
+ "normalized": false,
13
+ "rstrip": false,
14
+ "single_word": false
15
+ },
16
+ "eos_token": {
17
+ "content": "[SEP]",
18
+ "lstrip": false,
19
+ "normalized": false,
20
+ "rstrip": false,
21
+ "single_word": false
22
+ },
23
+ "mask_token": {
24
+ "content": "[MASK]",
25
+ "lstrip": false,
26
+ "normalized": false,
27
+ "rstrip": false,
28
+ "single_word": false
29
+ },
30
+ "pad_token": {
31
+ "content": "[PAD]",
32
+ "lstrip": false,
33
+ "normalized": false,
34
+ "rstrip": false,
35
+ "single_word": false
36
+ },
37
+ "sep_token": {
38
+ "content": "[SEP]",
39
+ "lstrip": false,
40
+ "normalized": false,
41
+ "rstrip": false,
42
+ "single_word": false
43
+ },
44
+ "unk_token": {
45
+ "content": "[UNK]",
46
+ "lstrip": false,
47
+ "normalized": true,
48
+ "rstrip": false,
49
+ "single_word": false
50
+ }
51
+ }
checkpoint-2862/tokenizer.json ADDED
The diff for this file is too large to render. See raw diff
 
checkpoint-2862/tokenizer_config.json ADDED
@@ -0,0 +1,63 @@
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
1
+ {
2
+ "added_tokens_decoder": {
3
+ "0": {
4
+ "content": "[PAD]",
5
+ "lstrip": false,
6
+ "normalized": false,
7
+ "rstrip": false,
8
+ "single_word": false,
9
+ "special": true
10
+ },
11
+ "1": {
12
+ "content": "[CLS]",
13
+ "lstrip": false,
14
+ "normalized": false,
15
+ "rstrip": false,
16
+ "single_word": false,
17
+ "special": true
18
+ },
19
+ "2": {
20
+ "content": "[SEP]",
21
+ "lstrip": false,
22
+ "normalized": false,
23
+ "rstrip": false,
24
+ "single_word": false,
25
+ "special": true
26
+ },
27
+ "3": {
28
+ "content": "[UNK]",
29
+ "lstrip": false,
30
+ "normalized": true,
31
+ "rstrip": false,
32
+ "single_word": false,
33
+ "special": true
34
+ },
35
+ "128000": {
36
+ "content": "[MASK]",
37
+ "lstrip": false,
38
+ "normalized": false,
39
+ "rstrip": false,
40
+ "single_word": false,
41
+ "special": true
42
+ }
43
+ },
44
+ "bos_token": "[CLS]",
45
+ "clean_up_tokenization_spaces": true,
46
+ "cls_token": "[CLS]",
47
+ "do_lower_case": false,
48
+ "eos_token": "[SEP]",
49
+ "extra_special_tokens": {},
50
+ "mask_token": "[MASK]",
51
+ "max_length": 512,
52
+ "model_max_length": 1000000000000000019884624838656,
53
+ "pad_token": "[PAD]",
54
+ "sep_token": "[SEP]",
55
+ "sp_model_kwargs": {},
56
+ "split_by_punct": false,
57
+ "stride": 0,
58
+ "tokenizer_class": "DebertaV2Tokenizer",
59
+ "truncation_side": "right",
60
+ "truncation_strategy": "longest_first",
61
+ "unk_token": "[UNK]",
62
+ "vocab_type": "spm"
63
+ }
checkpoint-2862/trainer_state.json ADDED
@@ -0,0 +1,2036 @@
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
1
+ {
2
+ "best_global_step": null,
3
+ "best_metric": null,
4
+ "best_model_checkpoint": null,
5
+ "epoch": 2.0,
6
+ "eval_steps": 500,
7
+ "global_step": 2862,
8
+ "is_hyper_param_search": false,
9
+ "is_local_process_zero": true,
10
+ "is_world_process_zero": true,
11
+ "log_history": [
12
+ {
13
+ "epoch": 0.0069881201956673656,
14
+ "grad_norm": 14.528067588806152,
15
+ "learning_rate": 1.99748427672956e-05,
16
+ "loss": 4.6827,
17
+ "step": 10
18
+ },
19
+ {
20
+ "epoch": 0.013976240391334731,
21
+ "grad_norm": 13.050820350646973,
22
+ "learning_rate": 1.994689028651293e-05,
23
+ "loss": 4.529,
24
+ "step": 20
25
+ },
26
+ {
27
+ "epoch": 0.020964360587002098,
28
+ "grad_norm": 2.765507698059082,
29
+ "learning_rate": 1.991893780573026e-05,
30
+ "loss": 3.4294,
31
+ "step": 30
32
+ },
33
+ {
34
+ "epoch": 0.027952480782669462,
35
+ "grad_norm": 16.95716667175293,
36
+ "learning_rate": 1.989098532494759e-05,
37
+ "loss": 3.352,
38
+ "step": 40
39
+ },
40
+ {
41
+ "epoch": 0.03494060097833683,
42
+ "grad_norm": 5.339346408843994,
43
+ "learning_rate": 1.986303284416492e-05,
44
+ "loss": 2.2654,
45
+ "step": 50
46
+ },
47
+ {
48
+ "epoch": 0.041928721174004195,
49
+ "grad_norm": 7.168309688568115,
50
+ "learning_rate": 1.983508036338225e-05,
51
+ "loss": 2.0076,
52
+ "step": 60
53
+ },
54
+ {
55
+ "epoch": 0.04891684136967156,
56
+ "grad_norm": 5.793034076690674,
57
+ "learning_rate": 1.9807127882599582e-05,
58
+ "loss": 1.6779,
59
+ "step": 70
60
+ },
61
+ {
62
+ "epoch": 0.055904961565338925,
63
+ "grad_norm": 6.093694686889648,
64
+ "learning_rate": 1.9779175401816913e-05,
65
+ "loss": 1.2564,
66
+ "step": 80
67
+ },
68
+ {
69
+ "epoch": 0.06289308176100629,
70
+ "grad_norm": 3.321807622909546,
71
+ "learning_rate": 1.9751222921034243e-05,
72
+ "loss": 1.4143,
73
+ "step": 90
74
+ },
75
+ {
76
+ "epoch": 0.06988120195667366,
77
+ "grad_norm": 2.9472098350524902,
78
+ "learning_rate": 1.9723270440251574e-05,
79
+ "loss": 1.0533,
80
+ "step": 100
81
+ },
82
+ {
83
+ "epoch": 0.07686932215234102,
84
+ "grad_norm": 8.515447616577148,
85
+ "learning_rate": 1.9695317959468904e-05,
86
+ "loss": 1.2895,
87
+ "step": 110
88
+ },
89
+ {
90
+ "epoch": 0.08385744234800839,
91
+ "grad_norm": 3.560924530029297,
92
+ "learning_rate": 1.9667365478686235e-05,
93
+ "loss": 1.097,
94
+ "step": 120
95
+ },
96
+ {
97
+ "epoch": 0.09084556254367575,
98
+ "grad_norm": 4.986349582672119,
99
+ "learning_rate": 1.9639412997903566e-05,
100
+ "loss": 0.9556,
101
+ "step": 130
102
+ },
103
+ {
104
+ "epoch": 0.09783368273934312,
105
+ "grad_norm": 1.27532160282135,
106
+ "learning_rate": 1.9611460517120896e-05,
107
+ "loss": 0.8306,
108
+ "step": 140
109
+ },
110
+ {
111
+ "epoch": 0.10482180293501048,
112
+ "grad_norm": 2.6796844005584717,
113
+ "learning_rate": 1.9583508036338227e-05,
114
+ "loss": 0.7747,
115
+ "step": 150
116
+ },
117
+ {
118
+ "epoch": 0.11180992313067785,
119
+ "grad_norm": 3.4592413902282715,
120
+ "learning_rate": 1.9555555555555557e-05,
121
+ "loss": 0.6187,
122
+ "step": 160
123
+ },
124
+ {
125
+ "epoch": 0.1187980433263452,
126
+ "grad_norm": 4.833241939544678,
127
+ "learning_rate": 1.9527603074772888e-05,
128
+ "loss": 0.6831,
129
+ "step": 170
130
+ },
131
+ {
132
+ "epoch": 0.12578616352201258,
133
+ "grad_norm": 1.7203434705734253,
134
+ "learning_rate": 1.949965059399022e-05,
135
+ "loss": 0.586,
136
+ "step": 180
137
+ },
138
+ {
139
+ "epoch": 0.13277428371767994,
140
+ "grad_norm": 1.333950161933899,
141
+ "learning_rate": 1.947169811320755e-05,
142
+ "loss": 0.6214,
143
+ "step": 190
144
+ },
145
+ {
146
+ "epoch": 0.13976240391334732,
147
+ "grad_norm": 1.5078257322311401,
148
+ "learning_rate": 1.944374563242488e-05,
149
+ "loss": 0.5652,
150
+ "step": 200
151
+ },
152
+ {
153
+ "epoch": 0.14675052410901468,
154
+ "grad_norm": 3.7645132541656494,
155
+ "learning_rate": 1.941579315164221e-05,
156
+ "loss": 0.5229,
157
+ "step": 210
158
+ },
159
+ {
160
+ "epoch": 0.15373864430468204,
161
+ "grad_norm": 1.2441000938415527,
162
+ "learning_rate": 1.938784067085954e-05,
163
+ "loss": 0.3669,
164
+ "step": 220
165
+ },
166
+ {
167
+ "epoch": 0.1607267645003494,
168
+ "grad_norm": 2.2875142097473145,
169
+ "learning_rate": 1.935988819007687e-05,
170
+ "loss": 0.4162,
171
+ "step": 230
172
+ },
173
+ {
174
+ "epoch": 0.16771488469601678,
175
+ "grad_norm": 2.1816585063934326,
176
+ "learning_rate": 1.9331935709294202e-05,
177
+ "loss": 0.3836,
178
+ "step": 240
179
+ },
180
+ {
181
+ "epoch": 0.17470300489168414,
182
+ "grad_norm": 3.4903249740600586,
183
+ "learning_rate": 1.9303983228511532e-05,
184
+ "loss": 0.4131,
185
+ "step": 250
186
+ },
187
+ {
188
+ "epoch": 0.1816911250873515,
189
+ "grad_norm": 2.1905159950256348,
190
+ "learning_rate": 1.9276030747728863e-05,
191
+ "loss": 0.3725,
192
+ "step": 260
193
+ },
194
+ {
195
+ "epoch": 0.18867924528301888,
196
+ "grad_norm": 3.6003665924072266,
197
+ "learning_rate": 1.9248078266946194e-05,
198
+ "loss": 0.4131,
199
+ "step": 270
200
+ },
201
+ {
202
+ "epoch": 0.19566736547868624,
203
+ "grad_norm": 3.012568235397339,
204
+ "learning_rate": 1.9220125786163524e-05,
205
+ "loss": 0.3991,
206
+ "step": 280
207
+ },
208
+ {
209
+ "epoch": 0.2026554856743536,
210
+ "grad_norm": 1.2049503326416016,
211
+ "learning_rate": 1.9192173305380855e-05,
212
+ "loss": 0.2732,
213
+ "step": 290
214
+ },
215
+ {
216
+ "epoch": 0.20964360587002095,
217
+ "grad_norm": 2.0943682193756104,
218
+ "learning_rate": 1.9164220824598185e-05,
219
+ "loss": 0.3052,
220
+ "step": 300
221
+ },
222
+ {
223
+ "epoch": 0.21663172606568834,
224
+ "grad_norm": 4.1970086097717285,
225
+ "learning_rate": 1.9136268343815516e-05,
226
+ "loss": 0.2544,
227
+ "step": 310
228
+ },
229
+ {
230
+ "epoch": 0.2236198462613557,
231
+ "grad_norm": 2.6826064586639404,
232
+ "learning_rate": 1.9108315863032846e-05,
233
+ "loss": 0.1921,
234
+ "step": 320
235
+ },
236
+ {
237
+ "epoch": 0.23060796645702306,
238
+ "grad_norm": 6.013642311096191,
239
+ "learning_rate": 1.9080363382250177e-05,
240
+ "loss": 0.3014,
241
+ "step": 330
242
+ },
243
+ {
244
+ "epoch": 0.2375960866526904,
245
+ "grad_norm": 4.2075629234313965,
246
+ "learning_rate": 1.9052410901467508e-05,
247
+ "loss": 0.2569,
248
+ "step": 340
249
+ },
250
+ {
251
+ "epoch": 0.2445842068483578,
252
+ "grad_norm": 0.5800023078918457,
253
+ "learning_rate": 1.9024458420684838e-05,
254
+ "loss": 0.1486,
255
+ "step": 350
256
+ },
257
+ {
258
+ "epoch": 0.25157232704402516,
259
+ "grad_norm": 8.123499870300293,
260
+ "learning_rate": 1.899650593990217e-05,
261
+ "loss": 0.2215,
262
+ "step": 360
263
+ },
264
+ {
265
+ "epoch": 0.2585604472396925,
266
+ "grad_norm": 4.439255237579346,
267
+ "learning_rate": 1.89685534591195e-05,
268
+ "loss": 0.1692,
269
+ "step": 370
270
+ },
271
+ {
272
+ "epoch": 0.2655485674353599,
273
+ "grad_norm": 4.177631855010986,
274
+ "learning_rate": 1.894060097833683e-05,
275
+ "loss": 0.1176,
276
+ "step": 380
277
+ },
278
+ {
279
+ "epoch": 0.27253668763102723,
280
+ "grad_norm": 0.5754140019416809,
281
+ "learning_rate": 1.891264849755416e-05,
282
+ "loss": 0.1538,
283
+ "step": 390
284
+ },
285
+ {
286
+ "epoch": 0.27952480782669464,
287
+ "grad_norm": 0.5612320303916931,
288
+ "learning_rate": 1.888469601677149e-05,
289
+ "loss": 0.1054,
290
+ "step": 400
291
+ },
292
+ {
293
+ "epoch": 0.286512928022362,
294
+ "grad_norm": 1.949729561805725,
295
+ "learning_rate": 1.8856743535988818e-05,
296
+ "loss": 0.0824,
297
+ "step": 410
298
+ },
299
+ {
300
+ "epoch": 0.29350104821802936,
301
+ "grad_norm": 3.153822898864746,
302
+ "learning_rate": 1.8828791055206152e-05,
303
+ "loss": 0.1174,
304
+ "step": 420
305
+ },
306
+ {
307
+ "epoch": 0.3004891684136967,
308
+ "grad_norm": 0.6545250415802002,
309
+ "learning_rate": 1.8800838574423483e-05,
310
+ "loss": 0.1305,
311
+ "step": 430
312
+ },
313
+ {
314
+ "epoch": 0.3074772886093641,
315
+ "grad_norm": 1.2621614933013916,
316
+ "learning_rate": 1.8772886093640813e-05,
317
+ "loss": 0.0684,
318
+ "step": 440
319
+ },
320
+ {
321
+ "epoch": 0.31446540880503143,
322
+ "grad_norm": 0.3028416335582733,
323
+ "learning_rate": 1.8744933612858144e-05,
324
+ "loss": 0.0863,
325
+ "step": 450
326
+ },
327
+ {
328
+ "epoch": 0.3214535290006988,
329
+ "grad_norm": 0.4452091455459595,
330
+ "learning_rate": 1.8716981132075474e-05,
331
+ "loss": 0.1016,
332
+ "step": 460
333
+ },
334
+ {
335
+ "epoch": 0.3284416491963662,
336
+ "grad_norm": 0.4046468436717987,
337
+ "learning_rate": 1.8689028651292805e-05,
338
+ "loss": 0.2317,
339
+ "step": 470
340
+ },
341
+ {
342
+ "epoch": 0.33542976939203356,
343
+ "grad_norm": 0.5121694803237915,
344
+ "learning_rate": 1.8661076170510135e-05,
345
+ "loss": 0.1965,
346
+ "step": 480
347
+ },
348
+ {
349
+ "epoch": 0.3424178895877009,
350
+ "grad_norm": 2.7892794609069824,
351
+ "learning_rate": 1.8633123689727466e-05,
352
+ "loss": 0.2082,
353
+ "step": 490
354
+ },
355
+ {
356
+ "epoch": 0.3494060097833683,
357
+ "grad_norm": 0.29012227058410645,
358
+ "learning_rate": 1.8605171208944793e-05,
359
+ "loss": 0.1393,
360
+ "step": 500
361
+ },
362
+ {
363
+ "epoch": 0.35639412997903563,
364
+ "grad_norm": 0.14969919621944427,
365
+ "learning_rate": 1.8577218728162124e-05,
366
+ "loss": 0.1917,
367
+ "step": 510
368
+ },
369
+ {
370
+ "epoch": 0.363382250174703,
371
+ "grad_norm": 2.5301618576049805,
372
+ "learning_rate": 1.8549266247379458e-05,
373
+ "loss": 0.0939,
374
+ "step": 520
375
+ },
376
+ {
377
+ "epoch": 0.37037037037037035,
378
+ "grad_norm": 0.5171930193901062,
379
+ "learning_rate": 1.8521313766596788e-05,
380
+ "loss": 0.0643,
381
+ "step": 530
382
+ },
383
+ {
384
+ "epoch": 0.37735849056603776,
385
+ "grad_norm": 0.23746764659881592,
386
+ "learning_rate": 1.849336128581412e-05,
387
+ "loss": 0.2094,
388
+ "step": 540
389
+ },
390
+ {
391
+ "epoch": 0.3843466107617051,
392
+ "grad_norm": 1.2422730922698975,
393
+ "learning_rate": 1.846540880503145e-05,
394
+ "loss": 0.109,
395
+ "step": 550
396
+ },
397
+ {
398
+ "epoch": 0.3913347309573725,
399
+ "grad_norm": 0.8466539978981018,
400
+ "learning_rate": 1.843745632424878e-05,
401
+ "loss": 0.1514,
402
+ "step": 560
403
+ },
404
+ {
405
+ "epoch": 0.39832285115303984,
406
+ "grad_norm": 0.201154425740242,
407
+ "learning_rate": 1.840950384346611e-05,
408
+ "loss": 0.1597,
409
+ "step": 570
410
+ },
411
+ {
412
+ "epoch": 0.4053109713487072,
413
+ "grad_norm": 0.19054199755191803,
414
+ "learning_rate": 1.838155136268344e-05,
415
+ "loss": 0.0955,
416
+ "step": 580
417
+ },
418
+ {
419
+ "epoch": 0.41229909154437455,
420
+ "grad_norm": 1.8036576509475708,
421
+ "learning_rate": 1.8353598881900768e-05,
422
+ "loss": 0.1369,
423
+ "step": 590
424
+ },
425
+ {
426
+ "epoch": 0.4192872117400419,
427
+ "grad_norm": 0.3122437596321106,
428
+ "learning_rate": 1.83256464011181e-05,
429
+ "loss": 0.1439,
430
+ "step": 600
431
+ },
432
+ {
433
+ "epoch": 0.42627533193570927,
434
+ "grad_norm": 2.6353561878204346,
435
+ "learning_rate": 1.829769392033543e-05,
436
+ "loss": 0.1566,
437
+ "step": 610
438
+ },
439
+ {
440
+ "epoch": 0.4332634521313767,
441
+ "grad_norm": 3.645707368850708,
442
+ "learning_rate": 1.8269741439552763e-05,
443
+ "loss": 0.2505,
444
+ "step": 620
445
+ },
446
+ {
447
+ "epoch": 0.44025157232704404,
448
+ "grad_norm": 0.20553793013095856,
449
+ "learning_rate": 1.8241788958770094e-05,
450
+ "loss": 0.0509,
451
+ "step": 630
452
+ },
453
+ {
454
+ "epoch": 0.4472396925227114,
455
+ "grad_norm": 3.8460299968719482,
456
+ "learning_rate": 1.8213836477987425e-05,
457
+ "loss": 0.118,
458
+ "step": 640
459
+ },
460
+ {
461
+ "epoch": 0.45422781271837875,
462
+ "grad_norm": 0.4825759828090668,
463
+ "learning_rate": 1.8185883997204755e-05,
464
+ "loss": 0.0671,
465
+ "step": 650
466
+ },
467
+ {
468
+ "epoch": 0.4612159329140461,
469
+ "grad_norm": 2.5147883892059326,
470
+ "learning_rate": 1.8157931516422086e-05,
471
+ "loss": 0.1381,
472
+ "step": 660
473
+ },
474
+ {
475
+ "epoch": 0.46820405310971347,
476
+ "grad_norm": 4.72428035736084,
477
+ "learning_rate": 1.8129979035639413e-05,
478
+ "loss": 0.1093,
479
+ "step": 670
480
+ },
481
+ {
482
+ "epoch": 0.4751921733053808,
483
+ "grad_norm": 3.091625928878784,
484
+ "learning_rate": 1.8102026554856743e-05,
485
+ "loss": 0.2363,
486
+ "step": 680
487
+ },
488
+ {
489
+ "epoch": 0.48218029350104824,
490
+ "grad_norm": 0.10304142534732819,
491
+ "learning_rate": 1.8074074074074074e-05,
492
+ "loss": 0.1347,
493
+ "step": 690
494
+ },
495
+ {
496
+ "epoch": 0.4891684136967156,
497
+ "grad_norm": 0.4455307424068451,
498
+ "learning_rate": 1.8046121593291405e-05,
499
+ "loss": 0.1064,
500
+ "step": 700
501
+ },
502
+ {
503
+ "epoch": 0.49615653389238296,
504
+ "grad_norm": 0.11594414710998535,
505
+ "learning_rate": 1.8018169112508735e-05,
506
+ "loss": 0.123,
507
+ "step": 710
508
+ },
509
+ {
510
+ "epoch": 0.5031446540880503,
511
+ "grad_norm": 0.1224653497338295,
512
+ "learning_rate": 1.799021663172607e-05,
513
+ "loss": 0.1263,
514
+ "step": 720
515
+ },
516
+ {
517
+ "epoch": 0.5101327742837177,
518
+ "grad_norm": 4.951889991760254,
519
+ "learning_rate": 1.79622641509434e-05,
520
+ "loss": 0.2111,
521
+ "step": 730
522
+ },
523
+ {
524
+ "epoch": 0.517120894479385,
525
+ "grad_norm": 0.16316205263137817,
526
+ "learning_rate": 1.793431167016073e-05,
527
+ "loss": 0.0366,
528
+ "step": 740
529
+ },
530
+ {
531
+ "epoch": 0.5241090146750524,
532
+ "grad_norm": 0.1896398663520813,
533
+ "learning_rate": 1.790635918937806e-05,
534
+ "loss": 0.1656,
535
+ "step": 750
536
+ },
537
+ {
538
+ "epoch": 0.5310971348707197,
539
+ "grad_norm": 0.3920069932937622,
540
+ "learning_rate": 1.7878406708595388e-05,
541
+ "loss": 0.2804,
542
+ "step": 760
543
+ },
544
+ {
545
+ "epoch": 0.5380852550663872,
546
+ "grad_norm": 0.12546010315418243,
547
+ "learning_rate": 1.785045422781272e-05,
548
+ "loss": 0.1404,
549
+ "step": 770
550
+ },
551
+ {
552
+ "epoch": 0.5450733752620545,
553
+ "grad_norm": 5.382510185241699,
554
+ "learning_rate": 1.782250174703005e-05,
555
+ "loss": 0.1194,
556
+ "step": 780
557
+ },
558
+ {
559
+ "epoch": 0.5520614954577219,
560
+ "grad_norm": 0.3077394366264343,
561
+ "learning_rate": 1.779454926624738e-05,
562
+ "loss": 0.1744,
563
+ "step": 790
564
+ },
565
+ {
566
+ "epoch": 0.5590496156533893,
567
+ "grad_norm": 0.41757112741470337,
568
+ "learning_rate": 1.776659678546471e-05,
569
+ "loss": 0.0729,
570
+ "step": 800
571
+ },
572
+ {
573
+ "epoch": 0.5660377358490566,
574
+ "grad_norm": 0.17300930619239807,
575
+ "learning_rate": 1.773864430468204e-05,
576
+ "loss": 0.1443,
577
+ "step": 810
578
+ },
579
+ {
580
+ "epoch": 0.573025856044724,
581
+ "grad_norm": 0.09419082850217819,
582
+ "learning_rate": 1.7710691823899375e-05,
583
+ "loss": 0.1587,
584
+ "step": 820
585
+ },
586
+ {
587
+ "epoch": 0.5800139762403913,
588
+ "grad_norm": 0.14113612473011017,
589
+ "learning_rate": 1.7682739343116705e-05,
590
+ "loss": 0.1196,
591
+ "step": 830
592
+ },
593
+ {
594
+ "epoch": 0.5870020964360587,
595
+ "grad_norm": 0.6690120697021484,
596
+ "learning_rate": 1.7654786862334036e-05,
597
+ "loss": 0.0619,
598
+ "step": 840
599
+ },
600
+ {
601
+ "epoch": 0.593990216631726,
602
+ "grad_norm": 0.5076020956039429,
603
+ "learning_rate": 1.7626834381551363e-05,
604
+ "loss": 0.2922,
605
+ "step": 850
606
+ },
607
+ {
608
+ "epoch": 0.6009783368273934,
609
+ "grad_norm": 0.12461690604686737,
610
+ "learning_rate": 1.7598881900768694e-05,
611
+ "loss": 0.0351,
612
+ "step": 860
613
+ },
614
+ {
615
+ "epoch": 0.6079664570230608,
616
+ "grad_norm": 5.236762523651123,
617
+ "learning_rate": 1.7570929419986024e-05,
618
+ "loss": 0.1571,
619
+ "step": 870
620
+ },
621
+ {
622
+ "epoch": 0.6149545772187281,
623
+ "grad_norm": 0.12950463593006134,
624
+ "learning_rate": 1.7542976939203355e-05,
625
+ "loss": 0.0204,
626
+ "step": 880
627
+ },
628
+ {
629
+ "epoch": 0.6219426974143956,
630
+ "grad_norm": 0.08026222884654999,
631
+ "learning_rate": 1.7515024458420685e-05,
632
+ "loss": 0.1185,
633
+ "step": 890
634
+ },
635
+ {
636
+ "epoch": 0.6289308176100629,
637
+ "grad_norm": 4.237085342407227,
638
+ "learning_rate": 1.7487071977638016e-05,
639
+ "loss": 0.0876,
640
+ "step": 900
641
+ },
642
+ {
643
+ "epoch": 0.6359189378057303,
644
+ "grad_norm": 0.916748046875,
645
+ "learning_rate": 1.7459119496855346e-05,
646
+ "loss": 0.0619,
647
+ "step": 910
648
+ },
649
+ {
650
+ "epoch": 0.6429070580013976,
651
+ "grad_norm": 4.51674747467041,
652
+ "learning_rate": 1.743116701607268e-05,
653
+ "loss": 0.1116,
654
+ "step": 920
655
+ },
656
+ {
657
+ "epoch": 0.649895178197065,
658
+ "grad_norm": 1.1469999551773071,
659
+ "learning_rate": 1.740321453529001e-05,
660
+ "loss": 0.0147,
661
+ "step": 930
662
+ },
663
+ {
664
+ "epoch": 0.6568832983927324,
665
+ "grad_norm": 2.3627333641052246,
666
+ "learning_rate": 1.7375262054507338e-05,
667
+ "loss": 0.1485,
668
+ "step": 940
669
+ },
670
+ {
671
+ "epoch": 0.6638714185883997,
672
+ "grad_norm": 8.216973304748535,
673
+ "learning_rate": 1.734730957372467e-05,
674
+ "loss": 0.0472,
675
+ "step": 950
676
+ },
677
+ {
678
+ "epoch": 0.6708595387840671,
679
+ "grad_norm": 3.5901684761047363,
680
+ "learning_rate": 1.7319357092942e-05,
681
+ "loss": 0.0979,
682
+ "step": 960
683
+ },
684
+ {
685
+ "epoch": 0.6778476589797344,
686
+ "grad_norm": 7.015125274658203,
687
+ "learning_rate": 1.729140461215933e-05,
688
+ "loss": 0.0537,
689
+ "step": 970
690
+ },
691
+ {
692
+ "epoch": 0.6848357791754018,
693
+ "grad_norm": 4.372157096862793,
694
+ "learning_rate": 1.726345213137666e-05,
695
+ "loss": 0.1152,
696
+ "step": 980
697
+ },
698
+ {
699
+ "epoch": 0.6918238993710691,
700
+ "grad_norm": 4.862729072570801,
701
+ "learning_rate": 1.723549965059399e-05,
702
+ "loss": 0.1131,
703
+ "step": 990
704
+ },
705
+ {
706
+ "epoch": 0.6988120195667366,
707
+ "grad_norm": 0.07732150703668594,
708
+ "learning_rate": 1.720754716981132e-05,
709
+ "loss": 0.0707,
710
+ "step": 1000
711
+ },
712
+ {
713
+ "epoch": 0.705800139762404,
714
+ "grad_norm": 0.19970574975013733,
715
+ "learning_rate": 1.7179594689028652e-05,
716
+ "loss": 0.0738,
717
+ "step": 1010
718
+ },
719
+ {
720
+ "epoch": 0.7127882599580713,
721
+ "grad_norm": 0.21677608788013458,
722
+ "learning_rate": 1.7151642208245983e-05,
723
+ "loss": 0.0992,
724
+ "step": 1020
725
+ },
726
+ {
727
+ "epoch": 0.7197763801537387,
728
+ "grad_norm": 0.10738188028335571,
729
+ "learning_rate": 1.7123689727463313e-05,
730
+ "loss": 0.0103,
731
+ "step": 1030
732
+ },
733
+ {
734
+ "epoch": 0.726764500349406,
735
+ "grad_norm": 3.7014143466949463,
736
+ "learning_rate": 1.7095737246680644e-05,
737
+ "loss": 0.0803,
738
+ "step": 1040
739
+ },
740
+ {
741
+ "epoch": 0.7337526205450734,
742
+ "grad_norm": 0.04619055986404419,
743
+ "learning_rate": 1.7067784765897974e-05,
744
+ "loss": 0.049,
745
+ "step": 1050
746
+ },
747
+ {
748
+ "epoch": 0.7407407407407407,
749
+ "grad_norm": 0.2753339409828186,
750
+ "learning_rate": 1.7039832285115305e-05,
751
+ "loss": 0.144,
752
+ "step": 1060
753
+ },
754
+ {
755
+ "epoch": 0.7477288609364081,
756
+ "grad_norm": 2.8835108280181885,
757
+ "learning_rate": 1.7011879804332635e-05,
758
+ "loss": 0.0885,
759
+ "step": 1070
760
+ },
761
+ {
762
+ "epoch": 0.7547169811320755,
763
+ "grad_norm": 0.22654394805431366,
764
+ "learning_rate": 1.6983927323549966e-05,
765
+ "loss": 0.0512,
766
+ "step": 1080
767
+ },
768
+ {
769
+ "epoch": 0.7617051013277428,
770
+ "grad_norm": 2.421400547027588,
771
+ "learning_rate": 1.6955974842767297e-05,
772
+ "loss": 0.1544,
773
+ "step": 1090
774
+ },
775
+ {
776
+ "epoch": 0.7686932215234102,
777
+ "grad_norm": 0.03806561604142189,
778
+ "learning_rate": 1.6928022361984627e-05,
779
+ "loss": 0.0225,
780
+ "step": 1100
781
+ },
782
+ {
783
+ "epoch": 0.7756813417190775,
784
+ "grad_norm": 0.08347579091787338,
785
+ "learning_rate": 1.6900069881201958e-05,
786
+ "loss": 0.1386,
787
+ "step": 1110
788
+ },
789
+ {
790
+ "epoch": 0.782669461914745,
791
+ "grad_norm": 0.2059401273727417,
792
+ "learning_rate": 1.687211740041929e-05,
793
+ "loss": 0.2406,
794
+ "step": 1120
795
+ },
796
+ {
797
+ "epoch": 0.7896575821104123,
798
+ "grad_norm": 0.060870036482810974,
799
+ "learning_rate": 1.684416491963662e-05,
800
+ "loss": 0.1158,
801
+ "step": 1130
802
+ },
803
+ {
804
+ "epoch": 0.7966457023060797,
805
+ "grad_norm": 0.14614495635032654,
806
+ "learning_rate": 1.681621243885395e-05,
807
+ "loss": 0.1938,
808
+ "step": 1140
809
+ },
810
+ {
811
+ "epoch": 0.803633822501747,
812
+ "grad_norm": 0.09593985229730606,
813
+ "learning_rate": 1.678825995807128e-05,
814
+ "loss": 0.0627,
815
+ "step": 1150
816
+ },
817
+ {
818
+ "epoch": 0.8106219426974144,
819
+ "grad_norm": 9.543107032775879,
820
+ "learning_rate": 1.676030747728861e-05,
821
+ "loss": 0.1365,
822
+ "step": 1160
823
+ },
824
+ {
825
+ "epoch": 0.8176100628930818,
826
+ "grad_norm": 0.09678485244512558,
827
+ "learning_rate": 1.673235499650594e-05,
828
+ "loss": 0.1438,
829
+ "step": 1170
830
+ },
831
+ {
832
+ "epoch": 0.8245981830887491,
833
+ "grad_norm": 0.24556872248649597,
834
+ "learning_rate": 1.6704402515723272e-05,
835
+ "loss": 0.2233,
836
+ "step": 1180
837
+ },
838
+ {
839
+ "epoch": 0.8315863032844165,
840
+ "grad_norm": 0.06892254948616028,
841
+ "learning_rate": 1.6676450034940602e-05,
842
+ "loss": 0.0601,
843
+ "step": 1190
844
+ },
845
+ {
846
+ "epoch": 0.8385744234800838,
847
+ "grad_norm": 0.6429911255836487,
848
+ "learning_rate": 1.6648497554157933e-05,
849
+ "loss": 0.132,
850
+ "step": 1200
851
+ },
852
+ {
853
+ "epoch": 0.8455625436757512,
854
+ "grad_norm": 0.03480253368616104,
855
+ "learning_rate": 1.6620545073375263e-05,
856
+ "loss": 0.0472,
857
+ "step": 1210
858
+ },
859
+ {
860
+ "epoch": 0.8525506638714185,
861
+ "grad_norm": 0.25101321935653687,
862
+ "learning_rate": 1.6592592592592594e-05,
863
+ "loss": 0.0485,
864
+ "step": 1220
865
+ },
866
+ {
867
+ "epoch": 0.859538784067086,
868
+ "grad_norm": 0.0525231771171093,
869
+ "learning_rate": 1.6564640111809925e-05,
870
+ "loss": 0.1824,
871
+ "step": 1230
872
+ },
873
+ {
874
+ "epoch": 0.8665269042627534,
875
+ "grad_norm": 4.694910526275635,
876
+ "learning_rate": 1.6536687631027255e-05,
877
+ "loss": 0.1666,
878
+ "step": 1240
879
+ },
880
+ {
881
+ "epoch": 0.8735150244584207,
882
+ "grad_norm": 0.09604217112064362,
883
+ "learning_rate": 1.6508735150244586e-05,
884
+ "loss": 0.0638,
885
+ "step": 1250
886
+ },
887
+ {
888
+ "epoch": 0.8805031446540881,
889
+ "grad_norm": 9.125068664550781,
890
+ "learning_rate": 1.6480782669461916e-05,
891
+ "loss": 0.0744,
892
+ "step": 1260
893
+ },
894
+ {
895
+ "epoch": 0.8874912648497554,
896
+ "grad_norm": 0.5258679986000061,
897
+ "learning_rate": 1.6452830188679247e-05,
898
+ "loss": 0.019,
899
+ "step": 1270
900
+ },
901
+ {
902
+ "epoch": 0.8944793850454228,
903
+ "grad_norm": 3.6146976947784424,
904
+ "learning_rate": 1.6424877707896577e-05,
905
+ "loss": 0.0141,
906
+ "step": 1280
907
+ },
908
+ {
909
+ "epoch": 0.9014675052410901,
910
+ "grad_norm": 0.05512338504195213,
911
+ "learning_rate": 1.6396925227113908e-05,
912
+ "loss": 0.0138,
913
+ "step": 1290
914
+ },
915
+ {
916
+ "epoch": 0.9084556254367575,
917
+ "grad_norm": 0.028468603268265724,
918
+ "learning_rate": 1.636897274633124e-05,
919
+ "loss": 0.0815,
920
+ "step": 1300
921
+ },
922
+ {
923
+ "epoch": 0.9154437456324249,
924
+ "grad_norm": 0.7173348665237427,
925
+ "learning_rate": 1.634102026554857e-05,
926
+ "loss": 0.0919,
927
+ "step": 1310
928
+ },
929
+ {
930
+ "epoch": 0.9224318658280922,
931
+ "grad_norm": 0.06138516217470169,
932
+ "learning_rate": 1.63130677847659e-05,
933
+ "loss": 0.1383,
934
+ "step": 1320
935
+ },
936
+ {
937
+ "epoch": 0.9294199860237596,
938
+ "grad_norm": 10.367135047912598,
939
+ "learning_rate": 1.628511530398323e-05,
940
+ "loss": 0.1265,
941
+ "step": 1330
942
+ },
943
+ {
944
+ "epoch": 0.9364081062194269,
945
+ "grad_norm": 0.3321629762649536,
946
+ "learning_rate": 1.625716282320056e-05,
947
+ "loss": 0.1149,
948
+ "step": 1340
949
+ },
950
+ {
951
+ "epoch": 0.9433962264150944,
952
+ "grad_norm": 0.21935391426086426,
953
+ "learning_rate": 1.622921034241789e-05,
954
+ "loss": 0.1147,
955
+ "step": 1350
956
+ },
957
+ {
958
+ "epoch": 0.9503843466107617,
959
+ "grad_norm": 4.630390644073486,
960
+ "learning_rate": 1.6201257861635222e-05,
961
+ "loss": 0.1159,
962
+ "step": 1360
963
+ },
964
+ {
965
+ "epoch": 0.9573724668064291,
966
+ "grad_norm": 0.03748277202248573,
967
+ "learning_rate": 1.6173305380852552e-05,
968
+ "loss": 0.0029,
969
+ "step": 1370
970
+ },
971
+ {
972
+ "epoch": 0.9643605870020965,
973
+ "grad_norm": 0.09677435457706451,
974
+ "learning_rate": 1.6145352900069883e-05,
975
+ "loss": 0.179,
976
+ "step": 1380
977
+ },
978
+ {
979
+ "epoch": 0.9713487071977638,
980
+ "grad_norm": 0.022979794070124626,
981
+ "learning_rate": 1.6117400419287214e-05,
982
+ "loss": 0.0034,
983
+ "step": 1390
984
+ },
985
+ {
986
+ "epoch": 0.9783368273934312,
987
+ "grad_norm": 3.3972229957580566,
988
+ "learning_rate": 1.6089447938504544e-05,
989
+ "loss": 0.1633,
990
+ "step": 1400
991
+ },
992
+ {
993
+ "epoch": 0.9853249475890985,
994
+ "grad_norm": 0.0410487987101078,
995
+ "learning_rate": 1.6061495457721875e-05,
996
+ "loss": 0.0758,
997
+ "step": 1410
998
+ },
999
+ {
1000
+ "epoch": 0.9923130677847659,
1001
+ "grad_norm": 0.1519143134355545,
1002
+ "learning_rate": 1.6033542976939205e-05,
1003
+ "loss": 0.0967,
1004
+ "step": 1420
1005
+ },
1006
+ {
1007
+ "epoch": 0.9993011879804332,
1008
+ "grad_norm": 0.11000066250562668,
1009
+ "learning_rate": 1.6005590496156536e-05,
1010
+ "loss": 0.1054,
1011
+ "step": 1430
1012
+ },
1013
+ {
1014
+ "epoch": 1.0062893081761006,
1015
+ "grad_norm": 9.595348358154297,
1016
+ "learning_rate": 1.5977638015373866e-05,
1017
+ "loss": 0.0367,
1018
+ "step": 1440
1019
+ },
1020
+ {
1021
+ "epoch": 1.013277428371768,
1022
+ "grad_norm": 0.044866789132356644,
1023
+ "learning_rate": 1.5949685534591197e-05,
1024
+ "loss": 0.145,
1025
+ "step": 1450
1026
+ },
1027
+ {
1028
+ "epoch": 1.0202655485674355,
1029
+ "grad_norm": 6.3100361824035645,
1030
+ "learning_rate": 1.5921733053808524e-05,
1031
+ "loss": 0.2702,
1032
+ "step": 1460
1033
+ },
1034
+ {
1035
+ "epoch": 1.0272536687631026,
1036
+ "grad_norm": 1.0666788816452026,
1037
+ "learning_rate": 1.5893780573025858e-05,
1038
+ "loss": 0.1257,
1039
+ "step": 1470
1040
+ },
1041
+ {
1042
+ "epoch": 1.03424178895877,
1043
+ "grad_norm": 15.676182746887207,
1044
+ "learning_rate": 1.586582809224319e-05,
1045
+ "loss": 0.0373,
1046
+ "step": 1480
1047
+ },
1048
+ {
1049
+ "epoch": 1.0412299091544375,
1050
+ "grad_norm": 13.464042663574219,
1051
+ "learning_rate": 1.583787561146052e-05,
1052
+ "loss": 0.1107,
1053
+ "step": 1490
1054
+ },
1055
+ {
1056
+ "epoch": 1.0482180293501049,
1057
+ "grad_norm": 0.029935991391539574,
1058
+ "learning_rate": 1.580992313067785e-05,
1059
+ "loss": 0.0778,
1060
+ "step": 1500
1061
+ },
1062
+ {
1063
+ "epoch": 1.0552061495457723,
1064
+ "grad_norm": 1.0842170715332031,
1065
+ "learning_rate": 1.578197064989518e-05,
1066
+ "loss": 0.2003,
1067
+ "step": 1510
1068
+ },
1069
+ {
1070
+ "epoch": 1.0621942697414395,
1071
+ "grad_norm": 0.8753694891929626,
1072
+ "learning_rate": 1.575401816911251e-05,
1073
+ "loss": 0.0869,
1074
+ "step": 1520
1075
+ },
1076
+ {
1077
+ "epoch": 1.069182389937107,
1078
+ "grad_norm": 0.03639127314090729,
1079
+ "learning_rate": 1.572606568832984e-05,
1080
+ "loss": 0.0035,
1081
+ "step": 1530
1082
+ },
1083
+ {
1084
+ "epoch": 1.0761705101327743,
1085
+ "grad_norm": 0.10639658570289612,
1086
+ "learning_rate": 1.5698113207547172e-05,
1087
+ "loss": 0.0709,
1088
+ "step": 1540
1089
+ },
1090
+ {
1091
+ "epoch": 1.0831586303284417,
1092
+ "grad_norm": 0.09013110399246216,
1093
+ "learning_rate": 1.56701607267645e-05,
1094
+ "loss": 0.0041,
1095
+ "step": 1550
1096
+ },
1097
+ {
1098
+ "epoch": 1.090146750524109,
1099
+ "grad_norm": 0.1398763209581375,
1100
+ "learning_rate": 1.564220824598183e-05,
1101
+ "loss": 0.0039,
1102
+ "step": 1560
1103
+ },
1104
+ {
1105
+ "epoch": 1.0971348707197763,
1106
+ "grad_norm": 0.13840065896511078,
1107
+ "learning_rate": 1.5614255765199164e-05,
1108
+ "loss": 0.0288,
1109
+ "step": 1570
1110
+ },
1111
+ {
1112
+ "epoch": 1.1041229909154437,
1113
+ "grad_norm": 0.03229377046227455,
1114
+ "learning_rate": 1.5586303284416494e-05,
1115
+ "loss": 0.023,
1116
+ "step": 1580
1117
+ },
1118
+ {
1119
+ "epoch": 1.1111111111111112,
1120
+ "grad_norm": 0.09332744032144547,
1121
+ "learning_rate": 1.5558350803633825e-05,
1122
+ "loss": 0.148,
1123
+ "step": 1590
1124
+ },
1125
+ {
1126
+ "epoch": 1.1180992313067786,
1127
+ "grad_norm": 0.24537114799022675,
1128
+ "learning_rate": 1.5530398322851156e-05,
1129
+ "loss": 0.0143,
1130
+ "step": 1600
1131
+ },
1132
+ {
1133
+ "epoch": 1.1250873515024458,
1134
+ "grad_norm": 0.02455071546137333,
1135
+ "learning_rate": 1.5502445842068486e-05,
1136
+ "loss": 0.0378,
1137
+ "step": 1610
1138
+ },
1139
+ {
1140
+ "epoch": 1.1320754716981132,
1141
+ "grad_norm": 0.027023492380976677,
1142
+ "learning_rate": 1.5474493361285817e-05,
1143
+ "loss": 0.0558,
1144
+ "step": 1620
1145
+ },
1146
+ {
1147
+ "epoch": 1.1390635918937806,
1148
+ "grad_norm": 0.048673246055841446,
1149
+ "learning_rate": 1.5446540880503147e-05,
1150
+ "loss": 0.0519,
1151
+ "step": 1630
1152
+ },
1153
+ {
1154
+ "epoch": 1.146051712089448,
1155
+ "grad_norm": 8.062767028808594,
1156
+ "learning_rate": 1.5418588399720474e-05,
1157
+ "loss": 0.0404,
1158
+ "step": 1640
1159
+ },
1160
+ {
1161
+ "epoch": 1.1530398322851152,
1162
+ "grad_norm": 8.358656883239746,
1163
+ "learning_rate": 1.5390635918937805e-05,
1164
+ "loss": 0.1054,
1165
+ "step": 1650
1166
+ },
1167
+ {
1168
+ "epoch": 1.1600279524807826,
1169
+ "grad_norm": 1.3057444095611572,
1170
+ "learning_rate": 1.5362683438155136e-05,
1171
+ "loss": 0.1049,
1172
+ "step": 1660
1173
+ },
1174
+ {
1175
+ "epoch": 1.16701607267645,
1176
+ "grad_norm": 0.018700115382671356,
1177
+ "learning_rate": 1.533473095737247e-05,
1178
+ "loss": 0.0106,
1179
+ "step": 1670
1180
+ },
1181
+ {
1182
+ "epoch": 1.1740041928721174,
1183
+ "grad_norm": 8.731624603271484,
1184
+ "learning_rate": 1.53067784765898e-05,
1185
+ "loss": 0.0541,
1186
+ "step": 1680
1187
+ },
1188
+ {
1189
+ "epoch": 1.1809923130677848,
1190
+ "grad_norm": 1.358834147453308,
1191
+ "learning_rate": 1.527882599580713e-05,
1192
+ "loss": 0.1515,
1193
+ "step": 1690
1194
+ },
1195
+ {
1196
+ "epoch": 1.187980433263452,
1197
+ "grad_norm": 0.05055036395788193,
1198
+ "learning_rate": 1.525087351502446e-05,
1199
+ "loss": 0.1181,
1200
+ "step": 1700
1201
+ },
1202
+ {
1203
+ "epoch": 1.1949685534591195,
1204
+ "grad_norm": 0.19085977971553802,
1205
+ "learning_rate": 1.5222921034241792e-05,
1206
+ "loss": 0.0624,
1207
+ "step": 1710
1208
+ },
1209
+ {
1210
+ "epoch": 1.2019566736547869,
1211
+ "grad_norm": 1.282493233680725,
1212
+ "learning_rate": 1.5194968553459122e-05,
1213
+ "loss": 0.016,
1214
+ "step": 1720
1215
+ },
1216
+ {
1217
+ "epoch": 1.2089447938504543,
1218
+ "grad_norm": 0.03596067801117897,
1219
+ "learning_rate": 1.5167016072676451e-05,
1220
+ "loss": 0.0211,
1221
+ "step": 1730
1222
+ },
1223
+ {
1224
+ "epoch": 1.2159329140461215,
1225
+ "grad_norm": 0.033778220415115356,
1226
+ "learning_rate": 1.5139063591893782e-05,
1227
+ "loss": 0.0105,
1228
+ "step": 1740
1229
+ },
1230
+ {
1231
+ "epoch": 1.2229210342417889,
1232
+ "grad_norm": 0.022970452904701233,
1233
+ "learning_rate": 1.5111111111111112e-05,
1234
+ "loss": 0.1634,
1235
+ "step": 1750
1236
+ },
1237
+ {
1238
+ "epoch": 1.2299091544374563,
1239
+ "grad_norm": 0.08840584754943848,
1240
+ "learning_rate": 1.5083158630328443e-05,
1241
+ "loss": 0.0062,
1242
+ "step": 1760
1243
+ },
1244
+ {
1245
+ "epoch": 1.2368972746331237,
1246
+ "grad_norm": 0.0703321248292923,
1247
+ "learning_rate": 1.5055206149545773e-05,
1248
+ "loss": 0.0554,
1249
+ "step": 1770
1250
+ },
1251
+ {
1252
+ "epoch": 1.2438853948287911,
1253
+ "grad_norm": 2.026752471923828,
1254
+ "learning_rate": 1.5027253668763104e-05,
1255
+ "loss": 0.03,
1256
+ "step": 1780
1257
+ },
1258
+ {
1259
+ "epoch": 1.2508735150244585,
1260
+ "grad_norm": 0.08599984645843506,
1261
+ "learning_rate": 1.4999301187980435e-05,
1262
+ "loss": 0.0303,
1263
+ "step": 1790
1264
+ },
1265
+ {
1266
+ "epoch": 1.2578616352201257,
1267
+ "grad_norm": 0.0998782068490982,
1268
+ "learning_rate": 1.4971348707197765e-05,
1269
+ "loss": 0.1389,
1270
+ "step": 1800
1271
+ },
1272
+ {
1273
+ "epoch": 1.2648497554157931,
1274
+ "grad_norm": 1.7642574310302734,
1275
+ "learning_rate": 1.4943396226415094e-05,
1276
+ "loss": 0.0556,
1277
+ "step": 1810
1278
+ },
1279
+ {
1280
+ "epoch": 1.2718378756114606,
1281
+ "grad_norm": 0.012091516517102718,
1282
+ "learning_rate": 1.4915443745632425e-05,
1283
+ "loss": 0.1116,
1284
+ "step": 1820
1285
+ },
1286
+ {
1287
+ "epoch": 1.2788259958071277,
1288
+ "grad_norm": 0.030048305168747902,
1289
+ "learning_rate": 1.4887491264849757e-05,
1290
+ "loss": 0.0433,
1291
+ "step": 1830
1292
+ },
1293
+ {
1294
+ "epoch": 1.2858141160027952,
1295
+ "grad_norm": 0.025840098038315773,
1296
+ "learning_rate": 1.4859538784067087e-05,
1297
+ "loss": 0.038,
1298
+ "step": 1840
1299
+ },
1300
+ {
1301
+ "epoch": 1.2928022361984626,
1302
+ "grad_norm": 0.4607846736907959,
1303
+ "learning_rate": 1.4831586303284418e-05,
1304
+ "loss": 0.0815,
1305
+ "step": 1850
1306
+ },
1307
+ {
1308
+ "epoch": 1.29979035639413,
1309
+ "grad_norm": 3.925401210784912,
1310
+ "learning_rate": 1.4803633822501749e-05,
1311
+ "loss": 0.0708,
1312
+ "step": 1860
1313
+ },
1314
+ {
1315
+ "epoch": 1.3067784765897974,
1316
+ "grad_norm": 5.653714656829834,
1317
+ "learning_rate": 1.4775681341719079e-05,
1318
+ "loss": 0.0166,
1319
+ "step": 1870
1320
+ },
1321
+ {
1322
+ "epoch": 1.3137665967854648,
1323
+ "grad_norm": 0.02135496586561203,
1324
+ "learning_rate": 1.474772886093641e-05,
1325
+ "loss": 0.0782,
1326
+ "step": 1880
1327
+ },
1328
+ {
1329
+ "epoch": 1.320754716981132,
1330
+ "grad_norm": 0.020742664113640785,
1331
+ "learning_rate": 1.471977638015374e-05,
1332
+ "loss": 0.0048,
1333
+ "step": 1890
1334
+ },
1335
+ {
1336
+ "epoch": 1.3277428371767994,
1337
+ "grad_norm": 0.02374901808798313,
1338
+ "learning_rate": 1.4691823899371069e-05,
1339
+ "loss": 0.081,
1340
+ "step": 1900
1341
+ },
1342
+ {
1343
+ "epoch": 1.3347309573724668,
1344
+ "grad_norm": 0.03886565566062927,
1345
+ "learning_rate": 1.46638714185884e-05,
1346
+ "loss": 0.005,
1347
+ "step": 1910
1348
+ },
1349
+ {
1350
+ "epoch": 1.3417190775681342,
1351
+ "grad_norm": 0.16429579257965088,
1352
+ "learning_rate": 1.463591893780573e-05,
1353
+ "loss": 0.0154,
1354
+ "step": 1920
1355
+ },
1356
+ {
1357
+ "epoch": 1.3487071977638014,
1358
+ "grad_norm": 0.15556152164936066,
1359
+ "learning_rate": 1.4607966457023063e-05,
1360
+ "loss": 0.0659,
1361
+ "step": 1930
1362
+ },
1363
+ {
1364
+ "epoch": 1.3556953179594688,
1365
+ "grad_norm": 0.022907955572009087,
1366
+ "learning_rate": 1.4580013976240393e-05,
1367
+ "loss": 0.068,
1368
+ "step": 1940
1369
+ },
1370
+ {
1371
+ "epoch": 1.3626834381551363,
1372
+ "grad_norm": 0.028623567894101143,
1373
+ "learning_rate": 1.4552061495457724e-05,
1374
+ "loss": 0.0023,
1375
+ "step": 1950
1376
+ },
1377
+ {
1378
+ "epoch": 1.3696715583508037,
1379
+ "grad_norm": 7.010847091674805,
1380
+ "learning_rate": 1.4524109014675054e-05,
1381
+ "loss": 0.147,
1382
+ "step": 1960
1383
+ },
1384
+ {
1385
+ "epoch": 1.376659678546471,
1386
+ "grad_norm": 0.03581179678440094,
1387
+ "learning_rate": 1.4496156533892385e-05,
1388
+ "loss": 0.1209,
1389
+ "step": 1970
1390
+ },
1391
+ {
1392
+ "epoch": 1.3836477987421385,
1393
+ "grad_norm": 0.019916867837309837,
1394
+ "learning_rate": 1.4468204053109715e-05,
1395
+ "loss": 0.0233,
1396
+ "step": 1980
1397
+ },
1398
+ {
1399
+ "epoch": 1.3906359189378057,
1400
+ "grad_norm": 10.045747756958008,
1401
+ "learning_rate": 1.4440251572327044e-05,
1402
+ "loss": 0.0416,
1403
+ "step": 1990
1404
+ },
1405
+ {
1406
+ "epoch": 1.397624039133473,
1407
+ "grad_norm": 0.04035484418272972,
1408
+ "learning_rate": 1.4412299091544375e-05,
1409
+ "loss": 0.0265,
1410
+ "step": 2000
1411
+ },
1412
+ {
1413
+ "epoch": 1.4046121593291405,
1414
+ "grad_norm": 0.7727859616279602,
1415
+ "learning_rate": 1.4384346610761705e-05,
1416
+ "loss": 0.0331,
1417
+ "step": 2010
1418
+ },
1419
+ {
1420
+ "epoch": 1.4116002795248077,
1421
+ "grad_norm": 0.018749171867966652,
1422
+ "learning_rate": 1.4356394129979036e-05,
1423
+ "loss": 0.0749,
1424
+ "step": 2020
1425
+ },
1426
+ {
1427
+ "epoch": 1.4185883997204751,
1428
+ "grad_norm": 5.582619667053223,
1429
+ "learning_rate": 1.4328441649196368e-05,
1430
+ "loss": 0.0622,
1431
+ "step": 2030
1432
+ },
1433
+ {
1434
+ "epoch": 1.4255765199161425,
1435
+ "grad_norm": 0.03347029909491539,
1436
+ "learning_rate": 1.4300489168413699e-05,
1437
+ "loss": 0.0021,
1438
+ "step": 2040
1439
+ },
1440
+ {
1441
+ "epoch": 1.43256464011181,
1442
+ "grad_norm": 0.1696143001317978,
1443
+ "learning_rate": 1.427253668763103e-05,
1444
+ "loss": 0.0021,
1445
+ "step": 2050
1446
+ },
1447
+ {
1448
+ "epoch": 1.4395527603074774,
1449
+ "grad_norm": 0.021968642249703407,
1450
+ "learning_rate": 1.424458420684836e-05,
1451
+ "loss": 0.0025,
1452
+ "step": 2060
1453
+ },
1454
+ {
1455
+ "epoch": 1.4465408805031448,
1456
+ "grad_norm": 0.007907007820904255,
1457
+ "learning_rate": 1.421663172606569e-05,
1458
+ "loss": 0.0478,
1459
+ "step": 2070
1460
+ },
1461
+ {
1462
+ "epoch": 1.453529000698812,
1463
+ "grad_norm": 0.035767070949077606,
1464
+ "learning_rate": 1.418867924528302e-05,
1465
+ "loss": 0.0012,
1466
+ "step": 2080
1467
+ },
1468
+ {
1469
+ "epoch": 1.4605171208944794,
1470
+ "grad_norm": 0.011390830390155315,
1471
+ "learning_rate": 1.416072676450035e-05,
1472
+ "loss": 0.1533,
1473
+ "step": 2090
1474
+ },
1475
+ {
1476
+ "epoch": 1.4675052410901468,
1477
+ "grad_norm": 7.848016738891602,
1478
+ "learning_rate": 1.413277428371768e-05,
1479
+ "loss": 0.0675,
1480
+ "step": 2100
1481
+ },
1482
+ {
1483
+ "epoch": 1.474493361285814,
1484
+ "grad_norm": 0.011802544817328453,
1485
+ "learning_rate": 1.4104821802935011e-05,
1486
+ "loss": 0.0261,
1487
+ "step": 2110
1488
+ },
1489
+ {
1490
+ "epoch": 1.4814814814814814,
1491
+ "grad_norm": 2.9148340225219727,
1492
+ "learning_rate": 1.4076869322152342e-05,
1493
+ "loss": 0.2831,
1494
+ "step": 2120
1495
+ },
1496
+ {
1497
+ "epoch": 1.4884696016771488,
1498
+ "grad_norm": 0.08199399709701538,
1499
+ "learning_rate": 1.4048916841369674e-05,
1500
+ "loss": 0.0088,
1501
+ "step": 2130
1502
+ },
1503
+ {
1504
+ "epoch": 1.4954577218728162,
1505
+ "grad_norm": 0.01199902594089508,
1506
+ "learning_rate": 1.4020964360587004e-05,
1507
+ "loss": 0.0693,
1508
+ "step": 2140
1509
+ },
1510
+ {
1511
+ "epoch": 1.5024458420684836,
1512
+ "grad_norm": 0.07146008312702179,
1513
+ "learning_rate": 1.3993011879804335e-05,
1514
+ "loss": 0.0746,
1515
+ "step": 2150
1516
+ },
1517
+ {
1518
+ "epoch": 1.509433962264151,
1519
+ "grad_norm": 0.02118564024567604,
1520
+ "learning_rate": 1.3965059399021666e-05,
1521
+ "loss": 0.003,
1522
+ "step": 2160
1523
+ },
1524
+ {
1525
+ "epoch": 1.5164220824598185,
1526
+ "grad_norm": 0.016972247511148453,
1527
+ "learning_rate": 1.3937106918238994e-05,
1528
+ "loss": 0.0106,
1529
+ "step": 2170
1530
+ },
1531
+ {
1532
+ "epoch": 1.5234102026554857,
1533
+ "grad_norm": 0.023427631705999374,
1534
+ "learning_rate": 1.3909154437456325e-05,
1535
+ "loss": 0.0444,
1536
+ "step": 2180
1537
+ },
1538
+ {
1539
+ "epoch": 1.530398322851153,
1540
+ "grad_norm": 0.03280281275510788,
1541
+ "learning_rate": 1.3881201956673656e-05,
1542
+ "loss": 0.0026,
1543
+ "step": 2190
1544
+ },
1545
+ {
1546
+ "epoch": 1.5373864430468203,
1547
+ "grad_norm": 0.06073632091283798,
1548
+ "learning_rate": 1.3853249475890986e-05,
1549
+ "loss": 0.194,
1550
+ "step": 2200
1551
+ },
1552
+ {
1553
+ "epoch": 1.5443745632424877,
1554
+ "grad_norm": 0.008114098571240902,
1555
+ "learning_rate": 1.3825296995108317e-05,
1556
+ "loss": 0.0016,
1557
+ "step": 2210
1558
+ },
1559
+ {
1560
+ "epoch": 1.551362683438155,
1561
+ "grad_norm": 0.0467991828918457,
1562
+ "learning_rate": 1.3797344514325647e-05,
1563
+ "loss": 0.1305,
1564
+ "step": 2220
1565
+ },
1566
+ {
1567
+ "epoch": 1.5583508036338225,
1568
+ "grad_norm": 0.06984475255012512,
1569
+ "learning_rate": 1.376939203354298e-05,
1570
+ "loss": 0.1386,
1571
+ "step": 2230
1572
+ },
1573
+ {
1574
+ "epoch": 1.56533892382949,
1575
+ "grad_norm": 0.014983494766056538,
1576
+ "learning_rate": 1.374143955276031e-05,
1577
+ "loss": 0.1399,
1578
+ "step": 2240
1579
+ },
1580
+ {
1581
+ "epoch": 1.5723270440251573,
1582
+ "grad_norm": 0.01568700559437275,
1583
+ "learning_rate": 1.3713487071977637e-05,
1584
+ "loss": 0.1084,
1585
+ "step": 2250
1586
+ },
1587
+ {
1588
+ "epoch": 1.5793151642208247,
1589
+ "grad_norm": 10.549098014831543,
1590
+ "learning_rate": 1.368553459119497e-05,
1591
+ "loss": 0.0082,
1592
+ "step": 2260
1593
+ },
1594
+ {
1595
+ "epoch": 1.586303284416492,
1596
+ "grad_norm": 0.02495586685836315,
1597
+ "learning_rate": 1.36575821104123e-05,
1598
+ "loss": 0.0019,
1599
+ "step": 2270
1600
+ },
1601
+ {
1602
+ "epoch": 1.5932914046121593,
1603
+ "grad_norm": 0.01876898854970932,
1604
+ "learning_rate": 1.362962962962963e-05,
1605
+ "loss": 0.1002,
1606
+ "step": 2280
1607
+ },
1608
+ {
1609
+ "epoch": 1.6002795248078265,
1610
+ "grad_norm": 0.044464971870183945,
1611
+ "learning_rate": 1.3601677148846961e-05,
1612
+ "loss": 0.0019,
1613
+ "step": 2290
1614
+ },
1615
+ {
1616
+ "epoch": 1.607267645003494,
1617
+ "grad_norm": 0.04108593240380287,
1618
+ "learning_rate": 1.3573724668064292e-05,
1619
+ "loss": 0.0753,
1620
+ "step": 2300
1621
+ },
1622
+ {
1623
+ "epoch": 1.6142557651991614,
1624
+ "grad_norm": 7.812976837158203,
1625
+ "learning_rate": 1.3545772187281622e-05,
1626
+ "loss": 0.2875,
1627
+ "step": 2310
1628
+ },
1629
+ {
1630
+ "epoch": 1.6212438853948288,
1631
+ "grad_norm": 4.261682987213135,
1632
+ "learning_rate": 1.3517819706498953e-05,
1633
+ "loss": 0.1059,
1634
+ "step": 2320
1635
+ },
1636
+ {
1637
+ "epoch": 1.6282320055904962,
1638
+ "grad_norm": 0.03309008479118347,
1639
+ "learning_rate": 1.3489867225716285e-05,
1640
+ "loss": 0.0051,
1641
+ "step": 2330
1642
+ },
1643
+ {
1644
+ "epoch": 1.6352201257861636,
1645
+ "grad_norm": 9.983598709106445,
1646
+ "learning_rate": 1.3461914744933612e-05,
1647
+ "loss": 0.0249,
1648
+ "step": 2340
1649
+ },
1650
+ {
1651
+ "epoch": 1.642208245981831,
1652
+ "grad_norm": 4.973055839538574,
1653
+ "learning_rate": 1.3433962264150943e-05,
1654
+ "loss": 0.1441,
1655
+ "step": 2350
1656
+ },
1657
+ {
1658
+ "epoch": 1.6491963661774982,
1659
+ "grad_norm": 9.258291244506836,
1660
+ "learning_rate": 1.3406009783368275e-05,
1661
+ "loss": 0.1151,
1662
+ "step": 2360
1663
+ },
1664
+ {
1665
+ "epoch": 1.6561844863731656,
1666
+ "grad_norm": 0.015135078690946102,
1667
+ "learning_rate": 1.3378057302585606e-05,
1668
+ "loss": 0.0458,
1669
+ "step": 2370
1670
+ },
1671
+ {
1672
+ "epoch": 1.663172606568833,
1673
+ "grad_norm": 0.027922656387090683,
1674
+ "learning_rate": 1.3350104821802936e-05,
1675
+ "loss": 0.012,
1676
+ "step": 2380
1677
+ },
1678
+ {
1679
+ "epoch": 1.6701607267645002,
1680
+ "grad_norm": 0.043312493711709976,
1681
+ "learning_rate": 1.3322152341020267e-05,
1682
+ "loss": 0.1028,
1683
+ "step": 2390
1684
+ },
1685
+ {
1686
+ "epoch": 1.6771488469601676,
1687
+ "grad_norm": 0.012087292037904263,
1688
+ "learning_rate": 1.3294199860237597e-05,
1689
+ "loss": 0.0475,
1690
+ "step": 2400
1691
+ },
1692
+ {
1693
+ "epoch": 1.684136967155835,
1694
+ "grad_norm": 0.5556979775428772,
1695
+ "learning_rate": 1.3266247379454928e-05,
1696
+ "loss": 0.0142,
1697
+ "step": 2410
1698
+ },
1699
+ {
1700
+ "epoch": 1.6911250873515025,
1701
+ "grad_norm": 5.196042537689209,
1702
+ "learning_rate": 1.3238294898672259e-05,
1703
+ "loss": 0.1002,
1704
+ "step": 2420
1705
+ },
1706
+ {
1707
+ "epoch": 1.6981132075471699,
1708
+ "grad_norm": 0.010861317627131939,
1709
+ "learning_rate": 1.3210342417889587e-05,
1710
+ "loss": 0.0953,
1711
+ "step": 2430
1712
+ },
1713
+ {
1714
+ "epoch": 1.7051013277428373,
1715
+ "grad_norm": 0.017955880612134933,
1716
+ "learning_rate": 1.3182389937106918e-05,
1717
+ "loss": 0.0642,
1718
+ "step": 2440
1719
+ },
1720
+ {
1721
+ "epoch": 1.7120894479385047,
1722
+ "grad_norm": 0.028966449201107025,
1723
+ "learning_rate": 1.3154437456324249e-05,
1724
+ "loss": 0.1619,
1725
+ "step": 2450
1726
+ },
1727
+ {
1728
+ "epoch": 1.719077568134172,
1729
+ "grad_norm": 0.8712167739868164,
1730
+ "learning_rate": 1.312648497554158e-05,
1731
+ "loss": 0.0324,
1732
+ "step": 2460
1733
+ },
1734
+ {
1735
+ "epoch": 1.7260656883298393,
1736
+ "grad_norm": 0.2874550521373749,
1737
+ "learning_rate": 1.3098532494758911e-05,
1738
+ "loss": 0.026,
1739
+ "step": 2470
1740
+ },
1741
+ {
1742
+ "epoch": 1.7330538085255065,
1743
+ "grad_norm": 0.010579444468021393,
1744
+ "learning_rate": 1.3070580013976242e-05,
1745
+ "loss": 0.0048,
1746
+ "step": 2480
1747
+ },
1748
+ {
1749
+ "epoch": 1.740041928721174,
1750
+ "grad_norm": 0.15773366391658783,
1751
+ "learning_rate": 1.3042627533193573e-05,
1752
+ "loss": 0.108,
1753
+ "step": 2490
1754
+ },
1755
+ {
1756
+ "epoch": 1.7470300489168413,
1757
+ "grad_norm": 0.10269400477409363,
1758
+ "learning_rate": 1.3014675052410903e-05,
1759
+ "loss": 0.0884,
1760
+ "step": 2500
1761
+ },
1762
+ {
1763
+ "epoch": 1.7540181691125087,
1764
+ "grad_norm": 0.012210669927299023,
1765
+ "learning_rate": 1.2986722571628234e-05,
1766
+ "loss": 0.0025,
1767
+ "step": 2510
1768
+ },
1769
+ {
1770
+ "epoch": 1.7610062893081762,
1771
+ "grad_norm": 0.046242229640483856,
1772
+ "learning_rate": 1.2958770090845563e-05,
1773
+ "loss": 0.0124,
1774
+ "step": 2520
1775
+ },
1776
+ {
1777
+ "epoch": 1.7679944095038436,
1778
+ "grad_norm": 12.532265663146973,
1779
+ "learning_rate": 1.2930817610062893e-05,
1780
+ "loss": 0.1502,
1781
+ "step": 2530
1782
+ },
1783
+ {
1784
+ "epoch": 1.774982529699511,
1785
+ "grad_norm": 0.018337048590183258,
1786
+ "learning_rate": 1.2902865129280224e-05,
1787
+ "loss": 0.0402,
1788
+ "step": 2540
1789
+ },
1790
+ {
1791
+ "epoch": 1.7819706498951782,
1792
+ "grad_norm": 0.9082455039024353,
1793
+ "learning_rate": 1.2874912648497554e-05,
1794
+ "loss": 0.0719,
1795
+ "step": 2550
1796
+ },
1797
+ {
1798
+ "epoch": 1.7889587700908456,
1799
+ "grad_norm": 0.6127817034721375,
1800
+ "learning_rate": 1.2846960167714887e-05,
1801
+ "loss": 0.011,
1802
+ "step": 2560
1803
+ },
1804
+ {
1805
+ "epoch": 1.7959468902865128,
1806
+ "grad_norm": 0.02741195261478424,
1807
+ "learning_rate": 1.2819007686932217e-05,
1808
+ "loss": 0.0436,
1809
+ "step": 2570
1810
+ },
1811
+ {
1812
+ "epoch": 1.8029350104821802,
1813
+ "grad_norm": 0.012224216014146805,
1814
+ "learning_rate": 1.2791055206149548e-05,
1815
+ "loss": 0.0026,
1816
+ "step": 2580
1817
+ },
1818
+ {
1819
+ "epoch": 1.8099231306778476,
1820
+ "grad_norm": 0.01689975894987583,
1821
+ "learning_rate": 1.2763102725366878e-05,
1822
+ "loss": 0.0067,
1823
+ "step": 2590
1824
+ },
1825
+ {
1826
+ "epoch": 1.816911250873515,
1827
+ "grad_norm": 0.010701341554522514,
1828
+ "learning_rate": 1.2735150244584207e-05,
1829
+ "loss": 0.008,
1830
+ "step": 2600
1831
+ },
1832
+ {
1833
+ "epoch": 1.8238993710691824,
1834
+ "grad_norm": 0.12163686007261276,
1835
+ "learning_rate": 1.2707197763801538e-05,
1836
+ "loss": 0.0767,
1837
+ "step": 2610
1838
+ },
1839
+ {
1840
+ "epoch": 1.8308874912648498,
1841
+ "grad_norm": 0.07165365666151047,
1842
+ "learning_rate": 1.2679245283018868e-05,
1843
+ "loss": 0.0028,
1844
+ "step": 2620
1845
+ },
1846
+ {
1847
+ "epoch": 1.8378756114605173,
1848
+ "grad_norm": 0.007849736139178276,
1849
+ "learning_rate": 1.2651292802236199e-05,
1850
+ "loss": 0.0488,
1851
+ "step": 2630
1852
+ },
1853
+ {
1854
+ "epoch": 1.8448637316561844,
1855
+ "grad_norm": 0.01707429252564907,
1856
+ "learning_rate": 1.262334032145353e-05,
1857
+ "loss": 0.0022,
1858
+ "step": 2640
1859
+ },
1860
+ {
1861
+ "epoch": 1.8518518518518519,
1862
+ "grad_norm": 6.567415237426758,
1863
+ "learning_rate": 1.259538784067086e-05,
1864
+ "loss": 0.0677,
1865
+ "step": 2650
1866
+ },
1867
+ {
1868
+ "epoch": 1.858839972047519,
1869
+ "grad_norm": 0.005324068479239941,
1870
+ "learning_rate": 1.2567435359888192e-05,
1871
+ "loss": 0.0057,
1872
+ "step": 2660
1873
+ },
1874
+ {
1875
+ "epoch": 1.8658280922431865,
1876
+ "grad_norm": 11.791135787963867,
1877
+ "learning_rate": 1.2539482879105523e-05,
1878
+ "loss": 0.0764,
1879
+ "step": 2670
1880
+ },
1881
+ {
1882
+ "epoch": 1.8728162124388539,
1883
+ "grad_norm": 0.005892631132155657,
1884
+ "learning_rate": 1.2511530398322853e-05,
1885
+ "loss": 0.0736,
1886
+ "step": 2680
1887
+ },
1888
+ {
1889
+ "epoch": 1.8798043326345213,
1890
+ "grad_norm": 0.04005246236920357,
1891
+ "learning_rate": 1.2483577917540182e-05,
1892
+ "loss": 0.0617,
1893
+ "step": 2690
1894
+ },
1895
+ {
1896
+ "epoch": 1.8867924528301887,
1897
+ "grad_norm": 0.010624396614730358,
1898
+ "learning_rate": 1.2455625436757513e-05,
1899
+ "loss": 0.0011,
1900
+ "step": 2700
1901
+ },
1902
+ {
1903
+ "epoch": 1.8937805730258561,
1904
+ "grad_norm": 0.01761520840227604,
1905
+ "learning_rate": 1.2427672955974843e-05,
1906
+ "loss": 0.1574,
1907
+ "step": 2710
1908
+ },
1909
+ {
1910
+ "epoch": 1.9007686932215235,
1911
+ "grad_norm": 0.05212152749300003,
1912
+ "learning_rate": 1.2399720475192174e-05,
1913
+ "loss": 0.0618,
1914
+ "step": 2720
1915
+ },
1916
+ {
1917
+ "epoch": 1.9077568134171907,
1918
+ "grad_norm": 0.012297810055315495,
1919
+ "learning_rate": 1.2371767994409504e-05,
1920
+ "loss": 0.0802,
1921
+ "step": 2730
1922
+ },
1923
+ {
1924
+ "epoch": 1.9147449336128581,
1925
+ "grad_norm": 1.4671796560287476,
1926
+ "learning_rate": 1.2343815513626835e-05,
1927
+ "loss": 0.0351,
1928
+ "step": 2740
1929
+ },
1930
+ {
1931
+ "epoch": 1.9217330538085255,
1932
+ "grad_norm": 1.28997802734375,
1933
+ "learning_rate": 1.2315863032844166e-05,
1934
+ "loss": 0.0069,
1935
+ "step": 2750
1936
+ },
1937
+ {
1938
+ "epoch": 1.9287211740041927,
1939
+ "grad_norm": 0.15810908377170563,
1940
+ "learning_rate": 1.2287910552061498e-05,
1941
+ "loss": 0.0037,
1942
+ "step": 2760
1943
+ },
1944
+ {
1945
+ "epoch": 1.9357092941998602,
1946
+ "grad_norm": 0.01939568482339382,
1947
+ "learning_rate": 1.2259958071278828e-05,
1948
+ "loss": 0.1375,
1949
+ "step": 2770
1950
+ },
1951
+ {
1952
+ "epoch": 1.9426974143955276,
1953
+ "grad_norm": 0.023233506828546524,
1954
+ "learning_rate": 1.2232005590496157e-05,
1955
+ "loss": 0.0016,
1956
+ "step": 2780
1957
+ },
1958
+ {
1959
+ "epoch": 1.949685534591195,
1960
+ "grad_norm": 13.808353424072266,
1961
+ "learning_rate": 1.2204053109713488e-05,
1962
+ "loss": 0.1251,
1963
+ "step": 2790
1964
+ },
1965
+ {
1966
+ "epoch": 1.9566736547868624,
1967
+ "grad_norm": 0.007430393248796463,
1968
+ "learning_rate": 1.2176100628930818e-05,
1969
+ "loss": 0.0957,
1970
+ "step": 2800
1971
+ },
1972
+ {
1973
+ "epoch": 1.9636617749825298,
1974
+ "grad_norm": 0.015328769572079182,
1975
+ "learning_rate": 1.2148148148148149e-05,
1976
+ "loss": 0.0019,
1977
+ "step": 2810
1978
+ },
1979
+ {
1980
+ "epoch": 1.9706498951781972,
1981
+ "grad_norm": 0.05328751727938652,
1982
+ "learning_rate": 1.212019566736548e-05,
1983
+ "loss": 0.2193,
1984
+ "step": 2820
1985
+ },
1986
+ {
1987
+ "epoch": 1.9776380153738644,
1988
+ "grad_norm": 0.05544736981391907,
1989
+ "learning_rate": 1.209224318658281e-05,
1990
+ "loss": 0.011,
1991
+ "step": 2830
1992
+ },
1993
+ {
1994
+ "epoch": 1.9846261355695318,
1995
+ "grad_norm": 0.006596289575099945,
1996
+ "learning_rate": 1.206429070580014e-05,
1997
+ "loss": 0.0454,
1998
+ "step": 2840
1999
+ },
2000
+ {
2001
+ "epoch": 1.991614255765199,
2002
+ "grad_norm": 7.7824015617370605,
2003
+ "learning_rate": 1.2036338225017471e-05,
2004
+ "loss": 0.024,
2005
+ "step": 2850
2006
+ },
2007
+ {
2008
+ "epoch": 1.9986023759608664,
2009
+ "grad_norm": 0.024772530421614647,
2010
+ "learning_rate": 1.2008385744234804e-05,
2011
+ "loss": 0.0109,
2012
+ "step": 2860
2013
+ }
2014
+ ],
2015
+ "logging_steps": 10,
2016
+ "max_steps": 7155,
2017
+ "num_input_tokens_seen": 0,
2018
+ "num_train_epochs": 5,
2019
+ "save_steps": 500,
2020
+ "stateful_callbacks": {
2021
+ "TrainerControl": {
2022
+ "args": {
2023
+ "should_epoch_stop": false,
2024
+ "should_evaluate": false,
2025
+ "should_log": false,
2026
+ "should_save": true,
2027
+ "should_training_stop": false
2028
+ },
2029
+ "attributes": {}
2030
+ }
2031
+ },
2032
+ "total_flos": 3093823953408000.0,
2033
+ "train_batch_size": 8,
2034
+ "trial_name": null,
2035
+ "trial_params": null
2036
+ }
checkpoint-2862/training_args.bin ADDED
@@ -0,0 +1,3 @@
 
 
 
 
1
+ version https://git-lfs.github.com/spec/v1
2
+ oid sha256:444dc926ee87ace38749c0cb481134ef6ff50da3debedc51baf9fb097d31d73d
3
+ size 5841
checkpoint-4293/README.md ADDED
@@ -0,0 +1,206 @@
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
1
+ ---
2
+ base_model: protectai/deberta-v3-base-prompt-injection-v2
3
+ library_name: peft
4
+ tags:
5
+ - base_model:adapter:protectai/deberta-v3-base-prompt-injection-v2
6
+ - lora
7
+ - transformers
8
+ ---
9
+
10
+ # Model Card for Model ID
11
+
12
+ <!-- Provide a quick summary of what the model is/does. -->
13
+
14
+
15
+
16
+ ## Model Details
17
+
18
+ ### Model Description
19
+
20
+ <!-- Provide a longer summary of what this model is. -->
21
+
22
+
23
+
24
+ - **Developed by:** [More Information Needed]
25
+ - **Funded by [optional]:** [More Information Needed]
26
+ - **Shared by [optional]:** [More Information Needed]
27
+ - **Model type:** [More Information Needed]
28
+ - **Language(s) (NLP):** [More Information Needed]
29
+ - **License:** [More Information Needed]
30
+ - **Finetuned from model [optional]:** [More Information Needed]
31
+
32
+ ### Model Sources [optional]
33
+
34
+ <!-- Provide the basic links for the model. -->
35
+
36
+ - **Repository:** [More Information Needed]
37
+ - **Paper [optional]:** [More Information Needed]
38
+ - **Demo [optional]:** [More Information Needed]
39
+
40
+ ## Uses
41
+
42
+ <!-- Address questions around how the model is intended to be used, including the foreseeable users of the model and those affected by the model. -->
43
+
44
+ ### Direct Use
45
+
46
+ <!-- This section is for the model use without fine-tuning or plugging into a larger ecosystem/app. -->
47
+
48
+ [More Information Needed]
49
+
50
+ ### Downstream Use [optional]
51
+
52
+ <!-- This section is for the model use when fine-tuned for a task, or when plugged into a larger ecosystem/app -->
53
+
54
+ [More Information Needed]
55
+
56
+ ### Out-of-Scope Use
57
+
58
+ <!-- This section addresses misuse, malicious use, and uses that the model will not work well for. -->
59
+
60
+ [More Information Needed]
61
+
62
+ ## Bias, Risks, and Limitations
63
+
64
+ <!-- This section is meant to convey both technical and sociotechnical limitations. -->
65
+
66
+ [More Information Needed]
67
+
68
+ ### Recommendations
69
+
70
+ <!-- This section is meant to convey recommendations with respect to the bias, risk, and technical limitations. -->
71
+
72
+ Users (both direct and downstream) should be made aware of the risks, biases and limitations of the model. More information needed for further recommendations.
73
+
74
+ ## How to Get Started with the Model
75
+
76
+ Use the code below to get started with the model.
77
+
78
+ [More Information Needed]
79
+
80
+ ## Training Details
81
+
82
+ ### Training Data
83
+
84
+ <!-- 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. -->
85
+
86
+ [More Information Needed]
87
+
88
+ ### Training Procedure
89
+
90
+ <!-- This relates heavily to the Technical Specifications. Content here should link to that section when it is relevant to the training procedure. -->
91
+
92
+ #### Preprocessing [optional]
93
+
94
+ [More Information Needed]
95
+
96
+
97
+ #### Training Hyperparameters
98
+
99
+ - **Training regime:** [More Information Needed] <!--fp32, fp16 mixed precision, bf16 mixed precision, bf16 non-mixed precision, fp16 non-mixed precision, fp8 mixed precision -->
100
+
101
+ #### Speeds, Sizes, Times [optional]
102
+
103
+ <!-- This section provides information about throughput, start/end time, checkpoint size if relevant, etc. -->
104
+
105
+ [More Information Needed]
106
+
107
+ ## Evaluation
108
+
109
+ <!-- This section describes the evaluation protocols and provides the results. -->
110
+
111
+ ### Testing Data, Factors & Metrics
112
+
113
+ #### Testing Data
114
+
115
+ <!-- This should link to a Dataset Card if possible. -->
116
+
117
+ [More Information Needed]
118
+
119
+ #### Factors
120
+
121
+ <!-- These are the things the evaluation is disaggregating by, e.g., subpopulations or domains. -->
122
+
123
+ [More Information Needed]
124
+
125
+ #### Metrics
126
+
127
+ <!-- These are the evaluation metrics being used, ideally with a description of why. -->
128
+
129
+ [More Information Needed]
130
+
131
+ ### Results
132
+
133
+ [More Information Needed]
134
+
135
+ #### Summary
136
+
137
+
138
+
139
+ ## Model Examination [optional]
140
+
141
+ <!-- Relevant interpretability work for the model goes here -->
142
+
143
+ [More Information Needed]
144
+
145
+ ## Environmental Impact
146
+
147
+ <!-- Total emissions (in grams of CO2eq) and additional considerations, such as electricity usage, go here. Edit the suggested text below accordingly -->
148
+
149
+ 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).
150
+
151
+ - **Hardware Type:** [More Information Needed]
152
+ - **Hours used:** [More Information Needed]
153
+ - **Cloud Provider:** [More Information Needed]
154
+ - **Compute Region:** [More Information Needed]
155
+ - **Carbon Emitted:** [More Information Needed]
156
+
157
+ ## Technical Specifications [optional]
158
+
159
+ ### Model Architecture and Objective
160
+
161
+ [More Information Needed]
162
+
163
+ ### Compute Infrastructure
164
+
165
+ [More Information Needed]
166
+
167
+ #### Hardware
168
+
169
+ [More Information Needed]
170
+
171
+ #### Software
172
+
173
+ [More Information Needed]
174
+
175
+ ## Citation [optional]
176
+
177
+ <!-- If there is a paper or blog post introducing the model, the APA and Bibtex information for that should go in this section. -->
178
+
179
+ **BibTeX:**
180
+
181
+ [More Information Needed]
182
+
183
+ **APA:**
184
+
185
+ [More Information Needed]
186
+
187
+ ## Glossary [optional]
188
+
189
+ <!-- If relevant, include terms and calculations in this section that can help readers understand the model or model card. -->
190
+
191
+ [More Information Needed]
192
+
193
+ ## More Information [optional]
194
+
195
+ [More Information Needed]
196
+
197
+ ## Model Card Authors [optional]
198
+
199
+ [More Information Needed]
200
+
201
+ ## Model Card Contact
202
+
203
+ [More Information Needed]
204
+ ### Framework versions
205
+
206
+ - PEFT 0.18.1
checkpoint-4293/adapter_config.json ADDED
@@ -0,0 +1,44 @@
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
1
+ {
2
+ "alora_invocation_tokens": null,
3
+ "alpha_pattern": {},
4
+ "arrow_config": null,
5
+ "auto_mapping": null,
6
+ "base_model_name_or_path": "protectai/deberta-v3-base-prompt-injection-v2",
7
+ "bias": "none",
8
+ "corda_config": null,
9
+ "ensure_weight_tying": false,
10
+ "eva_config": null,
11
+ "exclude_modules": null,
12
+ "fan_in_fan_out": false,
13
+ "inference_mode": true,
14
+ "init_lora_weights": true,
15
+ "layer_replication": null,
16
+ "layers_pattern": null,
17
+ "layers_to_transform": null,
18
+ "loftq_config": {},
19
+ "lora_alpha": 64,
20
+ "lora_bias": false,
21
+ "lora_dropout": 0.05,
22
+ "megatron_config": null,
23
+ "megatron_core": "megatron.core",
24
+ "modules_to_save": [
25
+ "classifier",
26
+ "score"
27
+ ],
28
+ "peft_type": "LORA",
29
+ "peft_version": "0.18.1",
30
+ "qalora_group_size": 16,
31
+ "r": 64,
32
+ "rank_pattern": {},
33
+ "revision": null,
34
+ "target_modules": [
35
+ "query_proj",
36
+ "value_proj"
37
+ ],
38
+ "target_parameters": null,
39
+ "task_type": "SEQ_CLS",
40
+ "trainable_token_indices": null,
41
+ "use_dora": false,
42
+ "use_qalora": false,
43
+ "use_rslora": false
44
+ }
checkpoint-4293/adapter_model.safetensors ADDED
@@ -0,0 +1,3 @@
 
 
 
 
1
+ version https://git-lfs.github.com/spec/v1
2
+ oid sha256:23a89efb0de7af674fe98d9ccacc58b5c5de7aca059db7c1290a1e1a2018c62a
3
+ size 9450704
checkpoint-4293/optimizer.pt ADDED
@@ -0,0 +1,3 @@
 
 
 
 
1
+ version https://git-lfs.github.com/spec/v1
2
+ oid sha256:2f0fe83ef6135d8c5def12e79a67dc1ee73ba76c405696e311cfd59819bf4cb3
3
+ size 18930123
checkpoint-4293/rng_state.pth ADDED
@@ -0,0 +1,3 @@
 
 
 
 
1
+ version https://git-lfs.github.com/spec/v1
2
+ oid sha256:d068ed9e3bc95c78ab2f76e0954f6cde56524e8b7ef883456d5027ce3a7c602b
3
+ size 14645
checkpoint-4293/scaler.pt ADDED
@@ -0,0 +1,3 @@
 
 
 
 
1
+ version https://git-lfs.github.com/spec/v1
2
+ oid sha256:1ff18aacfcfcd77c4b6500cba7558ecc92c94b133f0de4ac7c1fc8bc80cd0d6e
3
+ size 1383
checkpoint-4293/scheduler.pt ADDED
@@ -0,0 +1,3 @@
 
 
 
 
1
+ version https://git-lfs.github.com/spec/v1
2
+ oid sha256:7167b58368949946f4516433013fc7f395f142d0233671c9fc9f92aa46fef880
3
+ size 1465
checkpoint-4293/special_tokens_map.json ADDED
@@ -0,0 +1,51 @@
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
1
+ {
2
+ "bos_token": {
3
+ "content": "[CLS]",
4
+ "lstrip": false,
5
+ "normalized": false,
6
+ "rstrip": false,
7
+ "single_word": false
8
+ },
9
+ "cls_token": {
10
+ "content": "[CLS]",
11
+ "lstrip": false,
12
+ "normalized": false,
13
+ "rstrip": false,
14
+ "single_word": false
15
+ },
16
+ "eos_token": {
17
+ "content": "[SEP]",
18
+ "lstrip": false,
19
+ "normalized": false,
20
+ "rstrip": false,
21
+ "single_word": false
22
+ },
23
+ "mask_token": {
24
+ "content": "[MASK]",
25
+ "lstrip": false,
26
+ "normalized": false,
27
+ "rstrip": false,
28
+ "single_word": false
29
+ },
30
+ "pad_token": {
31
+ "content": "[PAD]",
32
+ "lstrip": false,
33
+ "normalized": false,
34
+ "rstrip": false,
35
+ "single_word": false
36
+ },
37
+ "sep_token": {
38
+ "content": "[SEP]",
39
+ "lstrip": false,
40
+ "normalized": false,
41
+ "rstrip": false,
42
+ "single_word": false
43
+ },
44
+ "unk_token": {
45
+ "content": "[UNK]",
46
+ "lstrip": false,
47
+ "normalized": true,
48
+ "rstrip": false,
49
+ "single_word": false
50
+ }
51
+ }
checkpoint-4293/tokenizer.json ADDED
The diff for this file is too large to render. See raw diff
 
checkpoint-4293/tokenizer_config.json ADDED
@@ -0,0 +1,63 @@
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
1
+ {
2
+ "added_tokens_decoder": {
3
+ "0": {
4
+ "content": "[PAD]",
5
+ "lstrip": false,
6
+ "normalized": false,
7
+ "rstrip": false,
8
+ "single_word": false,
9
+ "special": true
10
+ },
11
+ "1": {
12
+ "content": "[CLS]",
13
+ "lstrip": false,
14
+ "normalized": false,
15
+ "rstrip": false,
16
+ "single_word": false,
17
+ "special": true
18
+ },
19
+ "2": {
20
+ "content": "[SEP]",
21
+ "lstrip": false,
22
+ "normalized": false,
23
+ "rstrip": false,
24
+ "single_word": false,
25
+ "special": true
26
+ },
27
+ "3": {
28
+ "content": "[UNK]",
29
+ "lstrip": false,
30
+ "normalized": true,
31
+ "rstrip": false,
32
+ "single_word": false,
33
+ "special": true
34
+ },
35
+ "128000": {
36
+ "content": "[MASK]",
37
+ "lstrip": false,
38
+ "normalized": false,
39
+ "rstrip": false,
40
+ "single_word": false,
41
+ "special": true
42
+ }
43
+ },
44
+ "bos_token": "[CLS]",
45
+ "clean_up_tokenization_spaces": true,
46
+ "cls_token": "[CLS]",
47
+ "do_lower_case": false,
48
+ "eos_token": "[SEP]",
49
+ "extra_special_tokens": {},
50
+ "mask_token": "[MASK]",
51
+ "max_length": 512,
52
+ "model_max_length": 1000000000000000019884624838656,
53
+ "pad_token": "[PAD]",
54
+ "sep_token": "[SEP]",
55
+ "sp_model_kwargs": {},
56
+ "split_by_punct": false,
57
+ "stride": 0,
58
+ "tokenizer_class": "DebertaV2Tokenizer",
59
+ "truncation_side": "right",
60
+ "truncation_strategy": "longest_first",
61
+ "unk_token": "[UNK]",
62
+ "vocab_type": "spm"
63
+ }
checkpoint-4293/trainer_state.json ADDED
@@ -0,0 +1,3037 @@
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
1
+ {
2
+ "best_global_step": null,
3
+ "best_metric": null,
4
+ "best_model_checkpoint": null,
5
+ "epoch": 3.0,
6
+ "eval_steps": 500,
7
+ "global_step": 4293,
8
+ "is_hyper_param_search": false,
9
+ "is_local_process_zero": true,
10
+ "is_world_process_zero": true,
11
+ "log_history": [
12
+ {
13
+ "epoch": 0.0069881201956673656,
14
+ "grad_norm": 14.528067588806152,
15
+ "learning_rate": 1.99748427672956e-05,
16
+ "loss": 4.6827,
17
+ "step": 10
18
+ },
19
+ {
20
+ "epoch": 0.013976240391334731,
21
+ "grad_norm": 13.050820350646973,
22
+ "learning_rate": 1.994689028651293e-05,
23
+ "loss": 4.529,
24
+ "step": 20
25
+ },
26
+ {
27
+ "epoch": 0.020964360587002098,
28
+ "grad_norm": 2.765507698059082,
29
+ "learning_rate": 1.991893780573026e-05,
30
+ "loss": 3.4294,
31
+ "step": 30
32
+ },
33
+ {
34
+ "epoch": 0.027952480782669462,
35
+ "grad_norm": 16.95716667175293,
36
+ "learning_rate": 1.989098532494759e-05,
37
+ "loss": 3.352,
38
+ "step": 40
39
+ },
40
+ {
41
+ "epoch": 0.03494060097833683,
42
+ "grad_norm": 5.339346408843994,
43
+ "learning_rate": 1.986303284416492e-05,
44
+ "loss": 2.2654,
45
+ "step": 50
46
+ },
47
+ {
48
+ "epoch": 0.041928721174004195,
49
+ "grad_norm": 7.168309688568115,
50
+ "learning_rate": 1.983508036338225e-05,
51
+ "loss": 2.0076,
52
+ "step": 60
53
+ },
54
+ {
55
+ "epoch": 0.04891684136967156,
56
+ "grad_norm": 5.793034076690674,
57
+ "learning_rate": 1.9807127882599582e-05,
58
+ "loss": 1.6779,
59
+ "step": 70
60
+ },
61
+ {
62
+ "epoch": 0.055904961565338925,
63
+ "grad_norm": 6.093694686889648,
64
+ "learning_rate": 1.9779175401816913e-05,
65
+ "loss": 1.2564,
66
+ "step": 80
67
+ },
68
+ {
69
+ "epoch": 0.06289308176100629,
70
+ "grad_norm": 3.321807622909546,
71
+ "learning_rate": 1.9751222921034243e-05,
72
+ "loss": 1.4143,
73
+ "step": 90
74
+ },
75
+ {
76
+ "epoch": 0.06988120195667366,
77
+ "grad_norm": 2.9472098350524902,
78
+ "learning_rate": 1.9723270440251574e-05,
79
+ "loss": 1.0533,
80
+ "step": 100
81
+ },
82
+ {
83
+ "epoch": 0.07686932215234102,
84
+ "grad_norm": 8.515447616577148,
85
+ "learning_rate": 1.9695317959468904e-05,
86
+ "loss": 1.2895,
87
+ "step": 110
88
+ },
89
+ {
90
+ "epoch": 0.08385744234800839,
91
+ "grad_norm": 3.560924530029297,
92
+ "learning_rate": 1.9667365478686235e-05,
93
+ "loss": 1.097,
94
+ "step": 120
95
+ },
96
+ {
97
+ "epoch": 0.09084556254367575,
98
+ "grad_norm": 4.986349582672119,
99
+ "learning_rate": 1.9639412997903566e-05,
100
+ "loss": 0.9556,
101
+ "step": 130
102
+ },
103
+ {
104
+ "epoch": 0.09783368273934312,
105
+ "grad_norm": 1.27532160282135,
106
+ "learning_rate": 1.9611460517120896e-05,
107
+ "loss": 0.8306,
108
+ "step": 140
109
+ },
110
+ {
111
+ "epoch": 0.10482180293501048,
112
+ "grad_norm": 2.6796844005584717,
113
+ "learning_rate": 1.9583508036338227e-05,
114
+ "loss": 0.7747,
115
+ "step": 150
116
+ },
117
+ {
118
+ "epoch": 0.11180992313067785,
119
+ "grad_norm": 3.4592413902282715,
120
+ "learning_rate": 1.9555555555555557e-05,
121
+ "loss": 0.6187,
122
+ "step": 160
123
+ },
124
+ {
125
+ "epoch": 0.1187980433263452,
126
+ "grad_norm": 4.833241939544678,
127
+ "learning_rate": 1.9527603074772888e-05,
128
+ "loss": 0.6831,
129
+ "step": 170
130
+ },
131
+ {
132
+ "epoch": 0.12578616352201258,
133
+ "grad_norm": 1.7203434705734253,
134
+ "learning_rate": 1.949965059399022e-05,
135
+ "loss": 0.586,
136
+ "step": 180
137
+ },
138
+ {
139
+ "epoch": 0.13277428371767994,
140
+ "grad_norm": 1.333950161933899,
141
+ "learning_rate": 1.947169811320755e-05,
142
+ "loss": 0.6214,
143
+ "step": 190
144
+ },
145
+ {
146
+ "epoch": 0.13976240391334732,
147
+ "grad_norm": 1.5078257322311401,
148
+ "learning_rate": 1.944374563242488e-05,
149
+ "loss": 0.5652,
150
+ "step": 200
151
+ },
152
+ {
153
+ "epoch": 0.14675052410901468,
154
+ "grad_norm": 3.7645132541656494,
155
+ "learning_rate": 1.941579315164221e-05,
156
+ "loss": 0.5229,
157
+ "step": 210
158
+ },
159
+ {
160
+ "epoch": 0.15373864430468204,
161
+ "grad_norm": 1.2441000938415527,
162
+ "learning_rate": 1.938784067085954e-05,
163
+ "loss": 0.3669,
164
+ "step": 220
165
+ },
166
+ {
167
+ "epoch": 0.1607267645003494,
168
+ "grad_norm": 2.2875142097473145,
169
+ "learning_rate": 1.935988819007687e-05,
170
+ "loss": 0.4162,
171
+ "step": 230
172
+ },
173
+ {
174
+ "epoch": 0.16771488469601678,
175
+ "grad_norm": 2.1816585063934326,
176
+ "learning_rate": 1.9331935709294202e-05,
177
+ "loss": 0.3836,
178
+ "step": 240
179
+ },
180
+ {
181
+ "epoch": 0.17470300489168414,
182
+ "grad_norm": 3.4903249740600586,
183
+ "learning_rate": 1.9303983228511532e-05,
184
+ "loss": 0.4131,
185
+ "step": 250
186
+ },
187
+ {
188
+ "epoch": 0.1816911250873515,
189
+ "grad_norm": 2.1905159950256348,
190
+ "learning_rate": 1.9276030747728863e-05,
191
+ "loss": 0.3725,
192
+ "step": 260
193
+ },
194
+ {
195
+ "epoch": 0.18867924528301888,
196
+ "grad_norm": 3.6003665924072266,
197
+ "learning_rate": 1.9248078266946194e-05,
198
+ "loss": 0.4131,
199
+ "step": 270
200
+ },
201
+ {
202
+ "epoch": 0.19566736547868624,
203
+ "grad_norm": 3.012568235397339,
204
+ "learning_rate": 1.9220125786163524e-05,
205
+ "loss": 0.3991,
206
+ "step": 280
207
+ },
208
+ {
209
+ "epoch": 0.2026554856743536,
210
+ "grad_norm": 1.2049503326416016,
211
+ "learning_rate": 1.9192173305380855e-05,
212
+ "loss": 0.2732,
213
+ "step": 290
214
+ },
215
+ {
216
+ "epoch": 0.20964360587002095,
217
+ "grad_norm": 2.0943682193756104,
218
+ "learning_rate": 1.9164220824598185e-05,
219
+ "loss": 0.3052,
220
+ "step": 300
221
+ },
222
+ {
223
+ "epoch": 0.21663172606568834,
224
+ "grad_norm": 4.1970086097717285,
225
+ "learning_rate": 1.9136268343815516e-05,
226
+ "loss": 0.2544,
227
+ "step": 310
228
+ },
229
+ {
230
+ "epoch": 0.2236198462613557,
231
+ "grad_norm": 2.6826064586639404,
232
+ "learning_rate": 1.9108315863032846e-05,
233
+ "loss": 0.1921,
234
+ "step": 320
235
+ },
236
+ {
237
+ "epoch": 0.23060796645702306,
238
+ "grad_norm": 6.013642311096191,
239
+ "learning_rate": 1.9080363382250177e-05,
240
+ "loss": 0.3014,
241
+ "step": 330
242
+ },
243
+ {
244
+ "epoch": 0.2375960866526904,
245
+ "grad_norm": 4.2075629234313965,
246
+ "learning_rate": 1.9052410901467508e-05,
247
+ "loss": 0.2569,
248
+ "step": 340
249
+ },
250
+ {
251
+ "epoch": 0.2445842068483578,
252
+ "grad_norm": 0.5800023078918457,
253
+ "learning_rate": 1.9024458420684838e-05,
254
+ "loss": 0.1486,
255
+ "step": 350
256
+ },
257
+ {
258
+ "epoch": 0.25157232704402516,
259
+ "grad_norm": 8.123499870300293,
260
+ "learning_rate": 1.899650593990217e-05,
261
+ "loss": 0.2215,
262
+ "step": 360
263
+ },
264
+ {
265
+ "epoch": 0.2585604472396925,
266
+ "grad_norm": 4.439255237579346,
267
+ "learning_rate": 1.89685534591195e-05,
268
+ "loss": 0.1692,
269
+ "step": 370
270
+ },
271
+ {
272
+ "epoch": 0.2655485674353599,
273
+ "grad_norm": 4.177631855010986,
274
+ "learning_rate": 1.894060097833683e-05,
275
+ "loss": 0.1176,
276
+ "step": 380
277
+ },
278
+ {
279
+ "epoch": 0.27253668763102723,
280
+ "grad_norm": 0.5754140019416809,
281
+ "learning_rate": 1.891264849755416e-05,
282
+ "loss": 0.1538,
283
+ "step": 390
284
+ },
285
+ {
286
+ "epoch": 0.27952480782669464,
287
+ "grad_norm": 0.5612320303916931,
288
+ "learning_rate": 1.888469601677149e-05,
289
+ "loss": 0.1054,
290
+ "step": 400
291
+ },
292
+ {
293
+ "epoch": 0.286512928022362,
294
+ "grad_norm": 1.949729561805725,
295
+ "learning_rate": 1.8856743535988818e-05,
296
+ "loss": 0.0824,
297
+ "step": 410
298
+ },
299
+ {
300
+ "epoch": 0.29350104821802936,
301
+ "grad_norm": 3.153822898864746,
302
+ "learning_rate": 1.8828791055206152e-05,
303
+ "loss": 0.1174,
304
+ "step": 420
305
+ },
306
+ {
307
+ "epoch": 0.3004891684136967,
308
+ "grad_norm": 0.6545250415802002,
309
+ "learning_rate": 1.8800838574423483e-05,
310
+ "loss": 0.1305,
311
+ "step": 430
312
+ },
313
+ {
314
+ "epoch": 0.3074772886093641,
315
+ "grad_norm": 1.2621614933013916,
316
+ "learning_rate": 1.8772886093640813e-05,
317
+ "loss": 0.0684,
318
+ "step": 440
319
+ },
320
+ {
321
+ "epoch": 0.31446540880503143,
322
+ "grad_norm": 0.3028416335582733,
323
+ "learning_rate": 1.8744933612858144e-05,
324
+ "loss": 0.0863,
325
+ "step": 450
326
+ },
327
+ {
328
+ "epoch": 0.3214535290006988,
329
+ "grad_norm": 0.4452091455459595,
330
+ "learning_rate": 1.8716981132075474e-05,
331
+ "loss": 0.1016,
332
+ "step": 460
333
+ },
334
+ {
335
+ "epoch": 0.3284416491963662,
336
+ "grad_norm": 0.4046468436717987,
337
+ "learning_rate": 1.8689028651292805e-05,
338
+ "loss": 0.2317,
339
+ "step": 470
340
+ },
341
+ {
342
+ "epoch": 0.33542976939203356,
343
+ "grad_norm": 0.5121694803237915,
344
+ "learning_rate": 1.8661076170510135e-05,
345
+ "loss": 0.1965,
346
+ "step": 480
347
+ },
348
+ {
349
+ "epoch": 0.3424178895877009,
350
+ "grad_norm": 2.7892794609069824,
351
+ "learning_rate": 1.8633123689727466e-05,
352
+ "loss": 0.2082,
353
+ "step": 490
354
+ },
355
+ {
356
+ "epoch": 0.3494060097833683,
357
+ "grad_norm": 0.29012227058410645,
358
+ "learning_rate": 1.8605171208944793e-05,
359
+ "loss": 0.1393,
360
+ "step": 500
361
+ },
362
+ {
363
+ "epoch": 0.35639412997903563,
364
+ "grad_norm": 0.14969919621944427,
365
+ "learning_rate": 1.8577218728162124e-05,
366
+ "loss": 0.1917,
367
+ "step": 510
368
+ },
369
+ {
370
+ "epoch": 0.363382250174703,
371
+ "grad_norm": 2.5301618576049805,
372
+ "learning_rate": 1.8549266247379458e-05,
373
+ "loss": 0.0939,
374
+ "step": 520
375
+ },
376
+ {
377
+ "epoch": 0.37037037037037035,
378
+ "grad_norm": 0.5171930193901062,
379
+ "learning_rate": 1.8521313766596788e-05,
380
+ "loss": 0.0643,
381
+ "step": 530
382
+ },
383
+ {
384
+ "epoch": 0.37735849056603776,
385
+ "grad_norm": 0.23746764659881592,
386
+ "learning_rate": 1.849336128581412e-05,
387
+ "loss": 0.2094,
388
+ "step": 540
389
+ },
390
+ {
391
+ "epoch": 0.3843466107617051,
392
+ "grad_norm": 1.2422730922698975,
393
+ "learning_rate": 1.846540880503145e-05,
394
+ "loss": 0.109,
395
+ "step": 550
396
+ },
397
+ {
398
+ "epoch": 0.3913347309573725,
399
+ "grad_norm": 0.8466539978981018,
400
+ "learning_rate": 1.843745632424878e-05,
401
+ "loss": 0.1514,
402
+ "step": 560
403
+ },
404
+ {
405
+ "epoch": 0.39832285115303984,
406
+ "grad_norm": 0.201154425740242,
407
+ "learning_rate": 1.840950384346611e-05,
408
+ "loss": 0.1597,
409
+ "step": 570
410
+ },
411
+ {
412
+ "epoch": 0.4053109713487072,
413
+ "grad_norm": 0.19054199755191803,
414
+ "learning_rate": 1.838155136268344e-05,
415
+ "loss": 0.0955,
416
+ "step": 580
417
+ },
418
+ {
419
+ "epoch": 0.41229909154437455,
420
+ "grad_norm": 1.8036576509475708,
421
+ "learning_rate": 1.8353598881900768e-05,
422
+ "loss": 0.1369,
423
+ "step": 590
424
+ },
425
+ {
426
+ "epoch": 0.4192872117400419,
427
+ "grad_norm": 0.3122437596321106,
428
+ "learning_rate": 1.83256464011181e-05,
429
+ "loss": 0.1439,
430
+ "step": 600
431
+ },
432
+ {
433
+ "epoch": 0.42627533193570927,
434
+ "grad_norm": 2.6353561878204346,
435
+ "learning_rate": 1.829769392033543e-05,
436
+ "loss": 0.1566,
437
+ "step": 610
438
+ },
439
+ {
440
+ "epoch": 0.4332634521313767,
441
+ "grad_norm": 3.645707368850708,
442
+ "learning_rate": 1.8269741439552763e-05,
443
+ "loss": 0.2505,
444
+ "step": 620
445
+ },
446
+ {
447
+ "epoch": 0.44025157232704404,
448
+ "grad_norm": 0.20553793013095856,
449
+ "learning_rate": 1.8241788958770094e-05,
450
+ "loss": 0.0509,
451
+ "step": 630
452
+ },
453
+ {
454
+ "epoch": 0.4472396925227114,
455
+ "grad_norm": 3.8460299968719482,
456
+ "learning_rate": 1.8213836477987425e-05,
457
+ "loss": 0.118,
458
+ "step": 640
459
+ },
460
+ {
461
+ "epoch": 0.45422781271837875,
462
+ "grad_norm": 0.4825759828090668,
463
+ "learning_rate": 1.8185883997204755e-05,
464
+ "loss": 0.0671,
465
+ "step": 650
466
+ },
467
+ {
468
+ "epoch": 0.4612159329140461,
469
+ "grad_norm": 2.5147883892059326,
470
+ "learning_rate": 1.8157931516422086e-05,
471
+ "loss": 0.1381,
472
+ "step": 660
473
+ },
474
+ {
475
+ "epoch": 0.46820405310971347,
476
+ "grad_norm": 4.72428035736084,
477
+ "learning_rate": 1.8129979035639413e-05,
478
+ "loss": 0.1093,
479
+ "step": 670
480
+ },
481
+ {
482
+ "epoch": 0.4751921733053808,
483
+ "grad_norm": 3.091625928878784,
484
+ "learning_rate": 1.8102026554856743e-05,
485
+ "loss": 0.2363,
486
+ "step": 680
487
+ },
488
+ {
489
+ "epoch": 0.48218029350104824,
490
+ "grad_norm": 0.10304142534732819,
491
+ "learning_rate": 1.8074074074074074e-05,
492
+ "loss": 0.1347,
493
+ "step": 690
494
+ },
495
+ {
496
+ "epoch": 0.4891684136967156,
497
+ "grad_norm": 0.4455307424068451,
498
+ "learning_rate": 1.8046121593291405e-05,
499
+ "loss": 0.1064,
500
+ "step": 700
501
+ },
502
+ {
503
+ "epoch": 0.49615653389238296,
504
+ "grad_norm": 0.11594414710998535,
505
+ "learning_rate": 1.8018169112508735e-05,
506
+ "loss": 0.123,
507
+ "step": 710
508
+ },
509
+ {
510
+ "epoch": 0.5031446540880503,
511
+ "grad_norm": 0.1224653497338295,
512
+ "learning_rate": 1.799021663172607e-05,
513
+ "loss": 0.1263,
514
+ "step": 720
515
+ },
516
+ {
517
+ "epoch": 0.5101327742837177,
518
+ "grad_norm": 4.951889991760254,
519
+ "learning_rate": 1.79622641509434e-05,
520
+ "loss": 0.2111,
521
+ "step": 730
522
+ },
523
+ {
524
+ "epoch": 0.517120894479385,
525
+ "grad_norm": 0.16316205263137817,
526
+ "learning_rate": 1.793431167016073e-05,
527
+ "loss": 0.0366,
528
+ "step": 740
529
+ },
530
+ {
531
+ "epoch": 0.5241090146750524,
532
+ "grad_norm": 0.1896398663520813,
533
+ "learning_rate": 1.790635918937806e-05,
534
+ "loss": 0.1656,
535
+ "step": 750
536
+ },
537
+ {
538
+ "epoch": 0.5310971348707197,
539
+ "grad_norm": 0.3920069932937622,
540
+ "learning_rate": 1.7878406708595388e-05,
541
+ "loss": 0.2804,
542
+ "step": 760
543
+ },
544
+ {
545
+ "epoch": 0.5380852550663872,
546
+ "grad_norm": 0.12546010315418243,
547
+ "learning_rate": 1.785045422781272e-05,
548
+ "loss": 0.1404,
549
+ "step": 770
550
+ },
551
+ {
552
+ "epoch": 0.5450733752620545,
553
+ "grad_norm": 5.382510185241699,
554
+ "learning_rate": 1.782250174703005e-05,
555
+ "loss": 0.1194,
556
+ "step": 780
557
+ },
558
+ {
559
+ "epoch": 0.5520614954577219,
560
+ "grad_norm": 0.3077394366264343,
561
+ "learning_rate": 1.779454926624738e-05,
562
+ "loss": 0.1744,
563
+ "step": 790
564
+ },
565
+ {
566
+ "epoch": 0.5590496156533893,
567
+ "grad_norm": 0.41757112741470337,
568
+ "learning_rate": 1.776659678546471e-05,
569
+ "loss": 0.0729,
570
+ "step": 800
571
+ },
572
+ {
573
+ "epoch": 0.5660377358490566,
574
+ "grad_norm": 0.17300930619239807,
575
+ "learning_rate": 1.773864430468204e-05,
576
+ "loss": 0.1443,
577
+ "step": 810
578
+ },
579
+ {
580
+ "epoch": 0.573025856044724,
581
+ "grad_norm": 0.09419082850217819,
582
+ "learning_rate": 1.7710691823899375e-05,
583
+ "loss": 0.1587,
584
+ "step": 820
585
+ },
586
+ {
587
+ "epoch": 0.5800139762403913,
588
+ "grad_norm": 0.14113612473011017,
589
+ "learning_rate": 1.7682739343116705e-05,
590
+ "loss": 0.1196,
591
+ "step": 830
592
+ },
593
+ {
594
+ "epoch": 0.5870020964360587,
595
+ "grad_norm": 0.6690120697021484,
596
+ "learning_rate": 1.7654786862334036e-05,
597
+ "loss": 0.0619,
598
+ "step": 840
599
+ },
600
+ {
601
+ "epoch": 0.593990216631726,
602
+ "grad_norm": 0.5076020956039429,
603
+ "learning_rate": 1.7626834381551363e-05,
604
+ "loss": 0.2922,
605
+ "step": 850
606
+ },
607
+ {
608
+ "epoch": 0.6009783368273934,
609
+ "grad_norm": 0.12461690604686737,
610
+ "learning_rate": 1.7598881900768694e-05,
611
+ "loss": 0.0351,
612
+ "step": 860
613
+ },
614
+ {
615
+ "epoch": 0.6079664570230608,
616
+ "grad_norm": 5.236762523651123,
617
+ "learning_rate": 1.7570929419986024e-05,
618
+ "loss": 0.1571,
619
+ "step": 870
620
+ },
621
+ {
622
+ "epoch": 0.6149545772187281,
623
+ "grad_norm": 0.12950463593006134,
624
+ "learning_rate": 1.7542976939203355e-05,
625
+ "loss": 0.0204,
626
+ "step": 880
627
+ },
628
+ {
629
+ "epoch": 0.6219426974143956,
630
+ "grad_norm": 0.08026222884654999,
631
+ "learning_rate": 1.7515024458420685e-05,
632
+ "loss": 0.1185,
633
+ "step": 890
634
+ },
635
+ {
636
+ "epoch": 0.6289308176100629,
637
+ "grad_norm": 4.237085342407227,
638
+ "learning_rate": 1.7487071977638016e-05,
639
+ "loss": 0.0876,
640
+ "step": 900
641
+ },
642
+ {
643
+ "epoch": 0.6359189378057303,
644
+ "grad_norm": 0.916748046875,
645
+ "learning_rate": 1.7459119496855346e-05,
646
+ "loss": 0.0619,
647
+ "step": 910
648
+ },
649
+ {
650
+ "epoch": 0.6429070580013976,
651
+ "grad_norm": 4.51674747467041,
652
+ "learning_rate": 1.743116701607268e-05,
653
+ "loss": 0.1116,
654
+ "step": 920
655
+ },
656
+ {
657
+ "epoch": 0.649895178197065,
658
+ "grad_norm": 1.1469999551773071,
659
+ "learning_rate": 1.740321453529001e-05,
660
+ "loss": 0.0147,
661
+ "step": 930
662
+ },
663
+ {
664
+ "epoch": 0.6568832983927324,
665
+ "grad_norm": 2.3627333641052246,
666
+ "learning_rate": 1.7375262054507338e-05,
667
+ "loss": 0.1485,
668
+ "step": 940
669
+ },
670
+ {
671
+ "epoch": 0.6638714185883997,
672
+ "grad_norm": 8.216973304748535,
673
+ "learning_rate": 1.734730957372467e-05,
674
+ "loss": 0.0472,
675
+ "step": 950
676
+ },
677
+ {
678
+ "epoch": 0.6708595387840671,
679
+ "grad_norm": 3.5901684761047363,
680
+ "learning_rate": 1.7319357092942e-05,
681
+ "loss": 0.0979,
682
+ "step": 960
683
+ },
684
+ {
685
+ "epoch": 0.6778476589797344,
686
+ "grad_norm": 7.015125274658203,
687
+ "learning_rate": 1.729140461215933e-05,
688
+ "loss": 0.0537,
689
+ "step": 970
690
+ },
691
+ {
692
+ "epoch": 0.6848357791754018,
693
+ "grad_norm": 4.372157096862793,
694
+ "learning_rate": 1.726345213137666e-05,
695
+ "loss": 0.1152,
696
+ "step": 980
697
+ },
698
+ {
699
+ "epoch": 0.6918238993710691,
700
+ "grad_norm": 4.862729072570801,
701
+ "learning_rate": 1.723549965059399e-05,
702
+ "loss": 0.1131,
703
+ "step": 990
704
+ },
705
+ {
706
+ "epoch": 0.6988120195667366,
707
+ "grad_norm": 0.07732150703668594,
708
+ "learning_rate": 1.720754716981132e-05,
709
+ "loss": 0.0707,
710
+ "step": 1000
711
+ },
712
+ {
713
+ "epoch": 0.705800139762404,
714
+ "grad_norm": 0.19970574975013733,
715
+ "learning_rate": 1.7179594689028652e-05,
716
+ "loss": 0.0738,
717
+ "step": 1010
718
+ },
719
+ {
720
+ "epoch": 0.7127882599580713,
721
+ "grad_norm": 0.21677608788013458,
722
+ "learning_rate": 1.7151642208245983e-05,
723
+ "loss": 0.0992,
724
+ "step": 1020
725
+ },
726
+ {
727
+ "epoch": 0.7197763801537387,
728
+ "grad_norm": 0.10738188028335571,
729
+ "learning_rate": 1.7123689727463313e-05,
730
+ "loss": 0.0103,
731
+ "step": 1030
732
+ },
733
+ {
734
+ "epoch": 0.726764500349406,
735
+ "grad_norm": 3.7014143466949463,
736
+ "learning_rate": 1.7095737246680644e-05,
737
+ "loss": 0.0803,
738
+ "step": 1040
739
+ },
740
+ {
741
+ "epoch": 0.7337526205450734,
742
+ "grad_norm": 0.04619055986404419,
743
+ "learning_rate": 1.7067784765897974e-05,
744
+ "loss": 0.049,
745
+ "step": 1050
746
+ },
747
+ {
748
+ "epoch": 0.7407407407407407,
749
+ "grad_norm": 0.2753339409828186,
750
+ "learning_rate": 1.7039832285115305e-05,
751
+ "loss": 0.144,
752
+ "step": 1060
753
+ },
754
+ {
755
+ "epoch": 0.7477288609364081,
756
+ "grad_norm": 2.8835108280181885,
757
+ "learning_rate": 1.7011879804332635e-05,
758
+ "loss": 0.0885,
759
+ "step": 1070
760
+ },
761
+ {
762
+ "epoch": 0.7547169811320755,
763
+ "grad_norm": 0.22654394805431366,
764
+ "learning_rate": 1.6983927323549966e-05,
765
+ "loss": 0.0512,
766
+ "step": 1080
767
+ },
768
+ {
769
+ "epoch": 0.7617051013277428,
770
+ "grad_norm": 2.421400547027588,
771
+ "learning_rate": 1.6955974842767297e-05,
772
+ "loss": 0.1544,
773
+ "step": 1090
774
+ },
775
+ {
776
+ "epoch": 0.7686932215234102,
777
+ "grad_norm": 0.03806561604142189,
778
+ "learning_rate": 1.6928022361984627e-05,
779
+ "loss": 0.0225,
780
+ "step": 1100
781
+ },
782
+ {
783
+ "epoch": 0.7756813417190775,
784
+ "grad_norm": 0.08347579091787338,
785
+ "learning_rate": 1.6900069881201958e-05,
786
+ "loss": 0.1386,
787
+ "step": 1110
788
+ },
789
+ {
790
+ "epoch": 0.782669461914745,
791
+ "grad_norm": 0.2059401273727417,
792
+ "learning_rate": 1.687211740041929e-05,
793
+ "loss": 0.2406,
794
+ "step": 1120
795
+ },
796
+ {
797
+ "epoch": 0.7896575821104123,
798
+ "grad_norm": 0.060870036482810974,
799
+ "learning_rate": 1.684416491963662e-05,
800
+ "loss": 0.1158,
801
+ "step": 1130
802
+ },
803
+ {
804
+ "epoch": 0.7966457023060797,
805
+ "grad_norm": 0.14614495635032654,
806
+ "learning_rate": 1.681621243885395e-05,
807
+ "loss": 0.1938,
808
+ "step": 1140
809
+ },
810
+ {
811
+ "epoch": 0.803633822501747,
812
+ "grad_norm": 0.09593985229730606,
813
+ "learning_rate": 1.678825995807128e-05,
814
+ "loss": 0.0627,
815
+ "step": 1150
816
+ },
817
+ {
818
+ "epoch": 0.8106219426974144,
819
+ "grad_norm": 9.543107032775879,
820
+ "learning_rate": 1.676030747728861e-05,
821
+ "loss": 0.1365,
822
+ "step": 1160
823
+ },
824
+ {
825
+ "epoch": 0.8176100628930818,
826
+ "grad_norm": 0.09678485244512558,
827
+ "learning_rate": 1.673235499650594e-05,
828
+ "loss": 0.1438,
829
+ "step": 1170
830
+ },
831
+ {
832
+ "epoch": 0.8245981830887491,
833
+ "grad_norm": 0.24556872248649597,
834
+ "learning_rate": 1.6704402515723272e-05,
835
+ "loss": 0.2233,
836
+ "step": 1180
837
+ },
838
+ {
839
+ "epoch": 0.8315863032844165,
840
+ "grad_norm": 0.06892254948616028,
841
+ "learning_rate": 1.6676450034940602e-05,
842
+ "loss": 0.0601,
843
+ "step": 1190
844
+ },
845
+ {
846
+ "epoch": 0.8385744234800838,
847
+ "grad_norm": 0.6429911255836487,
848
+ "learning_rate": 1.6648497554157933e-05,
849
+ "loss": 0.132,
850
+ "step": 1200
851
+ },
852
+ {
853
+ "epoch": 0.8455625436757512,
854
+ "grad_norm": 0.03480253368616104,
855
+ "learning_rate": 1.6620545073375263e-05,
856
+ "loss": 0.0472,
857
+ "step": 1210
858
+ },
859
+ {
860
+ "epoch": 0.8525506638714185,
861
+ "grad_norm": 0.25101321935653687,
862
+ "learning_rate": 1.6592592592592594e-05,
863
+ "loss": 0.0485,
864
+ "step": 1220
865
+ },
866
+ {
867
+ "epoch": 0.859538784067086,
868
+ "grad_norm": 0.0525231771171093,
869
+ "learning_rate": 1.6564640111809925e-05,
870
+ "loss": 0.1824,
871
+ "step": 1230
872
+ },
873
+ {
874
+ "epoch": 0.8665269042627534,
875
+ "grad_norm": 4.694910526275635,
876
+ "learning_rate": 1.6536687631027255e-05,
877
+ "loss": 0.1666,
878
+ "step": 1240
879
+ },
880
+ {
881
+ "epoch": 0.8735150244584207,
882
+ "grad_norm": 0.09604217112064362,
883
+ "learning_rate": 1.6508735150244586e-05,
884
+ "loss": 0.0638,
885
+ "step": 1250
886
+ },
887
+ {
888
+ "epoch": 0.8805031446540881,
889
+ "grad_norm": 9.125068664550781,
890
+ "learning_rate": 1.6480782669461916e-05,
891
+ "loss": 0.0744,
892
+ "step": 1260
893
+ },
894
+ {
895
+ "epoch": 0.8874912648497554,
896
+ "grad_norm": 0.5258679986000061,
897
+ "learning_rate": 1.6452830188679247e-05,
898
+ "loss": 0.019,
899
+ "step": 1270
900
+ },
901
+ {
902
+ "epoch": 0.8944793850454228,
903
+ "grad_norm": 3.6146976947784424,
904
+ "learning_rate": 1.6424877707896577e-05,
905
+ "loss": 0.0141,
906
+ "step": 1280
907
+ },
908
+ {
909
+ "epoch": 0.9014675052410901,
910
+ "grad_norm": 0.05512338504195213,
911
+ "learning_rate": 1.6396925227113908e-05,
912
+ "loss": 0.0138,
913
+ "step": 1290
914
+ },
915
+ {
916
+ "epoch": 0.9084556254367575,
917
+ "grad_norm": 0.028468603268265724,
918
+ "learning_rate": 1.636897274633124e-05,
919
+ "loss": 0.0815,
920
+ "step": 1300
921
+ },
922
+ {
923
+ "epoch": 0.9154437456324249,
924
+ "grad_norm": 0.7173348665237427,
925
+ "learning_rate": 1.634102026554857e-05,
926
+ "loss": 0.0919,
927
+ "step": 1310
928
+ },
929
+ {
930
+ "epoch": 0.9224318658280922,
931
+ "grad_norm": 0.06138516217470169,
932
+ "learning_rate": 1.63130677847659e-05,
933
+ "loss": 0.1383,
934
+ "step": 1320
935
+ },
936
+ {
937
+ "epoch": 0.9294199860237596,
938
+ "grad_norm": 10.367135047912598,
939
+ "learning_rate": 1.628511530398323e-05,
940
+ "loss": 0.1265,
941
+ "step": 1330
942
+ },
943
+ {
944
+ "epoch": 0.9364081062194269,
945
+ "grad_norm": 0.3321629762649536,
946
+ "learning_rate": 1.625716282320056e-05,
947
+ "loss": 0.1149,
948
+ "step": 1340
949
+ },
950
+ {
951
+ "epoch": 0.9433962264150944,
952
+ "grad_norm": 0.21935391426086426,
953
+ "learning_rate": 1.622921034241789e-05,
954
+ "loss": 0.1147,
955
+ "step": 1350
956
+ },
957
+ {
958
+ "epoch": 0.9503843466107617,
959
+ "grad_norm": 4.630390644073486,
960
+ "learning_rate": 1.6201257861635222e-05,
961
+ "loss": 0.1159,
962
+ "step": 1360
963
+ },
964
+ {
965
+ "epoch": 0.9573724668064291,
966
+ "grad_norm": 0.03748277202248573,
967
+ "learning_rate": 1.6173305380852552e-05,
968
+ "loss": 0.0029,
969
+ "step": 1370
970
+ },
971
+ {
972
+ "epoch": 0.9643605870020965,
973
+ "grad_norm": 0.09677435457706451,
974
+ "learning_rate": 1.6145352900069883e-05,
975
+ "loss": 0.179,
976
+ "step": 1380
977
+ },
978
+ {
979
+ "epoch": 0.9713487071977638,
980
+ "grad_norm": 0.022979794070124626,
981
+ "learning_rate": 1.6117400419287214e-05,
982
+ "loss": 0.0034,
983
+ "step": 1390
984
+ },
985
+ {
986
+ "epoch": 0.9783368273934312,
987
+ "grad_norm": 3.3972229957580566,
988
+ "learning_rate": 1.6089447938504544e-05,
989
+ "loss": 0.1633,
990
+ "step": 1400
991
+ },
992
+ {
993
+ "epoch": 0.9853249475890985,
994
+ "grad_norm": 0.0410487987101078,
995
+ "learning_rate": 1.6061495457721875e-05,
996
+ "loss": 0.0758,
997
+ "step": 1410
998
+ },
999
+ {
1000
+ "epoch": 0.9923130677847659,
1001
+ "grad_norm": 0.1519143134355545,
1002
+ "learning_rate": 1.6033542976939205e-05,
1003
+ "loss": 0.0967,
1004
+ "step": 1420
1005
+ },
1006
+ {
1007
+ "epoch": 0.9993011879804332,
1008
+ "grad_norm": 0.11000066250562668,
1009
+ "learning_rate": 1.6005590496156536e-05,
1010
+ "loss": 0.1054,
1011
+ "step": 1430
1012
+ },
1013
+ {
1014
+ "epoch": 1.0062893081761006,
1015
+ "grad_norm": 9.595348358154297,
1016
+ "learning_rate": 1.5977638015373866e-05,
1017
+ "loss": 0.0367,
1018
+ "step": 1440
1019
+ },
1020
+ {
1021
+ "epoch": 1.013277428371768,
1022
+ "grad_norm": 0.044866789132356644,
1023
+ "learning_rate": 1.5949685534591197e-05,
1024
+ "loss": 0.145,
1025
+ "step": 1450
1026
+ },
1027
+ {
1028
+ "epoch": 1.0202655485674355,
1029
+ "grad_norm": 6.3100361824035645,
1030
+ "learning_rate": 1.5921733053808524e-05,
1031
+ "loss": 0.2702,
1032
+ "step": 1460
1033
+ },
1034
+ {
1035
+ "epoch": 1.0272536687631026,
1036
+ "grad_norm": 1.0666788816452026,
1037
+ "learning_rate": 1.5893780573025858e-05,
1038
+ "loss": 0.1257,
1039
+ "step": 1470
1040
+ },
1041
+ {
1042
+ "epoch": 1.03424178895877,
1043
+ "grad_norm": 15.676182746887207,
1044
+ "learning_rate": 1.586582809224319e-05,
1045
+ "loss": 0.0373,
1046
+ "step": 1480
1047
+ },
1048
+ {
1049
+ "epoch": 1.0412299091544375,
1050
+ "grad_norm": 13.464042663574219,
1051
+ "learning_rate": 1.583787561146052e-05,
1052
+ "loss": 0.1107,
1053
+ "step": 1490
1054
+ },
1055
+ {
1056
+ "epoch": 1.0482180293501049,
1057
+ "grad_norm": 0.029935991391539574,
1058
+ "learning_rate": 1.580992313067785e-05,
1059
+ "loss": 0.0778,
1060
+ "step": 1500
1061
+ },
1062
+ {
1063
+ "epoch": 1.0552061495457723,
1064
+ "grad_norm": 1.0842170715332031,
1065
+ "learning_rate": 1.578197064989518e-05,
1066
+ "loss": 0.2003,
1067
+ "step": 1510
1068
+ },
1069
+ {
1070
+ "epoch": 1.0621942697414395,
1071
+ "grad_norm": 0.8753694891929626,
1072
+ "learning_rate": 1.575401816911251e-05,
1073
+ "loss": 0.0869,
1074
+ "step": 1520
1075
+ },
1076
+ {
1077
+ "epoch": 1.069182389937107,
1078
+ "grad_norm": 0.03639127314090729,
1079
+ "learning_rate": 1.572606568832984e-05,
1080
+ "loss": 0.0035,
1081
+ "step": 1530
1082
+ },
1083
+ {
1084
+ "epoch": 1.0761705101327743,
1085
+ "grad_norm": 0.10639658570289612,
1086
+ "learning_rate": 1.5698113207547172e-05,
1087
+ "loss": 0.0709,
1088
+ "step": 1540
1089
+ },
1090
+ {
1091
+ "epoch": 1.0831586303284417,
1092
+ "grad_norm": 0.09013110399246216,
1093
+ "learning_rate": 1.56701607267645e-05,
1094
+ "loss": 0.0041,
1095
+ "step": 1550
1096
+ },
1097
+ {
1098
+ "epoch": 1.090146750524109,
1099
+ "grad_norm": 0.1398763209581375,
1100
+ "learning_rate": 1.564220824598183e-05,
1101
+ "loss": 0.0039,
1102
+ "step": 1560
1103
+ },
1104
+ {
1105
+ "epoch": 1.0971348707197763,
1106
+ "grad_norm": 0.13840065896511078,
1107
+ "learning_rate": 1.5614255765199164e-05,
1108
+ "loss": 0.0288,
1109
+ "step": 1570
1110
+ },
1111
+ {
1112
+ "epoch": 1.1041229909154437,
1113
+ "grad_norm": 0.03229377046227455,
1114
+ "learning_rate": 1.5586303284416494e-05,
1115
+ "loss": 0.023,
1116
+ "step": 1580
1117
+ },
1118
+ {
1119
+ "epoch": 1.1111111111111112,
1120
+ "grad_norm": 0.09332744032144547,
1121
+ "learning_rate": 1.5558350803633825e-05,
1122
+ "loss": 0.148,
1123
+ "step": 1590
1124
+ },
1125
+ {
1126
+ "epoch": 1.1180992313067786,
1127
+ "grad_norm": 0.24537114799022675,
1128
+ "learning_rate": 1.5530398322851156e-05,
1129
+ "loss": 0.0143,
1130
+ "step": 1600
1131
+ },
1132
+ {
1133
+ "epoch": 1.1250873515024458,
1134
+ "grad_norm": 0.02455071546137333,
1135
+ "learning_rate": 1.5502445842068486e-05,
1136
+ "loss": 0.0378,
1137
+ "step": 1610
1138
+ },
1139
+ {
1140
+ "epoch": 1.1320754716981132,
1141
+ "grad_norm": 0.027023492380976677,
1142
+ "learning_rate": 1.5474493361285817e-05,
1143
+ "loss": 0.0558,
1144
+ "step": 1620
1145
+ },
1146
+ {
1147
+ "epoch": 1.1390635918937806,
1148
+ "grad_norm": 0.048673246055841446,
1149
+ "learning_rate": 1.5446540880503147e-05,
1150
+ "loss": 0.0519,
1151
+ "step": 1630
1152
+ },
1153
+ {
1154
+ "epoch": 1.146051712089448,
1155
+ "grad_norm": 8.062767028808594,
1156
+ "learning_rate": 1.5418588399720474e-05,
1157
+ "loss": 0.0404,
1158
+ "step": 1640
1159
+ },
1160
+ {
1161
+ "epoch": 1.1530398322851152,
1162
+ "grad_norm": 8.358656883239746,
1163
+ "learning_rate": 1.5390635918937805e-05,
1164
+ "loss": 0.1054,
1165
+ "step": 1650
1166
+ },
1167
+ {
1168
+ "epoch": 1.1600279524807826,
1169
+ "grad_norm": 1.3057444095611572,
1170
+ "learning_rate": 1.5362683438155136e-05,
1171
+ "loss": 0.1049,
1172
+ "step": 1660
1173
+ },
1174
+ {
1175
+ "epoch": 1.16701607267645,
1176
+ "grad_norm": 0.018700115382671356,
1177
+ "learning_rate": 1.533473095737247e-05,
1178
+ "loss": 0.0106,
1179
+ "step": 1670
1180
+ },
1181
+ {
1182
+ "epoch": 1.1740041928721174,
1183
+ "grad_norm": 8.731624603271484,
1184
+ "learning_rate": 1.53067784765898e-05,
1185
+ "loss": 0.0541,
1186
+ "step": 1680
1187
+ },
1188
+ {
1189
+ "epoch": 1.1809923130677848,
1190
+ "grad_norm": 1.358834147453308,
1191
+ "learning_rate": 1.527882599580713e-05,
1192
+ "loss": 0.1515,
1193
+ "step": 1690
1194
+ },
1195
+ {
1196
+ "epoch": 1.187980433263452,
1197
+ "grad_norm": 0.05055036395788193,
1198
+ "learning_rate": 1.525087351502446e-05,
1199
+ "loss": 0.1181,
1200
+ "step": 1700
1201
+ },
1202
+ {
1203
+ "epoch": 1.1949685534591195,
1204
+ "grad_norm": 0.19085977971553802,
1205
+ "learning_rate": 1.5222921034241792e-05,
1206
+ "loss": 0.0624,
1207
+ "step": 1710
1208
+ },
1209
+ {
1210
+ "epoch": 1.2019566736547869,
1211
+ "grad_norm": 1.282493233680725,
1212
+ "learning_rate": 1.5194968553459122e-05,
1213
+ "loss": 0.016,
1214
+ "step": 1720
1215
+ },
1216
+ {
1217
+ "epoch": 1.2089447938504543,
1218
+ "grad_norm": 0.03596067801117897,
1219
+ "learning_rate": 1.5167016072676451e-05,
1220
+ "loss": 0.0211,
1221
+ "step": 1730
1222
+ },
1223
+ {
1224
+ "epoch": 1.2159329140461215,
1225
+ "grad_norm": 0.033778220415115356,
1226
+ "learning_rate": 1.5139063591893782e-05,
1227
+ "loss": 0.0105,
1228
+ "step": 1740
1229
+ },
1230
+ {
1231
+ "epoch": 1.2229210342417889,
1232
+ "grad_norm": 0.022970452904701233,
1233
+ "learning_rate": 1.5111111111111112e-05,
1234
+ "loss": 0.1634,
1235
+ "step": 1750
1236
+ },
1237
+ {
1238
+ "epoch": 1.2299091544374563,
1239
+ "grad_norm": 0.08840584754943848,
1240
+ "learning_rate": 1.5083158630328443e-05,
1241
+ "loss": 0.0062,
1242
+ "step": 1760
1243
+ },
1244
+ {
1245
+ "epoch": 1.2368972746331237,
1246
+ "grad_norm": 0.0703321248292923,
1247
+ "learning_rate": 1.5055206149545773e-05,
1248
+ "loss": 0.0554,
1249
+ "step": 1770
1250
+ },
1251
+ {
1252
+ "epoch": 1.2438853948287911,
1253
+ "grad_norm": 2.026752471923828,
1254
+ "learning_rate": 1.5027253668763104e-05,
1255
+ "loss": 0.03,
1256
+ "step": 1780
1257
+ },
1258
+ {
1259
+ "epoch": 1.2508735150244585,
1260
+ "grad_norm": 0.08599984645843506,
1261
+ "learning_rate": 1.4999301187980435e-05,
1262
+ "loss": 0.0303,
1263
+ "step": 1790
1264
+ },
1265
+ {
1266
+ "epoch": 1.2578616352201257,
1267
+ "grad_norm": 0.0998782068490982,
1268
+ "learning_rate": 1.4971348707197765e-05,
1269
+ "loss": 0.1389,
1270
+ "step": 1800
1271
+ },
1272
+ {
1273
+ "epoch": 1.2648497554157931,
1274
+ "grad_norm": 1.7642574310302734,
1275
+ "learning_rate": 1.4943396226415094e-05,
1276
+ "loss": 0.0556,
1277
+ "step": 1810
1278
+ },
1279
+ {
1280
+ "epoch": 1.2718378756114606,
1281
+ "grad_norm": 0.012091516517102718,
1282
+ "learning_rate": 1.4915443745632425e-05,
1283
+ "loss": 0.1116,
1284
+ "step": 1820
1285
+ },
1286
+ {
1287
+ "epoch": 1.2788259958071277,
1288
+ "grad_norm": 0.030048305168747902,
1289
+ "learning_rate": 1.4887491264849757e-05,
1290
+ "loss": 0.0433,
1291
+ "step": 1830
1292
+ },
1293
+ {
1294
+ "epoch": 1.2858141160027952,
1295
+ "grad_norm": 0.025840098038315773,
1296
+ "learning_rate": 1.4859538784067087e-05,
1297
+ "loss": 0.038,
1298
+ "step": 1840
1299
+ },
1300
+ {
1301
+ "epoch": 1.2928022361984626,
1302
+ "grad_norm": 0.4607846736907959,
1303
+ "learning_rate": 1.4831586303284418e-05,
1304
+ "loss": 0.0815,
1305
+ "step": 1850
1306
+ },
1307
+ {
1308
+ "epoch": 1.29979035639413,
1309
+ "grad_norm": 3.925401210784912,
1310
+ "learning_rate": 1.4803633822501749e-05,
1311
+ "loss": 0.0708,
1312
+ "step": 1860
1313
+ },
1314
+ {
1315
+ "epoch": 1.3067784765897974,
1316
+ "grad_norm": 5.653714656829834,
1317
+ "learning_rate": 1.4775681341719079e-05,
1318
+ "loss": 0.0166,
1319
+ "step": 1870
1320
+ },
1321
+ {
1322
+ "epoch": 1.3137665967854648,
1323
+ "grad_norm": 0.02135496586561203,
1324
+ "learning_rate": 1.474772886093641e-05,
1325
+ "loss": 0.0782,
1326
+ "step": 1880
1327
+ },
1328
+ {
1329
+ "epoch": 1.320754716981132,
1330
+ "grad_norm": 0.020742664113640785,
1331
+ "learning_rate": 1.471977638015374e-05,
1332
+ "loss": 0.0048,
1333
+ "step": 1890
1334
+ },
1335
+ {
1336
+ "epoch": 1.3277428371767994,
1337
+ "grad_norm": 0.02374901808798313,
1338
+ "learning_rate": 1.4691823899371069e-05,
1339
+ "loss": 0.081,
1340
+ "step": 1900
1341
+ },
1342
+ {
1343
+ "epoch": 1.3347309573724668,
1344
+ "grad_norm": 0.03886565566062927,
1345
+ "learning_rate": 1.46638714185884e-05,
1346
+ "loss": 0.005,
1347
+ "step": 1910
1348
+ },
1349
+ {
1350
+ "epoch": 1.3417190775681342,
1351
+ "grad_norm": 0.16429579257965088,
1352
+ "learning_rate": 1.463591893780573e-05,
1353
+ "loss": 0.0154,
1354
+ "step": 1920
1355
+ },
1356
+ {
1357
+ "epoch": 1.3487071977638014,
1358
+ "grad_norm": 0.15556152164936066,
1359
+ "learning_rate": 1.4607966457023063e-05,
1360
+ "loss": 0.0659,
1361
+ "step": 1930
1362
+ },
1363
+ {
1364
+ "epoch": 1.3556953179594688,
1365
+ "grad_norm": 0.022907955572009087,
1366
+ "learning_rate": 1.4580013976240393e-05,
1367
+ "loss": 0.068,
1368
+ "step": 1940
1369
+ },
1370
+ {
1371
+ "epoch": 1.3626834381551363,
1372
+ "grad_norm": 0.028623567894101143,
1373
+ "learning_rate": 1.4552061495457724e-05,
1374
+ "loss": 0.0023,
1375
+ "step": 1950
1376
+ },
1377
+ {
1378
+ "epoch": 1.3696715583508037,
1379
+ "grad_norm": 7.010847091674805,
1380
+ "learning_rate": 1.4524109014675054e-05,
1381
+ "loss": 0.147,
1382
+ "step": 1960
1383
+ },
1384
+ {
1385
+ "epoch": 1.376659678546471,
1386
+ "grad_norm": 0.03581179678440094,
1387
+ "learning_rate": 1.4496156533892385e-05,
1388
+ "loss": 0.1209,
1389
+ "step": 1970
1390
+ },
1391
+ {
1392
+ "epoch": 1.3836477987421385,
1393
+ "grad_norm": 0.019916867837309837,
1394
+ "learning_rate": 1.4468204053109715e-05,
1395
+ "loss": 0.0233,
1396
+ "step": 1980
1397
+ },
1398
+ {
1399
+ "epoch": 1.3906359189378057,
1400
+ "grad_norm": 10.045747756958008,
1401
+ "learning_rate": 1.4440251572327044e-05,
1402
+ "loss": 0.0416,
1403
+ "step": 1990
1404
+ },
1405
+ {
1406
+ "epoch": 1.397624039133473,
1407
+ "grad_norm": 0.04035484418272972,
1408
+ "learning_rate": 1.4412299091544375e-05,
1409
+ "loss": 0.0265,
1410
+ "step": 2000
1411
+ },
1412
+ {
1413
+ "epoch": 1.4046121593291405,
1414
+ "grad_norm": 0.7727859616279602,
1415
+ "learning_rate": 1.4384346610761705e-05,
1416
+ "loss": 0.0331,
1417
+ "step": 2010
1418
+ },
1419
+ {
1420
+ "epoch": 1.4116002795248077,
1421
+ "grad_norm": 0.018749171867966652,
1422
+ "learning_rate": 1.4356394129979036e-05,
1423
+ "loss": 0.0749,
1424
+ "step": 2020
1425
+ },
1426
+ {
1427
+ "epoch": 1.4185883997204751,
1428
+ "grad_norm": 5.582619667053223,
1429
+ "learning_rate": 1.4328441649196368e-05,
1430
+ "loss": 0.0622,
1431
+ "step": 2030
1432
+ },
1433
+ {
1434
+ "epoch": 1.4255765199161425,
1435
+ "grad_norm": 0.03347029909491539,
1436
+ "learning_rate": 1.4300489168413699e-05,
1437
+ "loss": 0.0021,
1438
+ "step": 2040
1439
+ },
1440
+ {
1441
+ "epoch": 1.43256464011181,
1442
+ "grad_norm": 0.1696143001317978,
1443
+ "learning_rate": 1.427253668763103e-05,
1444
+ "loss": 0.0021,
1445
+ "step": 2050
1446
+ },
1447
+ {
1448
+ "epoch": 1.4395527603074774,
1449
+ "grad_norm": 0.021968642249703407,
1450
+ "learning_rate": 1.424458420684836e-05,
1451
+ "loss": 0.0025,
1452
+ "step": 2060
1453
+ },
1454
+ {
1455
+ "epoch": 1.4465408805031448,
1456
+ "grad_norm": 0.007907007820904255,
1457
+ "learning_rate": 1.421663172606569e-05,
1458
+ "loss": 0.0478,
1459
+ "step": 2070
1460
+ },
1461
+ {
1462
+ "epoch": 1.453529000698812,
1463
+ "grad_norm": 0.035767070949077606,
1464
+ "learning_rate": 1.418867924528302e-05,
1465
+ "loss": 0.0012,
1466
+ "step": 2080
1467
+ },
1468
+ {
1469
+ "epoch": 1.4605171208944794,
1470
+ "grad_norm": 0.011390830390155315,
1471
+ "learning_rate": 1.416072676450035e-05,
1472
+ "loss": 0.1533,
1473
+ "step": 2090
1474
+ },
1475
+ {
1476
+ "epoch": 1.4675052410901468,
1477
+ "grad_norm": 7.848016738891602,
1478
+ "learning_rate": 1.413277428371768e-05,
1479
+ "loss": 0.0675,
1480
+ "step": 2100
1481
+ },
1482
+ {
1483
+ "epoch": 1.474493361285814,
1484
+ "grad_norm": 0.011802544817328453,
1485
+ "learning_rate": 1.4104821802935011e-05,
1486
+ "loss": 0.0261,
1487
+ "step": 2110
1488
+ },
1489
+ {
1490
+ "epoch": 1.4814814814814814,
1491
+ "grad_norm": 2.9148340225219727,
1492
+ "learning_rate": 1.4076869322152342e-05,
1493
+ "loss": 0.2831,
1494
+ "step": 2120
1495
+ },
1496
+ {
1497
+ "epoch": 1.4884696016771488,
1498
+ "grad_norm": 0.08199399709701538,
1499
+ "learning_rate": 1.4048916841369674e-05,
1500
+ "loss": 0.0088,
1501
+ "step": 2130
1502
+ },
1503
+ {
1504
+ "epoch": 1.4954577218728162,
1505
+ "grad_norm": 0.01199902594089508,
1506
+ "learning_rate": 1.4020964360587004e-05,
1507
+ "loss": 0.0693,
1508
+ "step": 2140
1509
+ },
1510
+ {
1511
+ "epoch": 1.5024458420684836,
1512
+ "grad_norm": 0.07146008312702179,
1513
+ "learning_rate": 1.3993011879804335e-05,
1514
+ "loss": 0.0746,
1515
+ "step": 2150
1516
+ },
1517
+ {
1518
+ "epoch": 1.509433962264151,
1519
+ "grad_norm": 0.02118564024567604,
1520
+ "learning_rate": 1.3965059399021666e-05,
1521
+ "loss": 0.003,
1522
+ "step": 2160
1523
+ },
1524
+ {
1525
+ "epoch": 1.5164220824598185,
1526
+ "grad_norm": 0.016972247511148453,
1527
+ "learning_rate": 1.3937106918238994e-05,
1528
+ "loss": 0.0106,
1529
+ "step": 2170
1530
+ },
1531
+ {
1532
+ "epoch": 1.5234102026554857,
1533
+ "grad_norm": 0.023427631705999374,
1534
+ "learning_rate": 1.3909154437456325e-05,
1535
+ "loss": 0.0444,
1536
+ "step": 2180
1537
+ },
1538
+ {
1539
+ "epoch": 1.530398322851153,
1540
+ "grad_norm": 0.03280281275510788,
1541
+ "learning_rate": 1.3881201956673656e-05,
1542
+ "loss": 0.0026,
1543
+ "step": 2190
1544
+ },
1545
+ {
1546
+ "epoch": 1.5373864430468203,
1547
+ "grad_norm": 0.06073632091283798,
1548
+ "learning_rate": 1.3853249475890986e-05,
1549
+ "loss": 0.194,
1550
+ "step": 2200
1551
+ },
1552
+ {
1553
+ "epoch": 1.5443745632424877,
1554
+ "grad_norm": 0.008114098571240902,
1555
+ "learning_rate": 1.3825296995108317e-05,
1556
+ "loss": 0.0016,
1557
+ "step": 2210
1558
+ },
1559
+ {
1560
+ "epoch": 1.551362683438155,
1561
+ "grad_norm": 0.0467991828918457,
1562
+ "learning_rate": 1.3797344514325647e-05,
1563
+ "loss": 0.1305,
1564
+ "step": 2220
1565
+ },
1566
+ {
1567
+ "epoch": 1.5583508036338225,
1568
+ "grad_norm": 0.06984475255012512,
1569
+ "learning_rate": 1.376939203354298e-05,
1570
+ "loss": 0.1386,
1571
+ "step": 2230
1572
+ },
1573
+ {
1574
+ "epoch": 1.56533892382949,
1575
+ "grad_norm": 0.014983494766056538,
1576
+ "learning_rate": 1.374143955276031e-05,
1577
+ "loss": 0.1399,
1578
+ "step": 2240
1579
+ },
1580
+ {
1581
+ "epoch": 1.5723270440251573,
1582
+ "grad_norm": 0.01568700559437275,
1583
+ "learning_rate": 1.3713487071977637e-05,
1584
+ "loss": 0.1084,
1585
+ "step": 2250
1586
+ },
1587
+ {
1588
+ "epoch": 1.5793151642208247,
1589
+ "grad_norm": 10.549098014831543,
1590
+ "learning_rate": 1.368553459119497e-05,
1591
+ "loss": 0.0082,
1592
+ "step": 2260
1593
+ },
1594
+ {
1595
+ "epoch": 1.586303284416492,
1596
+ "grad_norm": 0.02495586685836315,
1597
+ "learning_rate": 1.36575821104123e-05,
1598
+ "loss": 0.0019,
1599
+ "step": 2270
1600
+ },
1601
+ {
1602
+ "epoch": 1.5932914046121593,
1603
+ "grad_norm": 0.01876898854970932,
1604
+ "learning_rate": 1.362962962962963e-05,
1605
+ "loss": 0.1002,
1606
+ "step": 2280
1607
+ },
1608
+ {
1609
+ "epoch": 1.6002795248078265,
1610
+ "grad_norm": 0.044464971870183945,
1611
+ "learning_rate": 1.3601677148846961e-05,
1612
+ "loss": 0.0019,
1613
+ "step": 2290
1614
+ },
1615
+ {
1616
+ "epoch": 1.607267645003494,
1617
+ "grad_norm": 0.04108593240380287,
1618
+ "learning_rate": 1.3573724668064292e-05,
1619
+ "loss": 0.0753,
1620
+ "step": 2300
1621
+ },
1622
+ {
1623
+ "epoch": 1.6142557651991614,
1624
+ "grad_norm": 7.812976837158203,
1625
+ "learning_rate": 1.3545772187281622e-05,
1626
+ "loss": 0.2875,
1627
+ "step": 2310
1628
+ },
1629
+ {
1630
+ "epoch": 1.6212438853948288,
1631
+ "grad_norm": 4.261682987213135,
1632
+ "learning_rate": 1.3517819706498953e-05,
1633
+ "loss": 0.1059,
1634
+ "step": 2320
1635
+ },
1636
+ {
1637
+ "epoch": 1.6282320055904962,
1638
+ "grad_norm": 0.03309008479118347,
1639
+ "learning_rate": 1.3489867225716285e-05,
1640
+ "loss": 0.0051,
1641
+ "step": 2330
1642
+ },
1643
+ {
1644
+ "epoch": 1.6352201257861636,
1645
+ "grad_norm": 9.983598709106445,
1646
+ "learning_rate": 1.3461914744933612e-05,
1647
+ "loss": 0.0249,
1648
+ "step": 2340
1649
+ },
1650
+ {
1651
+ "epoch": 1.642208245981831,
1652
+ "grad_norm": 4.973055839538574,
1653
+ "learning_rate": 1.3433962264150943e-05,
1654
+ "loss": 0.1441,
1655
+ "step": 2350
1656
+ },
1657
+ {
1658
+ "epoch": 1.6491963661774982,
1659
+ "grad_norm": 9.258291244506836,
1660
+ "learning_rate": 1.3406009783368275e-05,
1661
+ "loss": 0.1151,
1662
+ "step": 2360
1663
+ },
1664
+ {
1665
+ "epoch": 1.6561844863731656,
1666
+ "grad_norm": 0.015135078690946102,
1667
+ "learning_rate": 1.3378057302585606e-05,
1668
+ "loss": 0.0458,
1669
+ "step": 2370
1670
+ },
1671
+ {
1672
+ "epoch": 1.663172606568833,
1673
+ "grad_norm": 0.027922656387090683,
1674
+ "learning_rate": 1.3350104821802936e-05,
1675
+ "loss": 0.012,
1676
+ "step": 2380
1677
+ },
1678
+ {
1679
+ "epoch": 1.6701607267645002,
1680
+ "grad_norm": 0.043312493711709976,
1681
+ "learning_rate": 1.3322152341020267e-05,
1682
+ "loss": 0.1028,
1683
+ "step": 2390
1684
+ },
1685
+ {
1686
+ "epoch": 1.6771488469601676,
1687
+ "grad_norm": 0.012087292037904263,
1688
+ "learning_rate": 1.3294199860237597e-05,
1689
+ "loss": 0.0475,
1690
+ "step": 2400
1691
+ },
1692
+ {
1693
+ "epoch": 1.684136967155835,
1694
+ "grad_norm": 0.5556979775428772,
1695
+ "learning_rate": 1.3266247379454928e-05,
1696
+ "loss": 0.0142,
1697
+ "step": 2410
1698
+ },
1699
+ {
1700
+ "epoch": 1.6911250873515025,
1701
+ "grad_norm": 5.196042537689209,
1702
+ "learning_rate": 1.3238294898672259e-05,
1703
+ "loss": 0.1002,
1704
+ "step": 2420
1705
+ },
1706
+ {
1707
+ "epoch": 1.6981132075471699,
1708
+ "grad_norm": 0.010861317627131939,
1709
+ "learning_rate": 1.3210342417889587e-05,
1710
+ "loss": 0.0953,
1711
+ "step": 2430
1712
+ },
1713
+ {
1714
+ "epoch": 1.7051013277428373,
1715
+ "grad_norm": 0.017955880612134933,
1716
+ "learning_rate": 1.3182389937106918e-05,
1717
+ "loss": 0.0642,
1718
+ "step": 2440
1719
+ },
1720
+ {
1721
+ "epoch": 1.7120894479385047,
1722
+ "grad_norm": 0.028966449201107025,
1723
+ "learning_rate": 1.3154437456324249e-05,
1724
+ "loss": 0.1619,
1725
+ "step": 2450
1726
+ },
1727
+ {
1728
+ "epoch": 1.719077568134172,
1729
+ "grad_norm": 0.8712167739868164,
1730
+ "learning_rate": 1.312648497554158e-05,
1731
+ "loss": 0.0324,
1732
+ "step": 2460
1733
+ },
1734
+ {
1735
+ "epoch": 1.7260656883298393,
1736
+ "grad_norm": 0.2874550521373749,
1737
+ "learning_rate": 1.3098532494758911e-05,
1738
+ "loss": 0.026,
1739
+ "step": 2470
1740
+ },
1741
+ {
1742
+ "epoch": 1.7330538085255065,
1743
+ "grad_norm": 0.010579444468021393,
1744
+ "learning_rate": 1.3070580013976242e-05,
1745
+ "loss": 0.0048,
1746
+ "step": 2480
1747
+ },
1748
+ {
1749
+ "epoch": 1.740041928721174,
1750
+ "grad_norm": 0.15773366391658783,
1751
+ "learning_rate": 1.3042627533193573e-05,
1752
+ "loss": 0.108,
1753
+ "step": 2490
1754
+ },
1755
+ {
1756
+ "epoch": 1.7470300489168413,
1757
+ "grad_norm": 0.10269400477409363,
1758
+ "learning_rate": 1.3014675052410903e-05,
1759
+ "loss": 0.0884,
1760
+ "step": 2500
1761
+ },
1762
+ {
1763
+ "epoch": 1.7540181691125087,
1764
+ "grad_norm": 0.012210669927299023,
1765
+ "learning_rate": 1.2986722571628234e-05,
1766
+ "loss": 0.0025,
1767
+ "step": 2510
1768
+ },
1769
+ {
1770
+ "epoch": 1.7610062893081762,
1771
+ "grad_norm": 0.046242229640483856,
1772
+ "learning_rate": 1.2958770090845563e-05,
1773
+ "loss": 0.0124,
1774
+ "step": 2520
1775
+ },
1776
+ {
1777
+ "epoch": 1.7679944095038436,
1778
+ "grad_norm": 12.532265663146973,
1779
+ "learning_rate": 1.2930817610062893e-05,
1780
+ "loss": 0.1502,
1781
+ "step": 2530
1782
+ },
1783
+ {
1784
+ "epoch": 1.774982529699511,
1785
+ "grad_norm": 0.018337048590183258,
1786
+ "learning_rate": 1.2902865129280224e-05,
1787
+ "loss": 0.0402,
1788
+ "step": 2540
1789
+ },
1790
+ {
1791
+ "epoch": 1.7819706498951782,
1792
+ "grad_norm": 0.9082455039024353,
1793
+ "learning_rate": 1.2874912648497554e-05,
1794
+ "loss": 0.0719,
1795
+ "step": 2550
1796
+ },
1797
+ {
1798
+ "epoch": 1.7889587700908456,
1799
+ "grad_norm": 0.6127817034721375,
1800
+ "learning_rate": 1.2846960167714887e-05,
1801
+ "loss": 0.011,
1802
+ "step": 2560
1803
+ },
1804
+ {
1805
+ "epoch": 1.7959468902865128,
1806
+ "grad_norm": 0.02741195261478424,
1807
+ "learning_rate": 1.2819007686932217e-05,
1808
+ "loss": 0.0436,
1809
+ "step": 2570
1810
+ },
1811
+ {
1812
+ "epoch": 1.8029350104821802,
1813
+ "grad_norm": 0.012224216014146805,
1814
+ "learning_rate": 1.2791055206149548e-05,
1815
+ "loss": 0.0026,
1816
+ "step": 2580
1817
+ },
1818
+ {
1819
+ "epoch": 1.8099231306778476,
1820
+ "grad_norm": 0.01689975894987583,
1821
+ "learning_rate": 1.2763102725366878e-05,
1822
+ "loss": 0.0067,
1823
+ "step": 2590
1824
+ },
1825
+ {
1826
+ "epoch": 1.816911250873515,
1827
+ "grad_norm": 0.010701341554522514,
1828
+ "learning_rate": 1.2735150244584207e-05,
1829
+ "loss": 0.008,
1830
+ "step": 2600
1831
+ },
1832
+ {
1833
+ "epoch": 1.8238993710691824,
1834
+ "grad_norm": 0.12163686007261276,
1835
+ "learning_rate": 1.2707197763801538e-05,
1836
+ "loss": 0.0767,
1837
+ "step": 2610
1838
+ },
1839
+ {
1840
+ "epoch": 1.8308874912648498,
1841
+ "grad_norm": 0.07165365666151047,
1842
+ "learning_rate": 1.2679245283018868e-05,
1843
+ "loss": 0.0028,
1844
+ "step": 2620
1845
+ },
1846
+ {
1847
+ "epoch": 1.8378756114605173,
1848
+ "grad_norm": 0.007849736139178276,
1849
+ "learning_rate": 1.2651292802236199e-05,
1850
+ "loss": 0.0488,
1851
+ "step": 2630
1852
+ },
1853
+ {
1854
+ "epoch": 1.8448637316561844,
1855
+ "grad_norm": 0.01707429252564907,
1856
+ "learning_rate": 1.262334032145353e-05,
1857
+ "loss": 0.0022,
1858
+ "step": 2640
1859
+ },
1860
+ {
1861
+ "epoch": 1.8518518518518519,
1862
+ "grad_norm": 6.567415237426758,
1863
+ "learning_rate": 1.259538784067086e-05,
1864
+ "loss": 0.0677,
1865
+ "step": 2650
1866
+ },
1867
+ {
1868
+ "epoch": 1.858839972047519,
1869
+ "grad_norm": 0.005324068479239941,
1870
+ "learning_rate": 1.2567435359888192e-05,
1871
+ "loss": 0.0057,
1872
+ "step": 2660
1873
+ },
1874
+ {
1875
+ "epoch": 1.8658280922431865,
1876
+ "grad_norm": 11.791135787963867,
1877
+ "learning_rate": 1.2539482879105523e-05,
1878
+ "loss": 0.0764,
1879
+ "step": 2670
1880
+ },
1881
+ {
1882
+ "epoch": 1.8728162124388539,
1883
+ "grad_norm": 0.005892631132155657,
1884
+ "learning_rate": 1.2511530398322853e-05,
1885
+ "loss": 0.0736,
1886
+ "step": 2680
1887
+ },
1888
+ {
1889
+ "epoch": 1.8798043326345213,
1890
+ "grad_norm": 0.04005246236920357,
1891
+ "learning_rate": 1.2483577917540182e-05,
1892
+ "loss": 0.0617,
1893
+ "step": 2690
1894
+ },
1895
+ {
1896
+ "epoch": 1.8867924528301887,
1897
+ "grad_norm": 0.010624396614730358,
1898
+ "learning_rate": 1.2455625436757513e-05,
1899
+ "loss": 0.0011,
1900
+ "step": 2700
1901
+ },
1902
+ {
1903
+ "epoch": 1.8937805730258561,
1904
+ "grad_norm": 0.01761520840227604,
1905
+ "learning_rate": 1.2427672955974843e-05,
1906
+ "loss": 0.1574,
1907
+ "step": 2710
1908
+ },
1909
+ {
1910
+ "epoch": 1.9007686932215235,
1911
+ "grad_norm": 0.05212152749300003,
1912
+ "learning_rate": 1.2399720475192174e-05,
1913
+ "loss": 0.0618,
1914
+ "step": 2720
1915
+ },
1916
+ {
1917
+ "epoch": 1.9077568134171907,
1918
+ "grad_norm": 0.012297810055315495,
1919
+ "learning_rate": 1.2371767994409504e-05,
1920
+ "loss": 0.0802,
1921
+ "step": 2730
1922
+ },
1923
+ {
1924
+ "epoch": 1.9147449336128581,
1925
+ "grad_norm": 1.4671796560287476,
1926
+ "learning_rate": 1.2343815513626835e-05,
1927
+ "loss": 0.0351,
1928
+ "step": 2740
1929
+ },
1930
+ {
1931
+ "epoch": 1.9217330538085255,
1932
+ "grad_norm": 1.28997802734375,
1933
+ "learning_rate": 1.2315863032844166e-05,
1934
+ "loss": 0.0069,
1935
+ "step": 2750
1936
+ },
1937
+ {
1938
+ "epoch": 1.9287211740041927,
1939
+ "grad_norm": 0.15810908377170563,
1940
+ "learning_rate": 1.2287910552061498e-05,
1941
+ "loss": 0.0037,
1942
+ "step": 2760
1943
+ },
1944
+ {
1945
+ "epoch": 1.9357092941998602,
1946
+ "grad_norm": 0.01939568482339382,
1947
+ "learning_rate": 1.2259958071278828e-05,
1948
+ "loss": 0.1375,
1949
+ "step": 2770
1950
+ },
1951
+ {
1952
+ "epoch": 1.9426974143955276,
1953
+ "grad_norm": 0.023233506828546524,
1954
+ "learning_rate": 1.2232005590496157e-05,
1955
+ "loss": 0.0016,
1956
+ "step": 2780
1957
+ },
1958
+ {
1959
+ "epoch": 1.949685534591195,
1960
+ "grad_norm": 13.808353424072266,
1961
+ "learning_rate": 1.2204053109713488e-05,
1962
+ "loss": 0.1251,
1963
+ "step": 2790
1964
+ },
1965
+ {
1966
+ "epoch": 1.9566736547868624,
1967
+ "grad_norm": 0.007430393248796463,
1968
+ "learning_rate": 1.2176100628930818e-05,
1969
+ "loss": 0.0957,
1970
+ "step": 2800
1971
+ },
1972
+ {
1973
+ "epoch": 1.9636617749825298,
1974
+ "grad_norm": 0.015328769572079182,
1975
+ "learning_rate": 1.2148148148148149e-05,
1976
+ "loss": 0.0019,
1977
+ "step": 2810
1978
+ },
1979
+ {
1980
+ "epoch": 1.9706498951781972,
1981
+ "grad_norm": 0.05328751727938652,
1982
+ "learning_rate": 1.212019566736548e-05,
1983
+ "loss": 0.2193,
1984
+ "step": 2820
1985
+ },
1986
+ {
1987
+ "epoch": 1.9776380153738644,
1988
+ "grad_norm": 0.05544736981391907,
1989
+ "learning_rate": 1.209224318658281e-05,
1990
+ "loss": 0.011,
1991
+ "step": 2830
1992
+ },
1993
+ {
1994
+ "epoch": 1.9846261355695318,
1995
+ "grad_norm": 0.006596289575099945,
1996
+ "learning_rate": 1.206429070580014e-05,
1997
+ "loss": 0.0454,
1998
+ "step": 2840
1999
+ },
2000
+ {
2001
+ "epoch": 1.991614255765199,
2002
+ "grad_norm": 7.7824015617370605,
2003
+ "learning_rate": 1.2036338225017471e-05,
2004
+ "loss": 0.024,
2005
+ "step": 2850
2006
+ },
2007
+ {
2008
+ "epoch": 1.9986023759608664,
2009
+ "grad_norm": 0.024772530421614647,
2010
+ "learning_rate": 1.2008385744234804e-05,
2011
+ "loss": 0.0109,
2012
+ "step": 2860
2013
+ },
2014
+ {
2015
+ "epoch": 2.005590496156534,
2016
+ "grad_norm": 0.026197150349617004,
2017
+ "learning_rate": 1.198043326345213e-05,
2018
+ "loss": 0.0362,
2019
+ "step": 2870
2020
+ },
2021
+ {
2022
+ "epoch": 2.0125786163522013,
2023
+ "grad_norm": 18.572996139526367,
2024
+ "learning_rate": 1.1952480782669463e-05,
2025
+ "loss": 0.0935,
2026
+ "step": 2880
2027
+ },
2028
+ {
2029
+ "epoch": 2.0195667365478687,
2030
+ "grad_norm": 0.17002998292446136,
2031
+ "learning_rate": 1.1924528301886794e-05,
2032
+ "loss": 0.0343,
2033
+ "step": 2890
2034
+ },
2035
+ {
2036
+ "epoch": 2.026554856743536,
2037
+ "grad_norm": 0.029017208144068718,
2038
+ "learning_rate": 1.1896575821104124e-05,
2039
+ "loss": 0.0013,
2040
+ "step": 2900
2041
+ },
2042
+ {
2043
+ "epoch": 2.0335429769392035,
2044
+ "grad_norm": 0.03208720684051514,
2045
+ "learning_rate": 1.1868623340321455e-05,
2046
+ "loss": 0.0595,
2047
+ "step": 2910
2048
+ },
2049
+ {
2050
+ "epoch": 2.040531097134871,
2051
+ "grad_norm": 0.03342962637543678,
2052
+ "learning_rate": 1.1840670859538785e-05,
2053
+ "loss": 0.0025,
2054
+ "step": 2920
2055
+ },
2056
+ {
2057
+ "epoch": 2.047519217330538,
2058
+ "grad_norm": 0.00888112373650074,
2059
+ "learning_rate": 1.1812718378756116e-05,
2060
+ "loss": 0.0007,
2061
+ "step": 2930
2062
+ },
2063
+ {
2064
+ "epoch": 2.0545073375262053,
2065
+ "grad_norm": 0.14470069110393524,
2066
+ "learning_rate": 1.1784765897973446e-05,
2067
+ "loss": 0.1185,
2068
+ "step": 2940
2069
+ },
2070
+ {
2071
+ "epoch": 2.0614954577218727,
2072
+ "grad_norm": 0.006698206998407841,
2073
+ "learning_rate": 1.1756813417190777e-05,
2074
+ "loss": 0.0027,
2075
+ "step": 2950
2076
+ },
2077
+ {
2078
+ "epoch": 2.06848357791754,
2079
+ "grad_norm": 0.013084967620670795,
2080
+ "learning_rate": 1.1728860936408106e-05,
2081
+ "loss": 0.1188,
2082
+ "step": 2960
2083
+ },
2084
+ {
2085
+ "epoch": 2.0754716981132075,
2086
+ "grad_norm": 4.599226951599121,
2087
+ "learning_rate": 1.1700908455625436e-05,
2088
+ "loss": 0.192,
2089
+ "step": 2970
2090
+ },
2091
+ {
2092
+ "epoch": 2.082459818308875,
2093
+ "grad_norm": 0.04704582691192627,
2094
+ "learning_rate": 1.1672955974842769e-05,
2095
+ "loss": 0.0214,
2096
+ "step": 2980
2097
+ },
2098
+ {
2099
+ "epoch": 2.0894479385045424,
2100
+ "grad_norm": 0.00998607650399208,
2101
+ "learning_rate": 1.16450034940601e-05,
2102
+ "loss": 0.0642,
2103
+ "step": 2990
2104
+ },
2105
+ {
2106
+ "epoch": 2.0964360587002098,
2107
+ "grad_norm": 11.119775772094727,
2108
+ "learning_rate": 1.161705101327743e-05,
2109
+ "loss": 0.0075,
2110
+ "step": 3000
2111
+ },
2112
+ {
2113
+ "epoch": 2.103424178895877,
2114
+ "grad_norm": 10.634467124938965,
2115
+ "learning_rate": 1.158909853249476e-05,
2116
+ "loss": 0.1313,
2117
+ "step": 3010
2118
+ },
2119
+ {
2120
+ "epoch": 2.1104122990915446,
2121
+ "grad_norm": 0.05366509407758713,
2122
+ "learning_rate": 1.1561146051712091e-05,
2123
+ "loss": 0.0151,
2124
+ "step": 3020
2125
+ },
2126
+ {
2127
+ "epoch": 2.1174004192872116,
2128
+ "grad_norm": 0.014267387799918652,
2129
+ "learning_rate": 1.1533193570929421e-05,
2130
+ "loss": 0.1542,
2131
+ "step": 3030
2132
+ },
2133
+ {
2134
+ "epoch": 2.124388539482879,
2135
+ "grad_norm": 0.008031471632421017,
2136
+ "learning_rate": 1.150524109014675e-05,
2137
+ "loss": 0.0015,
2138
+ "step": 3040
2139
+ },
2140
+ {
2141
+ "epoch": 2.1313766596785464,
2142
+ "grad_norm": 8.725931167602539,
2143
+ "learning_rate": 1.1477288609364081e-05,
2144
+ "loss": 0.0981,
2145
+ "step": 3050
2146
+ },
2147
+ {
2148
+ "epoch": 2.138364779874214,
2149
+ "grad_norm": 0.010948359034955502,
2150
+ "learning_rate": 1.1449336128581411e-05,
2151
+ "loss": 0.0513,
2152
+ "step": 3060
2153
+ },
2154
+ {
2155
+ "epoch": 2.145352900069881,
2156
+ "grad_norm": 0.22167250514030457,
2157
+ "learning_rate": 1.1421383647798742e-05,
2158
+ "loss": 0.0295,
2159
+ "step": 3070
2160
+ },
2161
+ {
2162
+ "epoch": 2.1523410202655486,
2163
+ "grad_norm": 0.038010116666555405,
2164
+ "learning_rate": 1.1393431167016074e-05,
2165
+ "loss": 0.001,
2166
+ "step": 3080
2167
+ },
2168
+ {
2169
+ "epoch": 2.159329140461216,
2170
+ "grad_norm": 0.014428159222006798,
2171
+ "learning_rate": 1.1365478686233405e-05,
2172
+ "loss": 0.0015,
2173
+ "step": 3090
2174
+ },
2175
+ {
2176
+ "epoch": 2.1663172606568835,
2177
+ "grad_norm": 0.027289502322673798,
2178
+ "learning_rate": 1.1337526205450735e-05,
2179
+ "loss": 0.0036,
2180
+ "step": 3100
2181
+ },
2182
+ {
2183
+ "epoch": 2.1733053808525504,
2184
+ "grad_norm": 1.7627710103988647,
2185
+ "learning_rate": 1.1309573724668066e-05,
2186
+ "loss": 0.0314,
2187
+ "step": 3110
2188
+ },
2189
+ {
2190
+ "epoch": 2.180293501048218,
2191
+ "grad_norm": 0.006555517669767141,
2192
+ "learning_rate": 1.1281621243885397e-05,
2193
+ "loss": 0.0017,
2194
+ "step": 3120
2195
+ },
2196
+ {
2197
+ "epoch": 2.1872816212438853,
2198
+ "grad_norm": 0.006436652038246393,
2199
+ "learning_rate": 1.1253668763102725e-05,
2200
+ "loss": 0.006,
2201
+ "step": 3130
2202
+ },
2203
+ {
2204
+ "epoch": 2.1942697414395527,
2205
+ "grad_norm": 0.010117010213434696,
2206
+ "learning_rate": 1.1225716282320056e-05,
2207
+ "loss": 0.0127,
2208
+ "step": 3140
2209
+ },
2210
+ {
2211
+ "epoch": 2.20125786163522,
2212
+ "grad_norm": 0.01584050804376602,
2213
+ "learning_rate": 1.1197763801537387e-05,
2214
+ "loss": 0.0011,
2215
+ "step": 3150
2216
+ },
2217
+ {
2218
+ "epoch": 2.2082459818308875,
2219
+ "grad_norm": 3.1698734760284424,
2220
+ "learning_rate": 1.1169811320754717e-05,
2221
+ "loss": 0.0643,
2222
+ "step": 3160
2223
+ },
2224
+ {
2225
+ "epoch": 2.215234102026555,
2226
+ "grad_norm": 20.464344024658203,
2227
+ "learning_rate": 1.1141858839972048e-05,
2228
+ "loss": 0.1347,
2229
+ "step": 3170
2230
+ },
2231
+ {
2232
+ "epoch": 2.2222222222222223,
2233
+ "grad_norm": 0.04433240368962288,
2234
+ "learning_rate": 1.111390635918938e-05,
2235
+ "loss": 0.0079,
2236
+ "step": 3180
2237
+ },
2238
+ {
2239
+ "epoch": 2.2292103424178897,
2240
+ "grad_norm": 0.04871518909931183,
2241
+ "learning_rate": 1.108595387840671e-05,
2242
+ "loss": 0.1057,
2243
+ "step": 3190
2244
+ },
2245
+ {
2246
+ "epoch": 2.236198462613557,
2247
+ "grad_norm": 21.398836135864258,
2248
+ "learning_rate": 1.1058001397624041e-05,
2249
+ "loss": 0.0091,
2250
+ "step": 3200
2251
+ },
2252
+ {
2253
+ "epoch": 2.243186582809224,
2254
+ "grad_norm": 0.023466596379876137,
2255
+ "learning_rate": 1.1030048916841372e-05,
2256
+ "loss": 0.0057,
2257
+ "step": 3210
2258
+ },
2259
+ {
2260
+ "epoch": 2.2501747030048915,
2261
+ "grad_norm": 0.011096821166574955,
2262
+ "learning_rate": 1.10020964360587e-05,
2263
+ "loss": 0.0048,
2264
+ "step": 3220
2265
+ },
2266
+ {
2267
+ "epoch": 2.257162823200559,
2268
+ "grad_norm": 7.414328098297119,
2269
+ "learning_rate": 1.0974143955276031e-05,
2270
+ "loss": 0.0135,
2271
+ "step": 3230
2272
+ },
2273
+ {
2274
+ "epoch": 2.2641509433962264,
2275
+ "grad_norm": 0.03595978021621704,
2276
+ "learning_rate": 1.0946191474493362e-05,
2277
+ "loss": 0.001,
2278
+ "step": 3240
2279
+ },
2280
+ {
2281
+ "epoch": 2.2711390635918938,
2282
+ "grad_norm": 17.89633560180664,
2283
+ "learning_rate": 1.0918238993710692e-05,
2284
+ "loss": 0.0867,
2285
+ "step": 3250
2286
+ },
2287
+ {
2288
+ "epoch": 2.278127183787561,
2289
+ "grad_norm": 0.009794089011847973,
2290
+ "learning_rate": 1.0890286512928023e-05,
2291
+ "loss": 0.0324,
2292
+ "step": 3260
2293
+ },
2294
+ {
2295
+ "epoch": 2.2851153039832286,
2296
+ "grad_norm": 0.04615940898656845,
2297
+ "learning_rate": 1.0862334032145353e-05,
2298
+ "loss": 0.0013,
2299
+ "step": 3270
2300
+ },
2301
+ {
2302
+ "epoch": 2.292103424178896,
2303
+ "grad_norm": 0.14327336847782135,
2304
+ "learning_rate": 1.0834381551362686e-05,
2305
+ "loss": 0.0009,
2306
+ "step": 3280
2307
+ },
2308
+ {
2309
+ "epoch": 2.2990915443745634,
2310
+ "grad_norm": 0.003527725813910365,
2311
+ "learning_rate": 1.0806429070580016e-05,
2312
+ "loss": 0.0751,
2313
+ "step": 3290
2314
+ },
2315
+ {
2316
+ "epoch": 2.3060796645702304,
2317
+ "grad_norm": 0.013885865919291973,
2318
+ "learning_rate": 1.0778476589797347e-05,
2319
+ "loss": 0.0537,
2320
+ "step": 3300
2321
+ },
2322
+ {
2323
+ "epoch": 2.313067784765898,
2324
+ "grad_norm": 0.01316741295158863,
2325
+ "learning_rate": 1.0750524109014676e-05,
2326
+ "loss": 0.0008,
2327
+ "step": 3310
2328
+ },
2329
+ {
2330
+ "epoch": 2.320055904961565,
2331
+ "grad_norm": 0.0060685062780976295,
2332
+ "learning_rate": 1.0722571628232006e-05,
2333
+ "loss": 0.0673,
2334
+ "step": 3320
2335
+ },
2336
+ {
2337
+ "epoch": 2.3270440251572326,
2338
+ "grad_norm": 0.021814968436956406,
2339
+ "learning_rate": 1.0694619147449337e-05,
2340
+ "loss": 0.0014,
2341
+ "step": 3330
2342
+ },
2343
+ {
2344
+ "epoch": 2.3340321453529,
2345
+ "grad_norm": 0.013100036419928074,
2346
+ "learning_rate": 1.0666666666666667e-05,
2347
+ "loss": 0.0756,
2348
+ "step": 3340
2349
+ },
2350
+ {
2351
+ "epoch": 2.3410202655485675,
2352
+ "grad_norm": 0.01045132428407669,
2353
+ "learning_rate": 1.0638714185883998e-05,
2354
+ "loss": 0.0459,
2355
+ "step": 3350
2356
+ },
2357
+ {
2358
+ "epoch": 2.348008385744235,
2359
+ "grad_norm": 0.006095814984291792,
2360
+ "learning_rate": 1.0610761705101328e-05,
2361
+ "loss": 0.0551,
2362
+ "step": 3360
2363
+ },
2364
+ {
2365
+ "epoch": 2.3549965059399023,
2366
+ "grad_norm": 0.011639703065156937,
2367
+ "learning_rate": 1.0582809224318659e-05,
2368
+ "loss": 0.0032,
2369
+ "step": 3370
2370
+ },
2371
+ {
2372
+ "epoch": 2.3619846261355697,
2373
+ "grad_norm": 0.0661187469959259,
2374
+ "learning_rate": 1.0554856743535991e-05,
2375
+ "loss": 0.0009,
2376
+ "step": 3380
2377
+ },
2378
+ {
2379
+ "epoch": 2.368972746331237,
2380
+ "grad_norm": 0.030025649815797806,
2381
+ "learning_rate": 1.0526904262753318e-05,
2382
+ "loss": 0.0096,
2383
+ "step": 3390
2384
+ },
2385
+ {
2386
+ "epoch": 2.375960866526904,
2387
+ "grad_norm": 0.05209723860025406,
2388
+ "learning_rate": 1.049895178197065e-05,
2389
+ "loss": 0.1485,
2390
+ "step": 3400
2391
+ },
2392
+ {
2393
+ "epoch": 2.3829489867225715,
2394
+ "grad_norm": 0.4318474233150482,
2395
+ "learning_rate": 1.0470999301187981e-05,
2396
+ "loss": 0.0011,
2397
+ "step": 3410
2398
+ },
2399
+ {
2400
+ "epoch": 2.389937106918239,
2401
+ "grad_norm": 0.011814878322184086,
2402
+ "learning_rate": 1.0443046820405312e-05,
2403
+ "loss": 0.0019,
2404
+ "step": 3420
2405
+ },
2406
+ {
2407
+ "epoch": 2.3969252271139063,
2408
+ "grad_norm": 27.78424644470215,
2409
+ "learning_rate": 1.0415094339622642e-05,
2410
+ "loss": 0.1376,
2411
+ "step": 3430
2412
+ },
2413
+ {
2414
+ "epoch": 2.4039133473095737,
2415
+ "grad_norm": 0.27662497758865356,
2416
+ "learning_rate": 1.0387141858839973e-05,
2417
+ "loss": 0.0009,
2418
+ "step": 3440
2419
+ },
2420
+ {
2421
+ "epoch": 2.410901467505241,
2422
+ "grad_norm": 0.0203730296343565,
2423
+ "learning_rate": 1.0359189378057304e-05,
2424
+ "loss": 0.001,
2425
+ "step": 3450
2426
+ },
2427
+ {
2428
+ "epoch": 2.4178895877009086,
2429
+ "grad_norm": 0.005023602396249771,
2430
+ "learning_rate": 1.0331236897274634e-05,
2431
+ "loss": 0.0026,
2432
+ "step": 3460
2433
+ },
2434
+ {
2435
+ "epoch": 2.424877707896576,
2436
+ "grad_norm": 0.005766034591943026,
2437
+ "learning_rate": 1.0303284416491965e-05,
2438
+ "loss": 0.0043,
2439
+ "step": 3470
2440
+ },
2441
+ {
2442
+ "epoch": 2.431865828092243,
2443
+ "grad_norm": 0.0078241853043437,
2444
+ "learning_rate": 1.0275331935709294e-05,
2445
+ "loss": 0.0666,
2446
+ "step": 3480
2447
+ },
2448
+ {
2449
+ "epoch": 2.4388539482879104,
2450
+ "grad_norm": 0.13950783014297485,
2451
+ "learning_rate": 1.0247379454926624e-05,
2452
+ "loss": 0.0299,
2453
+ "step": 3490
2454
+ },
2455
+ {
2456
+ "epoch": 2.4458420684835778,
2457
+ "grad_norm": 0.03160367161035538,
2458
+ "learning_rate": 1.0219426974143956e-05,
2459
+ "loss": 0.003,
2460
+ "step": 3500
2461
+ },
2462
+ {
2463
+ "epoch": 2.452830188679245,
2464
+ "grad_norm": 0.01961313560605049,
2465
+ "learning_rate": 1.0191474493361287e-05,
2466
+ "loss": 0.0103,
2467
+ "step": 3510
2468
+ },
2469
+ {
2470
+ "epoch": 2.4598183088749126,
2471
+ "grad_norm": 0.007672517094761133,
2472
+ "learning_rate": 1.0163522012578618e-05,
2473
+ "loss": 0.0037,
2474
+ "step": 3520
2475
+ },
2476
+ {
2477
+ "epoch": 2.46680642907058,
2478
+ "grad_norm": 0.005141376983374357,
2479
+ "learning_rate": 1.0135569531795948e-05,
2480
+ "loss": 0.0005,
2481
+ "step": 3530
2482
+ },
2483
+ {
2484
+ "epoch": 2.4737945492662474,
2485
+ "grad_norm": 0.027929285541176796,
2486
+ "learning_rate": 1.0107617051013279e-05,
2487
+ "loss": 0.0248,
2488
+ "step": 3540
2489
+ },
2490
+ {
2491
+ "epoch": 2.480782669461915,
2492
+ "grad_norm": 1.3310028314590454,
2493
+ "learning_rate": 1.007966457023061e-05,
2494
+ "loss": 0.124,
2495
+ "step": 3550
2496
+ },
2497
+ {
2498
+ "epoch": 2.4877707896575822,
2499
+ "grad_norm": 0.06481662392616272,
2500
+ "learning_rate": 1.005171208944794e-05,
2501
+ "loss": 0.0024,
2502
+ "step": 3560
2503
+ },
2504
+ {
2505
+ "epoch": 2.4947589098532497,
2506
+ "grad_norm": 0.008162762969732285,
2507
+ "learning_rate": 1.0023759608665269e-05,
2508
+ "loss": 0.0017,
2509
+ "step": 3570
2510
+ },
2511
+ {
2512
+ "epoch": 2.501747030048917,
2513
+ "grad_norm": 7.582877159118652,
2514
+ "learning_rate": 9.995807127882601e-06,
2515
+ "loss": 0.0838,
2516
+ "step": 3580
2517
+ },
2518
+ {
2519
+ "epoch": 2.508735150244584,
2520
+ "grad_norm": 0.06521458923816681,
2521
+ "learning_rate": 9.96785464709993e-06,
2522
+ "loss": 0.0009,
2523
+ "step": 3590
2524
+ },
2525
+ {
2526
+ "epoch": 2.5157232704402515,
2527
+ "grad_norm": 0.005363748874515295,
2528
+ "learning_rate": 9.939902166317262e-06,
2529
+ "loss": 0.0797,
2530
+ "step": 3600
2531
+ },
2532
+ {
2533
+ "epoch": 2.522711390635919,
2534
+ "grad_norm": 0.007538496516644955,
2535
+ "learning_rate": 9.911949685534593e-06,
2536
+ "loss": 0.039,
2537
+ "step": 3610
2538
+ },
2539
+ {
2540
+ "epoch": 2.5296995108315863,
2541
+ "grad_norm": 0.007935012690722942,
2542
+ "learning_rate": 9.883997204751923e-06,
2543
+ "loss": 0.0039,
2544
+ "step": 3620
2545
+ },
2546
+ {
2547
+ "epoch": 2.5366876310272537,
2548
+ "grad_norm": 0.005138902924954891,
2549
+ "learning_rate": 9.856044723969254e-06,
2550
+ "loss": 0.0006,
2551
+ "step": 3630
2552
+ },
2553
+ {
2554
+ "epoch": 2.543675751222921,
2555
+ "grad_norm": 0.012060786597430706,
2556
+ "learning_rate": 9.828092243186583e-06,
2557
+ "loss": 0.0013,
2558
+ "step": 3640
2559
+ },
2560
+ {
2561
+ "epoch": 2.5506638714185885,
2562
+ "grad_norm": 0.09823895990848541,
2563
+ "learning_rate": 9.800139762403915e-06,
2564
+ "loss": 0.0319,
2565
+ "step": 3650
2566
+ },
2567
+ {
2568
+ "epoch": 2.5576519916142555,
2569
+ "grad_norm": 0.009398439899086952,
2570
+ "learning_rate": 9.772187281621245e-06,
2571
+ "loss": 0.0012,
2572
+ "step": 3660
2573
+ },
2574
+ {
2575
+ "epoch": 2.564640111809923,
2576
+ "grad_norm": 0.009316815994679928,
2577
+ "learning_rate": 9.744234800838576e-06,
2578
+ "loss": 0.0008,
2579
+ "step": 3670
2580
+ },
2581
+ {
2582
+ "epoch": 2.5716282320055903,
2583
+ "grad_norm": 0.01164779718965292,
2584
+ "learning_rate": 9.716282320055905e-06,
2585
+ "loss": 0.0014,
2586
+ "step": 3680
2587
+ },
2588
+ {
2589
+ "epoch": 2.5786163522012577,
2590
+ "grad_norm": 7.828653335571289,
2591
+ "learning_rate": 9.688329839273235e-06,
2592
+ "loss": 0.0452,
2593
+ "step": 3690
2594
+ },
2595
+ {
2596
+ "epoch": 2.585604472396925,
2597
+ "grad_norm": 0.003899825969710946,
2598
+ "learning_rate": 9.660377358490568e-06,
2599
+ "loss": 0.0004,
2600
+ "step": 3700
2601
+ },
2602
+ {
2603
+ "epoch": 2.5925925925925926,
2604
+ "grad_norm": 6.803162097930908,
2605
+ "learning_rate": 9.632424877707898e-06,
2606
+ "loss": 0.0892,
2607
+ "step": 3710
2608
+ },
2609
+ {
2610
+ "epoch": 2.59958071278826,
2611
+ "grad_norm": 0.00998039823025465,
2612
+ "learning_rate": 9.604472396925227e-06,
2613
+ "loss": 0.123,
2614
+ "step": 3720
2615
+ },
2616
+ {
2617
+ "epoch": 2.6065688329839274,
2618
+ "grad_norm": 0.0067862533032894135,
2619
+ "learning_rate": 9.576519916142558e-06,
2620
+ "loss": 0.001,
2621
+ "step": 3730
2622
+ },
2623
+ {
2624
+ "epoch": 2.613556953179595,
2625
+ "grad_norm": 0.004215400665998459,
2626
+ "learning_rate": 9.548567435359888e-06,
2627
+ "loss": 0.177,
2628
+ "step": 3740
2629
+ },
2630
+ {
2631
+ "epoch": 2.620545073375262,
2632
+ "grad_norm": 0.02502995729446411,
2633
+ "learning_rate": 9.52061495457722e-06,
2634
+ "loss": 0.008,
2635
+ "step": 3750
2636
+ },
2637
+ {
2638
+ "epoch": 2.6275331935709296,
2639
+ "grad_norm": 5.30055046081543,
2640
+ "learning_rate": 9.492662473794551e-06,
2641
+ "loss": 0.1303,
2642
+ "step": 3760
2643
+ },
2644
+ {
2645
+ "epoch": 2.634521313766597,
2646
+ "grad_norm": 0.01970675028860569,
2647
+ "learning_rate": 9.46470999301188e-06,
2648
+ "loss": 0.0012,
2649
+ "step": 3770
2650
+ },
2651
+ {
2652
+ "epoch": 2.641509433962264,
2653
+ "grad_norm": 0.004884845577180386,
2654
+ "learning_rate": 9.43675751222921e-06,
2655
+ "loss": 0.0307,
2656
+ "step": 3780
2657
+ },
2658
+ {
2659
+ "epoch": 2.6484975541579314,
2660
+ "grad_norm": 0.08335088938474655,
2661
+ "learning_rate": 9.408805031446541e-06,
2662
+ "loss": 0.002,
2663
+ "step": 3790
2664
+ },
2665
+ {
2666
+ "epoch": 2.655485674353599,
2667
+ "grad_norm": 0.012866518460214138,
2668
+ "learning_rate": 9.380852550663873e-06,
2669
+ "loss": 0.0006,
2670
+ "step": 3800
2671
+ },
2672
+ {
2673
+ "epoch": 2.6624737945492662,
2674
+ "grad_norm": 0.024261610582470894,
2675
+ "learning_rate": 9.352900069881202e-06,
2676
+ "loss": 0.0012,
2677
+ "step": 3810
2678
+ },
2679
+ {
2680
+ "epoch": 2.6694619147449337,
2681
+ "grad_norm": 0.00420184712857008,
2682
+ "learning_rate": 9.324947589098533e-06,
2683
+ "loss": 0.0076,
2684
+ "step": 3820
2685
+ },
2686
+ {
2687
+ "epoch": 2.676450034940601,
2688
+ "grad_norm": 0.13258372247219086,
2689
+ "learning_rate": 9.296995108315863e-06,
2690
+ "loss": 0.0254,
2691
+ "step": 3830
2692
+ },
2693
+ {
2694
+ "epoch": 2.6834381551362685,
2695
+ "grad_norm": 0.05237485095858574,
2696
+ "learning_rate": 9.269042627533194e-06,
2697
+ "loss": 0.2556,
2698
+ "step": 3840
2699
+ },
2700
+ {
2701
+ "epoch": 2.6904262753319355,
2702
+ "grad_norm": 0.007042170502245426,
2703
+ "learning_rate": 9.241090146750526e-06,
2704
+ "loss": 0.0451,
2705
+ "step": 3850
2706
+ },
2707
+ {
2708
+ "epoch": 2.697414395527603,
2709
+ "grad_norm": 0.0603499561548233,
2710
+ "learning_rate": 9.213137665967855e-06,
2711
+ "loss": 0.018,
2712
+ "step": 3860
2713
+ },
2714
+ {
2715
+ "epoch": 2.7044025157232703,
2716
+ "grad_norm": 0.005806710571050644,
2717
+ "learning_rate": 9.185185185185186e-06,
2718
+ "loss": 0.0268,
2719
+ "step": 3870
2720
+ },
2721
+ {
2722
+ "epoch": 2.7113906359189377,
2723
+ "grad_norm": 0.019446423277258873,
2724
+ "learning_rate": 9.157232704402516e-06,
2725
+ "loss": 0.0013,
2726
+ "step": 3880
2727
+ },
2728
+ {
2729
+ "epoch": 2.718378756114605,
2730
+ "grad_norm": 0.046024855226278305,
2731
+ "learning_rate": 9.129280223619847e-06,
2732
+ "loss": 0.0094,
2733
+ "step": 3890
2734
+ },
2735
+ {
2736
+ "epoch": 2.7253668763102725,
2737
+ "grad_norm": 0.2116568386554718,
2738
+ "learning_rate": 9.101327742837177e-06,
2739
+ "loss": 0.0079,
2740
+ "step": 3900
2741
+ },
2742
+ {
2743
+ "epoch": 2.73235499650594,
2744
+ "grad_norm": 0.014955777674913406,
2745
+ "learning_rate": 9.073375262054508e-06,
2746
+ "loss": 0.0007,
2747
+ "step": 3910
2748
+ },
2749
+ {
2750
+ "epoch": 2.7393431167016074,
2751
+ "grad_norm": 0.7381608486175537,
2752
+ "learning_rate": 9.045422781271838e-06,
2753
+ "loss": 0.0017,
2754
+ "step": 3920
2755
+ },
2756
+ {
2757
+ "epoch": 2.7463312368972748,
2758
+ "grad_norm": 0.0042509473860263824,
2759
+ "learning_rate": 9.017470300489169e-06,
2760
+ "loss": 0.0328,
2761
+ "step": 3930
2762
+ },
2763
+ {
2764
+ "epoch": 2.753319357092942,
2765
+ "grad_norm": 0.008610788732767105,
2766
+ "learning_rate": 8.9895178197065e-06,
2767
+ "loss": 0.0033,
2768
+ "step": 3940
2769
+ },
2770
+ {
2771
+ "epoch": 2.7603074772886096,
2772
+ "grad_norm": 0.032417889684438705,
2773
+ "learning_rate": 8.96156533892383e-06,
2774
+ "loss": 0.0007,
2775
+ "step": 3950
2776
+ },
2777
+ {
2778
+ "epoch": 2.767295597484277,
2779
+ "grad_norm": 0.003039875766262412,
2780
+ "learning_rate": 8.93361285814116e-06,
2781
+ "loss": 0.0006,
2782
+ "step": 3960
2783
+ },
2784
+ {
2785
+ "epoch": 2.774283717679944,
2786
+ "grad_norm": 0.1048172265291214,
2787
+ "learning_rate": 8.905660377358491e-06,
2788
+ "loss": 0.0213,
2789
+ "step": 3970
2790
+ },
2791
+ {
2792
+ "epoch": 2.7812718378756114,
2793
+ "grad_norm": 0.0031095799058675766,
2794
+ "learning_rate": 8.877707896575822e-06,
2795
+ "loss": 0.0358,
2796
+ "step": 3980
2797
+ },
2798
+ {
2799
+ "epoch": 2.788259958071279,
2800
+ "grad_norm": 0.08362710475921631,
2801
+ "learning_rate": 8.849755415793152e-06,
2802
+ "loss": 0.0433,
2803
+ "step": 3990
2804
+ },
2805
+ {
2806
+ "epoch": 2.795248078266946,
2807
+ "grad_norm": 0.006064527668058872,
2808
+ "learning_rate": 8.821802935010483e-06,
2809
+ "loss": 0.0009,
2810
+ "step": 4000
2811
+ },
2812
+ {
2813
+ "epoch": 2.8022361984626136,
2814
+ "grad_norm": 0.009575744159519672,
2815
+ "learning_rate": 8.793850454227814e-06,
2816
+ "loss": 0.0003,
2817
+ "step": 4010
2818
+ },
2819
+ {
2820
+ "epoch": 2.809224318658281,
2821
+ "grad_norm": 0.017887070775032043,
2822
+ "learning_rate": 8.765897973445144e-06,
2823
+ "loss": 0.0152,
2824
+ "step": 4020
2825
+ },
2826
+ {
2827
+ "epoch": 2.8162124388539485,
2828
+ "grad_norm": 0.0015486754709854722,
2829
+ "learning_rate": 8.737945492662475e-06,
2830
+ "loss": 0.0692,
2831
+ "step": 4030
2832
+ },
2833
+ {
2834
+ "epoch": 2.8232005590496154,
2835
+ "grad_norm": 0.00923781655728817,
2836
+ "learning_rate": 8.709993011879805e-06,
2837
+ "loss": 0.0006,
2838
+ "step": 4040
2839
+ },
2840
+ {
2841
+ "epoch": 2.830188679245283,
2842
+ "grad_norm": 14.54555892944336,
2843
+ "learning_rate": 8.682040531097136e-06,
2844
+ "loss": 0.0804,
2845
+ "step": 4050
2846
+ },
2847
+ {
2848
+ "epoch": 2.8371767994409502,
2849
+ "grad_norm": 0.035101618617773056,
2850
+ "learning_rate": 8.654088050314466e-06,
2851
+ "loss": 0.0077,
2852
+ "step": 4060
2853
+ },
2854
+ {
2855
+ "epoch": 2.8441649196366177,
2856
+ "grad_norm": 0.22155028581619263,
2857
+ "learning_rate": 8.626135569531797e-06,
2858
+ "loss": 0.0257,
2859
+ "step": 4070
2860
+ },
2861
+ {
2862
+ "epoch": 2.851153039832285,
2863
+ "grad_norm": 0.013333957642316818,
2864
+ "learning_rate": 8.598183088749128e-06,
2865
+ "loss": 0.0033,
2866
+ "step": 4080
2867
+ },
2868
+ {
2869
+ "epoch": 2.8581411600279525,
2870
+ "grad_norm": 0.029933998361229897,
2871
+ "learning_rate": 8.570230607966458e-06,
2872
+ "loss": 0.0018,
2873
+ "step": 4090
2874
+ },
2875
+ {
2876
+ "epoch": 2.86512928022362,
2877
+ "grad_norm": 0.021358752623200417,
2878
+ "learning_rate": 8.542278127183789e-06,
2879
+ "loss": 0.0004,
2880
+ "step": 4100
2881
+ },
2882
+ {
2883
+ "epoch": 2.8721174004192873,
2884
+ "grad_norm": 0.0044855596497654915,
2885
+ "learning_rate": 8.51432564640112e-06,
2886
+ "loss": 0.0007,
2887
+ "step": 4110
2888
+ },
2889
+ {
2890
+ "epoch": 2.8791055206149547,
2891
+ "grad_norm": 4.667630672454834,
2892
+ "learning_rate": 8.486373165618448e-06,
2893
+ "loss": 0.0307,
2894
+ "step": 4120
2895
+ },
2896
+ {
2897
+ "epoch": 2.886093640810622,
2898
+ "grad_norm": 0.0070539130829274654,
2899
+ "learning_rate": 8.45842068483578e-06,
2900
+ "loss": 0.0006,
2901
+ "step": 4130
2902
+ },
2903
+ {
2904
+ "epoch": 2.8930817610062896,
2905
+ "grad_norm": 0.029348071664571762,
2906
+ "learning_rate": 8.430468204053111e-06,
2907
+ "loss": 0.0014,
2908
+ "step": 4140
2909
+ },
2910
+ {
2911
+ "epoch": 2.9000698812019565,
2912
+ "grad_norm": 0.12070006877183914,
2913
+ "learning_rate": 8.402515723270442e-06,
2914
+ "loss": 0.0043,
2915
+ "step": 4150
2916
+ },
2917
+ {
2918
+ "epoch": 2.907058001397624,
2919
+ "grad_norm": 0.00619628606364131,
2920
+ "learning_rate": 8.37456324248777e-06,
2921
+ "loss": 0.0004,
2922
+ "step": 4160
2923
+ },
2924
+ {
2925
+ "epoch": 2.9140461215932913,
2926
+ "grad_norm": 15.53851318359375,
2927
+ "learning_rate": 8.346610761705101e-06,
2928
+ "loss": 0.0074,
2929
+ "step": 4170
2930
+ },
2931
+ {
2932
+ "epoch": 2.9210342417889588,
2933
+ "grad_norm": 9.77151870727539,
2934
+ "learning_rate": 8.318658280922433e-06,
2935
+ "loss": 0.1214,
2936
+ "step": 4180
2937
+ },
2938
+ {
2939
+ "epoch": 2.928022361984626,
2940
+ "grad_norm": 0.0042813620530068874,
2941
+ "learning_rate": 8.290705800139764e-06,
2942
+ "loss": 0.0853,
2943
+ "step": 4190
2944
+ },
2945
+ {
2946
+ "epoch": 2.9350104821802936,
2947
+ "grad_norm": 0.0018544025951996446,
2948
+ "learning_rate": 8.262753319357094e-06,
2949
+ "loss": 0.0183,
2950
+ "step": 4200
2951
+ },
2952
+ {
2953
+ "epoch": 2.941998602375961,
2954
+ "grad_norm": 0.0065402681939303875,
2955
+ "learning_rate": 8.234800838574423e-06,
2956
+ "loss": 0.0746,
2957
+ "step": 4210
2958
+ },
2959
+ {
2960
+ "epoch": 2.948986722571628,
2961
+ "grad_norm": 0.0039556859992444515,
2962
+ "learning_rate": 8.206848357791754e-06,
2963
+ "loss": 0.0101,
2964
+ "step": 4220
2965
+ },
2966
+ {
2967
+ "epoch": 2.9559748427672954,
2968
+ "grad_norm": 0.8577067255973816,
2969
+ "learning_rate": 8.178895877009086e-06,
2970
+ "loss": 0.0323,
2971
+ "step": 4230
2972
+ },
2973
+ {
2974
+ "epoch": 2.962962962962963,
2975
+ "grad_norm": 0.013186552561819553,
2976
+ "learning_rate": 8.150943396226417e-06,
2977
+ "loss": 0.0135,
2978
+ "step": 4240
2979
+ },
2980
+ {
2981
+ "epoch": 2.96995108315863,
2982
+ "grad_norm": 0.008083418011665344,
2983
+ "learning_rate": 8.122990915443745e-06,
2984
+ "loss": 0.0133,
2985
+ "step": 4250
2986
+ },
2987
+ {
2988
+ "epoch": 2.9769392033542976,
2989
+ "grad_norm": 13.401435852050781,
2990
+ "learning_rate": 8.095038434661076e-06,
2991
+ "loss": 0.0557,
2992
+ "step": 4260
2993
+ },
2994
+ {
2995
+ "epoch": 2.983927323549965,
2996
+ "grad_norm": 0.00372733804397285,
2997
+ "learning_rate": 8.067085953878407e-06,
2998
+ "loss": 0.0125,
2999
+ "step": 4270
3000
+ },
3001
+ {
3002
+ "epoch": 2.9909154437456325,
3003
+ "grad_norm": 0.21677538752555847,
3004
+ "learning_rate": 8.039133473095739e-06,
3005
+ "loss": 0.0026,
3006
+ "step": 4280
3007
+ },
3008
+ {
3009
+ "epoch": 2.9979035639413,
3010
+ "grad_norm": 0.005610453896224499,
3011
+ "learning_rate": 8.011180992313068e-06,
3012
+ "loss": 0.0241,
3013
+ "step": 4290
3014
+ }
3015
+ ],
3016
+ "logging_steps": 10,
3017
+ "max_steps": 7155,
3018
+ "num_input_tokens_seen": 0,
3019
+ "num_train_epochs": 5,
3020
+ "save_steps": 500,
3021
+ "stateful_callbacks": {
3022
+ "TrainerControl": {
3023
+ "args": {
3024
+ "should_epoch_stop": false,
3025
+ "should_evaluate": false,
3026
+ "should_log": false,
3027
+ "should_save": true,
3028
+ "should_training_stop": false
3029
+ },
3030
+ "attributes": {}
3031
+ }
3032
+ },
3033
+ "total_flos": 4640735930112000.0,
3034
+ "train_batch_size": 8,
3035
+ "trial_name": null,
3036
+ "trial_params": null
3037
+ }
checkpoint-4293/training_args.bin ADDED
@@ -0,0 +1,3 @@
 
 
 
 
1
+ version https://git-lfs.github.com/spec/v1
2
+ oid sha256:444dc926ee87ace38749c0cb481134ef6ff50da3debedc51baf9fb097d31d73d
3
+ size 5841
checkpoint-5724/README.md ADDED
@@ -0,0 +1,206 @@
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
1
+ ---
2
+ base_model: protectai/deberta-v3-base-prompt-injection-v2
3
+ library_name: peft
4
+ tags:
5
+ - base_model:adapter:protectai/deberta-v3-base-prompt-injection-v2
6
+ - lora
7
+ - transformers
8
+ ---
9
+
10
+ # Model Card for Model ID
11
+
12
+ <!-- Provide a quick summary of what the model is/does. -->
13
+
14
+
15
+
16
+ ## Model Details
17
+
18
+ ### Model Description
19
+
20
+ <!-- Provide a longer summary of what this model is. -->
21
+
22
+
23
+
24
+ - **Developed by:** [More Information Needed]
25
+ - **Funded by [optional]:** [More Information Needed]
26
+ - **Shared by [optional]:** [More Information Needed]
27
+ - **Model type:** [More Information Needed]
28
+ - **Language(s) (NLP):** [More Information Needed]
29
+ - **License:** [More Information Needed]
30
+ - **Finetuned from model [optional]:** [More Information Needed]
31
+
32
+ ### Model Sources [optional]
33
+
34
+ <!-- Provide the basic links for the model. -->
35
+
36
+ - **Repository:** [More Information Needed]
37
+ - **Paper [optional]:** [More Information Needed]
38
+ - **Demo [optional]:** [More Information Needed]
39
+
40
+ ## Uses
41
+
42
+ <!-- Address questions around how the model is intended to be used, including the foreseeable users of the model and those affected by the model. -->
43
+
44
+ ### Direct Use
45
+
46
+ <!-- This section is for the model use without fine-tuning or plugging into a larger ecosystem/app. -->
47
+
48
+ [More Information Needed]
49
+
50
+ ### Downstream Use [optional]
51
+
52
+ <!-- This section is for the model use when fine-tuned for a task, or when plugged into a larger ecosystem/app -->
53
+
54
+ [More Information Needed]
55
+
56
+ ### Out-of-Scope Use
57
+
58
+ <!-- This section addresses misuse, malicious use, and uses that the model will not work well for. -->
59
+
60
+ [More Information Needed]
61
+
62
+ ## Bias, Risks, and Limitations
63
+
64
+ <!-- This section is meant to convey both technical and sociotechnical limitations. -->
65
+
66
+ [More Information Needed]
67
+
68
+ ### Recommendations
69
+
70
+ <!-- This section is meant to convey recommendations with respect to the bias, risk, and technical limitations. -->
71
+
72
+ Users (both direct and downstream) should be made aware of the risks, biases and limitations of the model. More information needed for further recommendations.
73
+
74
+ ## How to Get Started with the Model
75
+
76
+ Use the code below to get started with the model.
77
+
78
+ [More Information Needed]
79
+
80
+ ## Training Details
81
+
82
+ ### Training Data
83
+
84
+ <!-- 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. -->
85
+
86
+ [More Information Needed]
87
+
88
+ ### Training Procedure
89
+
90
+ <!-- This relates heavily to the Technical Specifications. Content here should link to that section when it is relevant to the training procedure. -->
91
+
92
+ #### Preprocessing [optional]
93
+
94
+ [More Information Needed]
95
+
96
+
97
+ #### Training Hyperparameters
98
+
99
+ - **Training regime:** [More Information Needed] <!--fp32, fp16 mixed precision, bf16 mixed precision, bf16 non-mixed precision, fp16 non-mixed precision, fp8 mixed precision -->
100
+
101
+ #### Speeds, Sizes, Times [optional]
102
+
103
+ <!-- This section provides information about throughput, start/end time, checkpoint size if relevant, etc. -->
104
+
105
+ [More Information Needed]
106
+
107
+ ## Evaluation
108
+
109
+ <!-- This section describes the evaluation protocols and provides the results. -->
110
+
111
+ ### Testing Data, Factors & Metrics
112
+
113
+ #### Testing Data
114
+
115
+ <!-- This should link to a Dataset Card if possible. -->
116
+
117
+ [More Information Needed]
118
+
119
+ #### Factors
120
+
121
+ <!-- These are the things the evaluation is disaggregating by, e.g., subpopulations or domains. -->
122
+
123
+ [More Information Needed]
124
+
125
+ #### Metrics
126
+
127
+ <!-- These are the evaluation metrics being used, ideally with a description of why. -->
128
+
129
+ [More Information Needed]
130
+
131
+ ### Results
132
+
133
+ [More Information Needed]
134
+
135
+ #### Summary
136
+
137
+
138
+
139
+ ## Model Examination [optional]
140
+
141
+ <!-- Relevant interpretability work for the model goes here -->
142
+
143
+ [More Information Needed]
144
+
145
+ ## Environmental Impact
146
+
147
+ <!-- Total emissions (in grams of CO2eq) and additional considerations, such as electricity usage, go here. Edit the suggested text below accordingly -->
148
+
149
+ 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).
150
+
151
+ - **Hardware Type:** [More Information Needed]
152
+ - **Hours used:** [More Information Needed]
153
+ - **Cloud Provider:** [More Information Needed]
154
+ - **Compute Region:** [More Information Needed]
155
+ - **Carbon Emitted:** [More Information Needed]
156
+
157
+ ## Technical Specifications [optional]
158
+
159
+ ### Model Architecture and Objective
160
+
161
+ [More Information Needed]
162
+
163
+ ### Compute Infrastructure
164
+
165
+ [More Information Needed]
166
+
167
+ #### Hardware
168
+
169
+ [More Information Needed]
170
+
171
+ #### Software
172
+
173
+ [More Information Needed]
174
+
175
+ ## Citation [optional]
176
+
177
+ <!-- If there is a paper or blog post introducing the model, the APA and Bibtex information for that should go in this section. -->
178
+
179
+ **BibTeX:**
180
+
181
+ [More Information Needed]
182
+
183
+ **APA:**
184
+
185
+ [More Information Needed]
186
+
187
+ ## Glossary [optional]
188
+
189
+ <!-- If relevant, include terms and calculations in this section that can help readers understand the model or model card. -->
190
+
191
+ [More Information Needed]
192
+
193
+ ## More Information [optional]
194
+
195
+ [More Information Needed]
196
+
197
+ ## Model Card Authors [optional]
198
+
199
+ [More Information Needed]
200
+
201
+ ## Model Card Contact
202
+
203
+ [More Information Needed]
204
+ ### Framework versions
205
+
206
+ - PEFT 0.18.1
checkpoint-5724/adapter_config.json ADDED
@@ -0,0 +1,44 @@
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
1
+ {
2
+ "alora_invocation_tokens": null,
3
+ "alpha_pattern": {},
4
+ "arrow_config": null,
5
+ "auto_mapping": null,
6
+ "base_model_name_or_path": "protectai/deberta-v3-base-prompt-injection-v2",
7
+ "bias": "none",
8
+ "corda_config": null,
9
+ "ensure_weight_tying": false,
10
+ "eva_config": null,
11
+ "exclude_modules": null,
12
+ "fan_in_fan_out": false,
13
+ "inference_mode": true,
14
+ "init_lora_weights": true,
15
+ "layer_replication": null,
16
+ "layers_pattern": null,
17
+ "layers_to_transform": null,
18
+ "loftq_config": {},
19
+ "lora_alpha": 64,
20
+ "lora_bias": false,
21
+ "lora_dropout": 0.05,
22
+ "megatron_config": null,
23
+ "megatron_core": "megatron.core",
24
+ "modules_to_save": [
25
+ "classifier",
26
+ "score"
27
+ ],
28
+ "peft_type": "LORA",
29
+ "peft_version": "0.18.1",
30
+ "qalora_group_size": 16,
31
+ "r": 64,
32
+ "rank_pattern": {},
33
+ "revision": null,
34
+ "target_modules": [
35
+ "query_proj",
36
+ "value_proj"
37
+ ],
38
+ "target_parameters": null,
39
+ "task_type": "SEQ_CLS",
40
+ "trainable_token_indices": null,
41
+ "use_dora": false,
42
+ "use_qalora": false,
43
+ "use_rslora": false
44
+ }
checkpoint-5724/adapter_model.safetensors ADDED
@@ -0,0 +1,3 @@
 
 
 
 
1
+ version https://git-lfs.github.com/spec/v1
2
+ oid sha256:891b998b604087ead9f153b1346aecaa8f82a33093cf58b6b981df9380120b64
3
+ size 9450704
checkpoint-5724/optimizer.pt ADDED
@@ -0,0 +1,3 @@
 
 
 
 
1
+ version https://git-lfs.github.com/spec/v1
2
+ oid sha256:c75a79049ced5b6b19146f6aed7b120bde428b88d8317fe82bfd98acc7a48259
3
+ size 18930123
checkpoint-5724/rng_state.pth ADDED
@@ -0,0 +1,3 @@
 
 
 
 
1
+ version https://git-lfs.github.com/spec/v1
2
+ oid sha256:76564799adaa2e1d925b375a2fb86ec3a8c8c7d44af72e32c316a7a2f8a20a78
3
+ size 14645
checkpoint-5724/scaler.pt ADDED
@@ -0,0 +1,3 @@
 
 
 
 
1
+ version https://git-lfs.github.com/spec/v1
2
+ oid sha256:43ef4f4f4f5453d378cb604946396b9badfe476e3d04552b5b89c8cc00f63bcd
3
+ size 1383
checkpoint-5724/scheduler.pt ADDED
@@ -0,0 +1,3 @@
 
 
 
 
1
+ version https://git-lfs.github.com/spec/v1
2
+ oid sha256:5f3d98a917e29550ae8742277c173360967e315119a2ac7e273fd73b4e3ad814
3
+ size 1465
checkpoint-5724/special_tokens_map.json ADDED
@@ -0,0 +1,51 @@
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
1
+ {
2
+ "bos_token": {
3
+ "content": "[CLS]",
4
+ "lstrip": false,
5
+ "normalized": false,
6
+ "rstrip": false,
7
+ "single_word": false
8
+ },
9
+ "cls_token": {
10
+ "content": "[CLS]",
11
+ "lstrip": false,
12
+ "normalized": false,
13
+ "rstrip": false,
14
+ "single_word": false
15
+ },
16
+ "eos_token": {
17
+ "content": "[SEP]",
18
+ "lstrip": false,
19
+ "normalized": false,
20
+ "rstrip": false,
21
+ "single_word": false
22
+ },
23
+ "mask_token": {
24
+ "content": "[MASK]",
25
+ "lstrip": false,
26
+ "normalized": false,
27
+ "rstrip": false,
28
+ "single_word": false
29
+ },
30
+ "pad_token": {
31
+ "content": "[PAD]",
32
+ "lstrip": false,
33
+ "normalized": false,
34
+ "rstrip": false,
35
+ "single_word": false
36
+ },
37
+ "sep_token": {
38
+ "content": "[SEP]",
39
+ "lstrip": false,
40
+ "normalized": false,
41
+ "rstrip": false,
42
+ "single_word": false
43
+ },
44
+ "unk_token": {
45
+ "content": "[UNK]",
46
+ "lstrip": false,
47
+ "normalized": true,
48
+ "rstrip": false,
49
+ "single_word": false
50
+ }
51
+ }
checkpoint-5724/tokenizer.json ADDED
The diff for this file is too large to render. See raw diff
 
checkpoint-5724/tokenizer_config.json ADDED
@@ -0,0 +1,63 @@
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
1
+ {
2
+ "added_tokens_decoder": {
3
+ "0": {
4
+ "content": "[PAD]",
5
+ "lstrip": false,
6
+ "normalized": false,
7
+ "rstrip": false,
8
+ "single_word": false,
9
+ "special": true
10
+ },
11
+ "1": {
12
+ "content": "[CLS]",
13
+ "lstrip": false,
14
+ "normalized": false,
15
+ "rstrip": false,
16
+ "single_word": false,
17
+ "special": true
18
+ },
19
+ "2": {
20
+ "content": "[SEP]",
21
+ "lstrip": false,
22
+ "normalized": false,
23
+ "rstrip": false,
24
+ "single_word": false,
25
+ "special": true
26
+ },
27
+ "3": {
28
+ "content": "[UNK]",
29
+ "lstrip": false,
30
+ "normalized": true,
31
+ "rstrip": false,
32
+ "single_word": false,
33
+ "special": true
34
+ },
35
+ "128000": {
36
+ "content": "[MASK]",
37
+ "lstrip": false,
38
+ "normalized": false,
39
+ "rstrip": false,
40
+ "single_word": false,
41
+ "special": true
42
+ }
43
+ },
44
+ "bos_token": "[CLS]",
45
+ "clean_up_tokenization_spaces": true,
46
+ "cls_token": "[CLS]",
47
+ "do_lower_case": false,
48
+ "eos_token": "[SEP]",
49
+ "extra_special_tokens": {},
50
+ "mask_token": "[MASK]",
51
+ "max_length": 512,
52
+ "model_max_length": 1000000000000000019884624838656,
53
+ "pad_token": "[PAD]",
54
+ "sep_token": "[SEP]",
55
+ "sp_model_kwargs": {},
56
+ "split_by_punct": false,
57
+ "stride": 0,
58
+ "tokenizer_class": "DebertaV2Tokenizer",
59
+ "truncation_side": "right",
60
+ "truncation_strategy": "longest_first",
61
+ "unk_token": "[UNK]",
62
+ "vocab_type": "spm"
63
+ }
checkpoint-5724/trainer_state.json ADDED
The diff for this file is too large to render. See raw diff
 
checkpoint-5724/training_args.bin ADDED
@@ -0,0 +1,3 @@
 
 
 
 
1
+ version https://git-lfs.github.com/spec/v1
2
+ oid sha256:444dc926ee87ace38749c0cb481134ef6ff50da3debedc51baf9fb097d31d73d
3
+ size 5841
checkpoint-7155/README.md ADDED
@@ -0,0 +1,206 @@
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
1
+ ---
2
+ base_model: protectai/deberta-v3-base-prompt-injection-v2
3
+ library_name: peft
4
+ tags:
5
+ - base_model:adapter:protectai/deberta-v3-base-prompt-injection-v2
6
+ - lora
7
+ - transformers
8
+ ---
9
+
10
+ # Model Card for Model ID
11
+
12
+ <!-- Provide a quick summary of what the model is/does. -->
13
+
14
+
15
+
16
+ ## Model Details
17
+
18
+ ### Model Description
19
+
20
+ <!-- Provide a longer summary of what this model is. -->
21
+
22
+
23
+
24
+ - **Developed by:** [More Information Needed]
25
+ - **Funded by [optional]:** [More Information Needed]
26
+ - **Shared by [optional]:** [More Information Needed]
27
+ - **Model type:** [More Information Needed]
28
+ - **Language(s) (NLP):** [More Information Needed]
29
+ - **License:** [More Information Needed]
30
+ - **Finetuned from model [optional]:** [More Information Needed]
31
+
32
+ ### Model Sources [optional]
33
+
34
+ <!-- Provide the basic links for the model. -->
35
+
36
+ - **Repository:** [More Information Needed]
37
+ - **Paper [optional]:** [More Information Needed]
38
+ - **Demo [optional]:** [More Information Needed]
39
+
40
+ ## Uses
41
+
42
+ <!-- Address questions around how the model is intended to be used, including the foreseeable users of the model and those affected by the model. -->
43
+
44
+ ### Direct Use
45
+
46
+ <!-- This section is for the model use without fine-tuning or plugging into a larger ecosystem/app. -->
47
+
48
+ [More Information Needed]
49
+
50
+ ### Downstream Use [optional]
51
+
52
+ <!-- This section is for the model use when fine-tuned for a task, or when plugged into a larger ecosystem/app -->
53
+
54
+ [More Information Needed]
55
+
56
+ ### Out-of-Scope Use
57
+
58
+ <!-- This section addresses misuse, malicious use, and uses that the model will not work well for. -->
59
+
60
+ [More Information Needed]
61
+
62
+ ## Bias, Risks, and Limitations
63
+
64
+ <!-- This section is meant to convey both technical and sociotechnical limitations. -->
65
+
66
+ [More Information Needed]
67
+
68
+ ### Recommendations
69
+
70
+ <!-- This section is meant to convey recommendations with respect to the bias, risk, and technical limitations. -->
71
+
72
+ Users (both direct and downstream) should be made aware of the risks, biases and limitations of the model. More information needed for further recommendations.
73
+
74
+ ## How to Get Started with the Model
75
+
76
+ Use the code below to get started with the model.
77
+
78
+ [More Information Needed]
79
+
80
+ ## Training Details
81
+
82
+ ### Training Data
83
+
84
+ <!-- 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. -->
85
+
86
+ [More Information Needed]
87
+
88
+ ### Training Procedure
89
+
90
+ <!-- This relates heavily to the Technical Specifications. Content here should link to that section when it is relevant to the training procedure. -->
91
+
92
+ #### Preprocessing [optional]
93
+
94
+ [More Information Needed]
95
+
96
+
97
+ #### Training Hyperparameters
98
+
99
+ - **Training regime:** [More Information Needed] <!--fp32, fp16 mixed precision, bf16 mixed precision, bf16 non-mixed precision, fp16 non-mixed precision, fp8 mixed precision -->
100
+
101
+ #### Speeds, Sizes, Times [optional]
102
+
103
+ <!-- This section provides information about throughput, start/end time, checkpoint size if relevant, etc. -->
104
+
105
+ [More Information Needed]
106
+
107
+ ## Evaluation
108
+
109
+ <!-- This section describes the evaluation protocols and provides the results. -->
110
+
111
+ ### Testing Data, Factors & Metrics
112
+
113
+ #### Testing Data
114
+
115
+ <!-- This should link to a Dataset Card if possible. -->
116
+
117
+ [More Information Needed]
118
+
119
+ #### Factors
120
+
121
+ <!-- These are the things the evaluation is disaggregating by, e.g., subpopulations or domains. -->
122
+
123
+ [More Information Needed]
124
+
125
+ #### Metrics
126
+
127
+ <!-- These are the evaluation metrics being used, ideally with a description of why. -->
128
+
129
+ [More Information Needed]
130
+
131
+ ### Results
132
+
133
+ [More Information Needed]
134
+
135
+ #### Summary
136
+
137
+
138
+
139
+ ## Model Examination [optional]
140
+
141
+ <!-- Relevant interpretability work for the model goes here -->
142
+
143
+ [More Information Needed]
144
+
145
+ ## Environmental Impact
146
+
147
+ <!-- Total emissions (in grams of CO2eq) and additional considerations, such as electricity usage, go here. Edit the suggested text below accordingly -->
148
+
149
+ 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).
150
+
151
+ - **Hardware Type:** [More Information Needed]
152
+ - **Hours used:** [More Information Needed]
153
+ - **Cloud Provider:** [More Information Needed]
154
+ - **Compute Region:** [More Information Needed]
155
+ - **Carbon Emitted:** [More Information Needed]
156
+
157
+ ## Technical Specifications [optional]
158
+
159
+ ### Model Architecture and Objective
160
+
161
+ [More Information Needed]
162
+
163
+ ### Compute Infrastructure
164
+
165
+ [More Information Needed]
166
+
167
+ #### Hardware
168
+
169
+ [More Information Needed]
170
+
171
+ #### Software
172
+
173
+ [More Information Needed]
174
+
175
+ ## Citation [optional]
176
+
177
+ <!-- If there is a paper or blog post introducing the model, the APA and Bibtex information for that should go in this section. -->
178
+
179
+ **BibTeX:**
180
+
181
+ [More Information Needed]
182
+
183
+ **APA:**
184
+
185
+ [More Information Needed]
186
+
187
+ ## Glossary [optional]
188
+
189
+ <!-- If relevant, include terms and calculations in this section that can help readers understand the model or model card. -->
190
+
191
+ [More Information Needed]
192
+
193
+ ## More Information [optional]
194
+
195
+ [More Information Needed]
196
+
197
+ ## Model Card Authors [optional]
198
+
199
+ [More Information Needed]
200
+
201
+ ## Model Card Contact
202
+
203
+ [More Information Needed]
204
+ ### Framework versions
205
+
206
+ - PEFT 0.18.1