Upload folder using huggingface_hub
Browse files- README.md +204 -0
- adapter_config.json +27 -0
- adapter_model.safetensors +3 -0
- optimizer.pt +3 -0
- rng_state.pth +3 -0
- scheduler.pt +3 -0
- special_tokens_map.json +34 -0
- tokenizer.model +3 -0
- tokenizer_config.json +72 -0
- trainer_state.json +3381 -0
- training_args.bin +3 -0
README.md
ADDED
@@ -0,0 +1,204 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
1 |
+
---
|
2 |
+
library_name: peft
|
3 |
+
base_model: google/gemma-2b-it
|
4 |
+
---
|
5 |
+
|
6 |
+
# Model Card for Model ID
|
7 |
+
|
8 |
+
<!-- Provide a quick summary of what the model is/does. -->
|
9 |
+
|
10 |
+
|
11 |
+
|
12 |
+
## Model Details
|
13 |
+
|
14 |
+
### Model Description
|
15 |
+
|
16 |
+
<!-- Provide a longer summary of what this model is. -->
|
17 |
+
|
18 |
+
|
19 |
+
|
20 |
+
- **Developed by:** [More Information Needed]
|
21 |
+
- **Funded by [optional]:** [More Information Needed]
|
22 |
+
- **Shared by [optional]:** [More Information Needed]
|
23 |
+
- **Model type:** [More Information Needed]
|
24 |
+
- **Language(s) (NLP):** [More Information Needed]
|
25 |
+
- **License:** [More Information Needed]
|
26 |
+
- **Finetuned from model [optional]:** [More Information Needed]
|
27 |
+
|
28 |
+
### Model Sources [optional]
|
29 |
+
|
30 |
+
<!-- Provide the basic links for the model. -->
|
31 |
+
|
32 |
+
- **Repository:** [More Information Needed]
|
33 |
+
- **Paper [optional]:** [More Information Needed]
|
34 |
+
- **Demo [optional]:** [More Information Needed]
|
35 |
+
|
36 |
+
## Uses
|
37 |
+
|
38 |
+
<!-- Address questions around how the model is intended to be used, including the foreseeable users of the model and those affected by the model. -->
|
39 |
+
|
40 |
+
### Direct Use
|
41 |
+
|
42 |
+
<!-- This section is for the model use without fine-tuning or plugging into a larger ecosystem/app. -->
|
43 |
+
|
44 |
+
[More Information Needed]
|
45 |
+
|
46 |
+
### Downstream Use [optional]
|
47 |
+
|
48 |
+
<!-- This section is for the model use when fine-tuned for a task, or when plugged into a larger ecosystem/app -->
|
49 |
+
|
50 |
+
[More Information Needed]
|
51 |
+
|
52 |
+
### Out-of-Scope Use
|
53 |
+
|
54 |
+
<!-- This section addresses misuse, malicious use, and uses that the model will not work well for. -->
|
55 |
+
|
56 |
+
[More Information Needed]
|
57 |
+
|
58 |
+
## Bias, Risks, and Limitations
|
59 |
+
|
60 |
+
<!-- This section is meant to convey both technical and sociotechnical limitations. -->
|
61 |
+
|
62 |
+
[More Information Needed]
|
63 |
+
|
64 |
+
### Recommendations
|
65 |
+
|
66 |
+
<!-- This section is meant to convey recommendations with respect to the bias, risk, and technical limitations. -->
|
67 |
+
|
68 |
+
Users (both direct and downstream) should be made aware of the risks, biases and limitations of the model. More information needed for further recommendations.
|
69 |
+
|
70 |
+
## How to Get Started with the Model
|
71 |
+
|
72 |
+
Use the code below to get started with the model.
|
73 |
+
|
74 |
+
[More Information Needed]
|
75 |
+
|
76 |
+
## Training Details
|
77 |
+
|
78 |
+
### Training Data
|
79 |
+
|
80 |
+
<!-- This should link to a Dataset Card, perhaps with a short stub of information on what the training data is all about as well as documentation related to data pre-processing or additional filtering. -->
|
81 |
+
|
82 |
+
[More Information Needed]
|
83 |
+
|
84 |
+
### Training Procedure
|
85 |
+
|
86 |
+
<!-- This relates heavily to the Technical Specifications. Content here should link to that section when it is relevant to the training procedure. -->
|
87 |
+
|
88 |
+
#### Preprocessing [optional]
|
89 |
+
|
90 |
+
[More Information Needed]
|
91 |
+
|
92 |
+
|
93 |
+
#### Training Hyperparameters
|
94 |
+
|
95 |
+
- **Training regime:** [More Information Needed] <!--fp32, fp16 mixed precision, bf16 mixed precision, bf16 non-mixed precision, fp16 non-mixed precision, fp8 mixed precision -->
|
96 |
+
|
97 |
+
#### Speeds, Sizes, Times [optional]
|
98 |
+
|
99 |
+
<!-- This section provides information about throughput, start/end time, checkpoint size if relevant, etc. -->
|
100 |
+
|
101 |
+
[More Information Needed]
|
102 |
+
|
103 |
+
## Evaluation
|
104 |
+
|
105 |
+
<!-- This section describes the evaluation protocols and provides the results. -->
|
106 |
+
|
107 |
+
### Testing Data, Factors & Metrics
|
108 |
+
|
109 |
+
#### Testing Data
|
110 |
+
|
111 |
+
<!-- This should link to a Dataset Card if possible. -->
|
112 |
+
|
113 |
+
[More Information Needed]
|
114 |
+
|
115 |
+
#### Factors
|
116 |
+
|
117 |
+
<!-- These are the things the evaluation is disaggregating by, e.g., subpopulations or domains. -->
|
118 |
+
|
119 |
+
[More Information Needed]
|
120 |
+
|
121 |
+
#### Metrics
|
122 |
+
|
123 |
+
<!-- These are the evaluation metrics being used, ideally with a description of why. -->
|
124 |
+
|
125 |
+
[More Information Needed]
|
126 |
+
|
127 |
+
### Results
|
128 |
+
|
129 |
+
[More Information Needed]
|
130 |
+
|
131 |
+
#### Summary
|
132 |
+
|
133 |
+
|
134 |
+
|
135 |
+
## Model Examination [optional]
|
136 |
+
|
137 |
+
<!-- Relevant interpretability work for the model goes here -->
|
138 |
+
|
139 |
+
[More Information Needed]
|
140 |
+
|
141 |
+
## Environmental Impact
|
142 |
+
|
143 |
+
<!-- Total emissions (in grams of CO2eq) and additional considerations, such as electricity usage, go here. Edit the suggested text below accordingly -->
|
144 |
+
|
145 |
+
Carbon emissions can be estimated using the [Machine Learning Impact calculator](https://mlco2.github.io/impact#compute) presented in [Lacoste et al. (2019)](https://arxiv.org/abs/1910.09700).
|
146 |
+
|
147 |
+
- **Hardware Type:** [More Information Needed]
|
148 |
+
- **Hours used:** [More Information Needed]
|
149 |
+
- **Cloud Provider:** [More Information Needed]
|
150 |
+
- **Compute Region:** [More Information Needed]
|
151 |
+
- **Carbon Emitted:** [More Information Needed]
|
152 |
+
|
153 |
+
## Technical Specifications [optional]
|
154 |
+
|
155 |
+
### Model Architecture and Objective
|
156 |
+
|
157 |
+
[More Information Needed]
|
158 |
+
|
159 |
+
### Compute Infrastructure
|
160 |
+
|
161 |
+
[More Information Needed]
|
162 |
+
|
163 |
+
#### Hardware
|
164 |
+
|
165 |
+
[More Information Needed]
|
166 |
+
|
167 |
+
#### Software
|
168 |
+
|
169 |
+
[More Information Needed]
|
170 |
+
|
171 |
+
## Citation [optional]
|
172 |
+
|
173 |
+
<!-- If there is a paper or blog post introducing the model, the APA and Bibtex information for that should go in this section. -->
|
174 |
+
|
175 |
+
**BibTeX:**
|
176 |
+
|
177 |
+
[More Information Needed]
|
178 |
+
|
179 |
+
**APA:**
|
180 |
+
|
181 |
+
[More Information Needed]
|
182 |
+
|
183 |
+
## Glossary [optional]
|
184 |
+
|
185 |
+
<!-- If relevant, include terms and calculations in this section that can help readers understand the model or model card. -->
|
186 |
+
|
187 |
+
[More Information Needed]
|
188 |
+
|
189 |
+
## More Information [optional]
|
190 |
+
|
191 |
+
[More Information Needed]
|
192 |
+
|
193 |
+
## Model Card Authors [optional]
|
194 |
+
|
195 |
+
[More Information Needed]
|
196 |
+
|
197 |
+
## Model Card Contact
|
198 |
+
|
199 |
+
[More Information Needed]
|
200 |
+
|
201 |
+
|
202 |
+
### Framework versions
|
203 |
+
|
204 |
+
- PEFT 0.8.2
|
adapter_config.json
ADDED
@@ -0,0 +1,27 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
1 |
+
{
|
2 |
+
"alpha_pattern": {},
|
3 |
+
"auto_mapping": null,
|
4 |
+
"base_model_name_or_path": "google/gemma-2b-it",
|
5 |
+
"bias": "none",
|
6 |
+
"fan_in_fan_out": false,
|
7 |
+
"inference_mode": true,
|
8 |
+
"init_lora_weights": true,
|
9 |
+
"layers_pattern": null,
|
10 |
+
"layers_to_transform": null,
|
11 |
+
"loftq_config": {},
|
12 |
+
"lora_alpha": 64,
|
13 |
+
"lora_dropout": 0.1,
|
14 |
+
"megatron_config": null,
|
15 |
+
"megatron_core": "megatron.core",
|
16 |
+
"modules_to_save": null,
|
17 |
+
"peft_type": "LORA",
|
18 |
+
"r": 32,
|
19 |
+
"rank_pattern": {},
|
20 |
+
"revision": null,
|
21 |
+
"target_modules": [
|
22 |
+
"q_proj",
|
23 |
+
"v_proj"
|
24 |
+
],
|
25 |
+
"task_type": "CAUSAL_LM",
|
26 |
+
"use_rslora": false
|
27 |
+
}
|
adapter_model.safetensors
ADDED
@@ -0,0 +1,3 @@
|
|
|
|
|
|
|
|
|
1 |
+
version https://git-lfs.github.com/spec/v1
|
2 |
+
oid sha256:56bd20cab045f8d5d8a31d2d4c04b548a4f64b4cd7feef336e1e2c30360ef100
|
3 |
+
size 14755192
|
optimizer.pt
ADDED
@@ -0,0 +1,3 @@
|
|
|
|
|
|
|
|
|
1 |
+
version https://git-lfs.github.com/spec/v1
|
2 |
+
oid sha256:1e68596349d7d9c612bdefe867b2754908eae9f64ca63d17861ceb4949c8c35b
|
3 |
+
size 29552186
|
rng_state.pth
ADDED
@@ -0,0 +1,3 @@
|
|
|
|
|
|
|
|
|
1 |
+
version https://git-lfs.github.com/spec/v1
|
2 |
+
oid sha256:7708c61cf9981be27a3a7ee9f447070cf59c6c04619c28f1bf0afc5cb5a09def
|
3 |
+
size 14244
|
scheduler.pt
ADDED
@@ -0,0 +1,3 @@
|
|
|
|
|
|
|
|
|
1 |
+
version https://git-lfs.github.com/spec/v1
|
2 |
+
oid sha256:b8b36ff0b55f2aa8f7f11f528e2c380ce7e2763940bd36cf0d5265cd02f4ba92
|
3 |
+
size 1064
|
special_tokens_map.json
ADDED
@@ -0,0 +1,34 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
1 |
+
{
|
2 |
+
"additional_special_tokens": [
|
3 |
+
"<start_of_turn>",
|
4 |
+
"<end_of_turn>"
|
5 |
+
],
|
6 |
+
"bos_token": {
|
7 |
+
"content": "<bos>",
|
8 |
+
"lstrip": false,
|
9 |
+
"normalized": false,
|
10 |
+
"rstrip": false,
|
11 |
+
"single_word": false
|
12 |
+
},
|
13 |
+
"eos_token": {
|
14 |
+
"content": "<end_of_turn>",
|
15 |
+
"lstrip": false,
|
16 |
+
"normalized": false,
|
17 |
+
"rstrip": false,
|
18 |
+
"single_word": false
|
19 |
+
},
|
20 |
+
"pad_token": {
|
21 |
+
"content": "<pad>",
|
22 |
+
"lstrip": false,
|
23 |
+
"normalized": false,
|
24 |
+
"rstrip": false,
|
25 |
+
"single_word": false
|
26 |
+
},
|
27 |
+
"unk_token": {
|
28 |
+
"content": "<unk>",
|
29 |
+
"lstrip": false,
|
30 |
+
"normalized": false,
|
31 |
+
"rstrip": false,
|
32 |
+
"single_word": false
|
33 |
+
}
|
34 |
+
}
|
tokenizer.model
ADDED
@@ -0,0 +1,3 @@
|
|
|
|
|
|
|
|
|
1 |
+
version https://git-lfs.github.com/spec/v1
|
2 |
+
oid sha256:6969e64047744a44bb3abfb5c50f8de0f7ed8b571d5444426ef931f651d1a0ef
|
3 |
+
size 4241111
|
tokenizer_config.json
ADDED
@@ -0,0 +1,72 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
1 |
+
{
|
2 |
+
"add_bos_token": true,
|
3 |
+
"add_eos_token": false,
|
4 |
+
"added_tokens_decoder": {
|
5 |
+
"0": {
|
6 |
+
"content": "<pad>",
|
7 |
+
"lstrip": false,
|
8 |
+
"normalized": false,
|
9 |
+
"rstrip": false,
|
10 |
+
"single_word": false,
|
11 |
+
"special": true
|
12 |
+
},
|
13 |
+
"1": {
|
14 |
+
"content": "<eos>",
|
15 |
+
"lstrip": false,
|
16 |
+
"normalized": false,
|
17 |
+
"rstrip": false,
|
18 |
+
"single_word": false,
|
19 |
+
"special": true
|
20 |
+
},
|
21 |
+
"2": {
|
22 |
+
"content": "<bos>",
|
23 |
+
"lstrip": false,
|
24 |
+
"normalized": false,
|
25 |
+
"rstrip": false,
|
26 |
+
"single_word": false,
|
27 |
+
"special": true
|
28 |
+
},
|
29 |
+
"3": {
|
30 |
+
"content": "<unk>",
|
31 |
+
"lstrip": false,
|
32 |
+
"normalized": false,
|
33 |
+
"rstrip": false,
|
34 |
+
"single_word": false,
|
35 |
+
"special": true
|
36 |
+
},
|
37 |
+
"106": {
|
38 |
+
"content": "<start_of_turn>",
|
39 |
+
"lstrip": false,
|
40 |
+
"normalized": false,
|
41 |
+
"rstrip": false,
|
42 |
+
"single_word": false,
|
43 |
+
"special": true
|
44 |
+
},
|
45 |
+
"107": {
|
46 |
+
"content": "<end_of_turn>",
|
47 |
+
"lstrip": false,
|
48 |
+
"normalized": false,
|
49 |
+
"rstrip": false,
|
50 |
+
"single_word": false,
|
51 |
+
"special": true
|
52 |
+
}
|
53 |
+
},
|
54 |
+
"additional_special_tokens": [
|
55 |
+
"<start_of_turn>",
|
56 |
+
"<end_of_turn>"
|
57 |
+
],
|
58 |
+
"bos_token": "<bos>",
|
59 |
+
"chat_template": "{% if messages[0]['role'] == 'system' %}{{ raise_exception('System role not supported') }}{% endif %}{% for message in messages %}{% if (message['role'] == 'user') != (loop.index0 % 2 == 0) %}{{ raise_exception('Conversation roles must alternate user/assistant/user/assistant/...') }}{% endif %}{% if (message['role'] == 'assistant') %}{% set role = 'model' %}{% else %}{% set role = message['role'] %}{% endif %}{{ '<start_of_turn>' + role + '\n' + message['content'] | trim + '<end_of_turn>\n' }}{% endfor %}{% if add_generation_prompt %}{{'<start_of_turn>model\n'}}{% endif %}",
|
60 |
+
"clean_up_tokenization_spaces": false,
|
61 |
+
"eos_token": "<end_of_turn>",
|
62 |
+
"legacy": null,
|
63 |
+
"model_max_length": 1000000000000000019884624838656,
|
64 |
+
"pad_token": "<pad>",
|
65 |
+
"padding_side": "right",
|
66 |
+
"sp_model_kwargs": {},
|
67 |
+
"spaces_between_special_tokens": false,
|
68 |
+
"split_special_tokens": false,
|
69 |
+
"tokenizer_class": "GemmaTokenizer",
|
70 |
+
"unk_token": "<unk>",
|
71 |
+
"use_default_system_prompt": false
|
72 |
+
}
|
trainer_state.json
ADDED
@@ -0,0 +1,3381 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
1 |
+
{
|
2 |
+
"best_metric": null,
|
3 |
+
"best_model_checkpoint": null,
|
4 |
+
"epoch": 0.96,
|
5 |
+
"eval_steps": 500,
|
6 |
+
"global_step": 2400,
|
7 |
+
"is_hyper_param_search": false,
|
8 |
+
"is_local_process_zero": true,
|
9 |
+
"is_world_process_zero": true,
|
10 |
+
"log_history": [
|
11 |
+
{
|
12 |
+
"epoch": 0.0,
|
13 |
+
"grad_norm": 14.472631454467773,
|
14 |
+
"learning_rate": 2.0000000000000003e-06,
|
15 |
+
"loss": 5.6365,
|
16 |
+
"step": 5
|
17 |
+
},
|
18 |
+
{
|
19 |
+
"epoch": 0.0,
|
20 |
+
"grad_norm": NaN,
|
21 |
+
"learning_rate": 6e-06,
|
22 |
+
"loss": 7.9556,
|
23 |
+
"step": 10
|
24 |
+
},
|
25 |
+
{
|
26 |
+
"epoch": 0.01,
|
27 |
+
"grad_norm": 9.6599702835083,
|
28 |
+
"learning_rate": 1.1000000000000001e-05,
|
29 |
+
"loss": 6.5246,
|
30 |
+
"step": 15
|
31 |
+
},
|
32 |
+
{
|
33 |
+
"epoch": 0.01,
|
34 |
+
"grad_norm": 40.07120895385742,
|
35 |
+
"learning_rate": 1.6000000000000003e-05,
|
36 |
+
"loss": 7.7818,
|
37 |
+
"step": 20
|
38 |
+
},
|
39 |
+
{
|
40 |
+
"epoch": 0.01,
|
41 |
+
"grad_norm": 5.662575721740723,
|
42 |
+
"learning_rate": 2.1e-05,
|
43 |
+
"loss": 5.5899,
|
44 |
+
"step": 25
|
45 |
+
},
|
46 |
+
{
|
47 |
+
"epoch": 0.01,
|
48 |
+
"grad_norm": NaN,
|
49 |
+
"learning_rate": 2.5e-05,
|
50 |
+
"loss": 7.1589,
|
51 |
+
"step": 30
|
52 |
+
},
|
53 |
+
{
|
54 |
+
"epoch": 0.01,
|
55 |
+
"grad_norm": 7.6021246910095215,
|
56 |
+
"learning_rate": 3e-05,
|
57 |
+
"loss": 4.9643,
|
58 |
+
"step": 35
|
59 |
+
},
|
60 |
+
{
|
61 |
+
"epoch": 0.02,
|
62 |
+
"grad_norm": 7.397390365600586,
|
63 |
+
"learning_rate": 3.5e-05,
|
64 |
+
"loss": 5.944,
|
65 |
+
"step": 40
|
66 |
+
},
|
67 |
+
{
|
68 |
+
"epoch": 0.02,
|
69 |
+
"grad_norm": 10.122846603393555,
|
70 |
+
"learning_rate": 4e-05,
|
71 |
+
"loss": 4.2179,
|
72 |
+
"step": 45
|
73 |
+
},
|
74 |
+
{
|
75 |
+
"epoch": 0.02,
|
76 |
+
"grad_norm": 7.520490646362305,
|
77 |
+
"learning_rate": 4.5e-05,
|
78 |
+
"loss": 3.9576,
|
79 |
+
"step": 50
|
80 |
+
},
|
81 |
+
{
|
82 |
+
"epoch": 0.02,
|
83 |
+
"grad_norm": 12.957700729370117,
|
84 |
+
"learning_rate": 5e-05,
|
85 |
+
"loss": 4.6752,
|
86 |
+
"step": 55
|
87 |
+
},
|
88 |
+
{
|
89 |
+
"epoch": 0.02,
|
90 |
+
"grad_norm": 11.252644538879395,
|
91 |
+
"learning_rate": 4.999948617395915e-05,
|
92 |
+
"loss": 4.586,
|
93 |
+
"step": 60
|
94 |
+
},
|
95 |
+
{
|
96 |
+
"epoch": 0.03,
|
97 |
+
"grad_norm": 14.29404354095459,
|
98 |
+
"learning_rate": 4.9997944716957985e-05,
|
99 |
+
"loss": 4.2009,
|
100 |
+
"step": 65
|
101 |
+
},
|
102 |
+
{
|
103 |
+
"epoch": 0.03,
|
104 |
+
"grad_norm": 12.03260612487793,
|
105 |
+
"learning_rate": 4.9995375692359755e-05,
|
106 |
+
"loss": 4.7383,
|
107 |
+
"step": 70
|
108 |
+
},
|
109 |
+
{
|
110 |
+
"epoch": 0.03,
|
111 |
+
"grad_norm": 7.339493751525879,
|
112 |
+
"learning_rate": 4.9991779205767e-05,
|
113 |
+
"loss": 4.0294,
|
114 |
+
"step": 75
|
115 |
+
},
|
116 |
+
{
|
117 |
+
"epoch": 0.03,
|
118 |
+
"grad_norm": 5.51794958114624,
|
119 |
+
"learning_rate": 4.99871554050172e-05,
|
120 |
+
"loss": 3.8225,
|
121 |
+
"step": 80
|
122 |
+
},
|
123 |
+
{
|
124 |
+
"epoch": 0.03,
|
125 |
+
"grad_norm": 10.195987701416016,
|
126 |
+
"learning_rate": 4.9981504480176696e-05,
|
127 |
+
"loss": 4.1299,
|
128 |
+
"step": 85
|
129 |
+
},
|
130 |
+
{
|
131 |
+
"epoch": 0.04,
|
132 |
+
"grad_norm": 14.785408973693848,
|
133 |
+
"learning_rate": 4.997482666353287e-05,
|
134 |
+
"loss": 4.1754,
|
135 |
+
"step": 90
|
136 |
+
},
|
137 |
+
{
|
138 |
+
"epoch": 0.04,
|
139 |
+
"grad_norm": 9.586732864379883,
|
140 |
+
"learning_rate": 4.996712222958461e-05,
|
141 |
+
"loss": 3.2669,
|
142 |
+
"step": 95
|
143 |
+
},
|
144 |
+
{
|
145 |
+
"epoch": 0.04,
|
146 |
+
"grad_norm": 25.427532196044922,
|
147 |
+
"learning_rate": 4.9958391495031026e-05,
|
148 |
+
"loss": 4.1074,
|
149 |
+
"step": 100
|
150 |
+
},
|
151 |
+
{
|
152 |
+
"epoch": 0.04,
|
153 |
+
"grad_norm": 7.9541730880737305,
|
154 |
+
"learning_rate": 4.994863481875841e-05,
|
155 |
+
"loss": 4.2565,
|
156 |
+
"step": 105
|
157 |
+
},
|
158 |
+
{
|
159 |
+
"epoch": 0.04,
|
160 |
+
"grad_norm": 8.371864318847656,
|
161 |
+
"learning_rate": 4.993785260182552e-05,
|
162 |
+
"loss": 3.3137,
|
163 |
+
"step": 110
|
164 |
+
},
|
165 |
+
{
|
166 |
+
"epoch": 0.05,
|
167 |
+
"grad_norm": 9.015960693359375,
|
168 |
+
"learning_rate": 4.992604528744705e-05,
|
169 |
+
"loss": 3.4324,
|
170 |
+
"step": 115
|
171 |
+
},
|
172 |
+
{
|
173 |
+
"epoch": 0.05,
|
174 |
+
"grad_norm": 6.063784599304199,
|
175 |
+
"learning_rate": 4.991321336097546e-05,
|
176 |
+
"loss": 3.3621,
|
177 |
+
"step": 120
|
178 |
+
},
|
179 |
+
{
|
180 |
+
"epoch": 0.05,
|
181 |
+
"grad_norm": 7.260740280151367,
|
182 |
+
"learning_rate": 4.989935734988098e-05,
|
183 |
+
"loss": 4.7443,
|
184 |
+
"step": 125
|
185 |
+
},
|
186 |
+
{
|
187 |
+
"epoch": 0.05,
|
188 |
+
"grad_norm": 9.744063377380371,
|
189 |
+
"learning_rate": 4.9884477823729956e-05,
|
190 |
+
"loss": 4.085,
|
191 |
+
"step": 130
|
192 |
+
},
|
193 |
+
{
|
194 |
+
"epoch": 0.05,
|
195 |
+
"grad_norm": 9.156418800354004,
|
196 |
+
"learning_rate": 4.986857539416144e-05,
|
197 |
+
"loss": 3.5387,
|
198 |
+
"step": 135
|
199 |
+
},
|
200 |
+
{
|
201 |
+
"epoch": 0.06,
|
202 |
+
"grad_norm": 7.296212673187256,
|
203 |
+
"learning_rate": 4.9851650714862006e-05,
|
204 |
+
"loss": 3.348,
|
205 |
+
"step": 140
|
206 |
+
},
|
207 |
+
{
|
208 |
+
"epoch": 0.06,
|
209 |
+
"grad_norm": 8.042045593261719,
|
210 |
+
"learning_rate": 4.983370448153896e-05,
|
211 |
+
"loss": 3.5621,
|
212 |
+
"step": 145
|
213 |
+
},
|
214 |
+
{
|
215 |
+
"epoch": 0.06,
|
216 |
+
"grad_norm": 12.138426780700684,
|
217 |
+
"learning_rate": 4.981473743189163e-05,
|
218 |
+
"loss": 3.4145,
|
219 |
+
"step": 150
|
220 |
+
},
|
221 |
+
{
|
222 |
+
"epoch": 0.06,
|
223 |
+
"grad_norm": 12.3366060256958,
|
224 |
+
"learning_rate": 4.979475034558115e-05,
|
225 |
+
"loss": 3.9396,
|
226 |
+
"step": 155
|
227 |
+
},
|
228 |
+
{
|
229 |
+
"epoch": 0.06,
|
230 |
+
"grad_norm": 6.696476936340332,
|
231 |
+
"learning_rate": 4.977374404419837e-05,
|
232 |
+
"loss": 3.5372,
|
233 |
+
"step": 160
|
234 |
+
},
|
235 |
+
{
|
236 |
+
"epoch": 0.07,
|
237 |
+
"grad_norm": 6.023372173309326,
|
238 |
+
"learning_rate": 4.975171939123005e-05,
|
239 |
+
"loss": 3.2419,
|
240 |
+
"step": 165
|
241 |
+
},
|
242 |
+
{
|
243 |
+
"epoch": 0.07,
|
244 |
+
"grad_norm": 7.861164093017578,
|
245 |
+
"learning_rate": 4.9728677292023405e-05,
|
246 |
+
"loss": 3.7565,
|
247 |
+
"step": 170
|
248 |
+
},
|
249 |
+
{
|
250 |
+
"epoch": 0.07,
|
251 |
+
"grad_norm": 6.757652282714844,
|
252 |
+
"learning_rate": 4.970461869374889e-05,
|
253 |
+
"loss": 3.2182,
|
254 |
+
"step": 175
|
255 |
+
},
|
256 |
+
{
|
257 |
+
"epoch": 0.07,
|
258 |
+
"grad_norm": 9.434078216552734,
|
259 |
+
"learning_rate": 4.967954458536126e-05,
|
260 |
+
"loss": 3.229,
|
261 |
+
"step": 180
|
262 |
+
},
|
263 |
+
{
|
264 |
+
"epoch": 0.07,
|
265 |
+
"grad_norm": 27.395694732666016,
|
266 |
+
"learning_rate": 4.965345599755887e-05,
|
267 |
+
"loss": 3.903,
|
268 |
+
"step": 185
|
269 |
+
},
|
270 |
+
{
|
271 |
+
"epoch": 0.08,
|
272 |
+
"grad_norm": 6.09558629989624,
|
273 |
+
"learning_rate": 4.962635400274142e-05,
|
274 |
+
"loss": 3.1898,
|
275 |
+
"step": 190
|
276 |
+
},
|
277 |
+
{
|
278 |
+
"epoch": 0.08,
|
279 |
+
"grad_norm": 4.992712497711182,
|
280 |
+
"learning_rate": 4.959823971496574e-05,
|
281 |
+
"loss": 3.2185,
|
282 |
+
"step": 195
|
283 |
+
},
|
284 |
+
{
|
285 |
+
"epoch": 0.08,
|
286 |
+
"grad_norm": 7.1479411125183105,
|
287 |
+
"learning_rate": 4.95691142899001e-05,
|
288 |
+
"loss": 3.2792,
|
289 |
+
"step": 200
|
290 |
+
},
|
291 |
+
{
|
292 |
+
"epoch": 0.08,
|
293 |
+
"grad_norm": 8.431621551513672,
|
294 |
+
"learning_rate": 4.9538978924776634e-05,
|
295 |
+
"loss": 3.1418,
|
296 |
+
"step": 205
|
297 |
+
},
|
298 |
+
{
|
299 |
+
"epoch": 0.08,
|
300 |
+
"grad_norm": 23.00255584716797,
|
301 |
+
"learning_rate": 4.9507834858342186e-05,
|
302 |
+
"loss": 3.9976,
|
303 |
+
"step": 210
|
304 |
+
},
|
305 |
+
{
|
306 |
+
"epoch": 0.09,
|
307 |
+
"grad_norm": 10.47463321685791,
|
308 |
+
"learning_rate": 4.9475683370807326e-05,
|
309 |
+
"loss": 3.4157,
|
310 |
+
"step": 215
|
311 |
+
},
|
312 |
+
{
|
313 |
+
"epoch": 0.09,
|
314 |
+
"grad_norm": 18.375465393066406,
|
315 |
+
"learning_rate": 4.9442525783793794e-05,
|
316 |
+
"loss": 3.5929,
|
317 |
+
"step": 220
|
318 |
+
},
|
319 |
+
{
|
320 |
+
"epoch": 0.09,
|
321 |
+
"grad_norm": 8.78403091430664,
|
322 |
+
"learning_rate": 4.940836346028011e-05,
|
323 |
+
"loss": 3.1698,
|
324 |
+
"step": 225
|
325 |
+
},
|
326 |
+
{
|
327 |
+
"epoch": 0.09,
|
328 |
+
"grad_norm": 7.213769912719727,
|
329 |
+
"learning_rate": 4.937319780454559e-05,
|
330 |
+
"loss": 3.9643,
|
331 |
+
"step": 230
|
332 |
+
},
|
333 |
+
{
|
334 |
+
"epoch": 0.09,
|
335 |
+
"grad_norm": 7.919612884521484,
|
336 |
+
"learning_rate": 4.933703026211262e-05,
|
337 |
+
"loss": 3.6494,
|
338 |
+
"step": 235
|
339 |
+
},
|
340 |
+
{
|
341 |
+
"epoch": 0.1,
|
342 |
+
"grad_norm": 4.598107814788818,
|
343 |
+
"learning_rate": 4.9299862319687204e-05,
|
344 |
+
"loss": 3.5917,
|
345 |
+
"step": 240
|
346 |
+
},
|
347 |
+
{
|
348 |
+
"epoch": 0.1,
|
349 |
+
"grad_norm": 37.90977096557617,
|
350 |
+
"learning_rate": 4.926169550509787e-05,
|
351 |
+
"loss": 3.7991,
|
352 |
+
"step": 245
|
353 |
+
},
|
354 |
+
{
|
355 |
+
"epoch": 0.1,
|
356 |
+
"grad_norm": 12.698101997375488,
|
357 |
+
"learning_rate": 4.9222531387232885e-05,
|
358 |
+
"loss": 3.5116,
|
359 |
+
"step": 250
|
360 |
+
},
|
361 |
+
{
|
362 |
+
"epoch": 0.1,
|
363 |
+
"grad_norm": 9.609795570373535,
|
364 |
+
"learning_rate": 4.9182371575975736e-05,
|
365 |
+
"loss": 3.4126,
|
366 |
+
"step": 255
|
367 |
+
},
|
368 |
+
{
|
369 |
+
"epoch": 0.1,
|
370 |
+
"grad_norm": 8.108003616333008,
|
371 |
+
"learning_rate": 4.914121772213898e-05,
|
372 |
+
"loss": 3.2962,
|
373 |
+
"step": 260
|
374 |
+
},
|
375 |
+
{
|
376 |
+
"epoch": 0.11,
|
377 |
+
"grad_norm": 11.94726848602295,
|
378 |
+
"learning_rate": 4.909907151739633e-05,
|
379 |
+
"loss": 3.6953,
|
380 |
+
"step": 265
|
381 |
+
},
|
382 |
+
{
|
383 |
+
"epoch": 0.11,
|
384 |
+
"grad_norm": 8.730710983276367,
|
385 |
+
"learning_rate": 4.905593469421323e-05,
|
386 |
+
"loss": 3.4841,
|
387 |
+
"step": 270
|
388 |
+
},
|
389 |
+
{
|
390 |
+
"epoch": 0.11,
|
391 |
+
"grad_norm": 6.423217296600342,
|
392 |
+
"learning_rate": 4.9011809025775486e-05,
|
393 |
+
"loss": 3.1157,
|
394 |
+
"step": 275
|
395 |
+
},
|
396 |
+
{
|
397 |
+
"epoch": 0.11,
|
398 |
+
"grad_norm": 10.736145973205566,
|
399 |
+
"learning_rate": 4.8966696325916515e-05,
|
400 |
+
"loss": 3.244,
|
401 |
+
"step": 280
|
402 |
+
},
|
403 |
+
{
|
404 |
+
"epoch": 0.11,
|
405 |
+
"grad_norm": 21.597434997558594,
|
406 |
+
"learning_rate": 4.892059844904272e-05,
|
407 |
+
"loss": 3.172,
|
408 |
+
"step": 285
|
409 |
+
},
|
410 |
+
{
|
411 |
+
"epoch": 0.12,
|
412 |
+
"grad_norm": 23.540430068969727,
|
413 |
+
"learning_rate": 4.887351729005726e-05,
|
414 |
+
"loss": 3.1889,
|
415 |
+
"step": 290
|
416 |
+
},
|
417 |
+
{
|
418 |
+
"epoch": 0.12,
|
419 |
+
"grad_norm": 11.286609649658203,
|
420 |
+
"learning_rate": 4.882545478428218e-05,
|
421 |
+
"loss": 3.5211,
|
422 |
+
"step": 295
|
423 |
+
},
|
424 |
+
{
|
425 |
+
"epoch": 0.12,
|
426 |
+
"grad_norm": 7.204074382781982,
|
427 |
+
"learning_rate": 4.877641290737884e-05,
|
428 |
+
"loss": 3.255,
|
429 |
+
"step": 300
|
430 |
+
},
|
431 |
+
{
|
432 |
+
"epoch": 0.12,
|
433 |
+
"grad_norm": 25.35500717163086,
|
434 |
+
"learning_rate": 4.8726393675266716e-05,
|
435 |
+
"loss": 3.4832,
|
436 |
+
"step": 305
|
437 |
+
},
|
438 |
+
{
|
439 |
+
"epoch": 0.12,
|
440 |
+
"grad_norm": 6.76464319229126,
|
441 |
+
"learning_rate": 4.8675399144040537e-05,
|
442 |
+
"loss": 2.9075,
|
443 |
+
"step": 310
|
444 |
+
},
|
445 |
+
{
|
446 |
+
"epoch": 0.13,
|
447 |
+
"grad_norm": 10.19654655456543,
|
448 |
+
"learning_rate": 4.862343140988573e-05,
|
449 |
+
"loss": 3.0107,
|
450 |
+
"step": 315
|
451 |
+
},
|
452 |
+
{
|
453 |
+
"epoch": 0.13,
|
454 |
+
"grad_norm": 7.545323848724365,
|
455 |
+
"learning_rate": 4.8570492608992325e-05,
|
456 |
+
"loss": 2.7479,
|
457 |
+
"step": 320
|
458 |
+
},
|
459 |
+
{
|
460 |
+
"epoch": 0.13,
|
461 |
+
"grad_norm": 15.334161758422852,
|
462 |
+
"learning_rate": 4.851658491746707e-05,
|
463 |
+
"loss": 2.8654,
|
464 |
+
"step": 325
|
465 |
+
},
|
466 |
+
{
|
467 |
+
"epoch": 0.13,
|
468 |
+
"grad_norm": 14.28708553314209,
|
469 |
+
"learning_rate": 4.846171055124401e-05,
|
470 |
+
"loss": 3.1851,
|
471 |
+
"step": 330
|
472 |
+
},
|
473 |
+
{
|
474 |
+
"epoch": 0.13,
|
475 |
+
"grad_norm": 22.403162002563477,
|
476 |
+
"learning_rate": 4.8405871765993433e-05,
|
477 |
+
"loss": 2.8862,
|
478 |
+
"step": 335
|
479 |
+
},
|
480 |
+
{
|
481 |
+
"epoch": 0.14,
|
482 |
+
"grad_norm": 5.840334415435791,
|
483 |
+
"learning_rate": 4.834907085702908e-05,
|
484 |
+
"loss": 2.968,
|
485 |
+
"step": 340
|
486 |
+
},
|
487 |
+
{
|
488 |
+
"epoch": 0.14,
|
489 |
+
"grad_norm": 12.307807922363281,
|
490 |
+
"learning_rate": 4.829131015921385e-05,
|
491 |
+
"loss": 2.9458,
|
492 |
+
"step": 345
|
493 |
+
},
|
494 |
+
{
|
495 |
+
"epoch": 0.14,
|
496 |
+
"grad_norm": 12.92584228515625,
|
497 |
+
"learning_rate": 4.82325920468638e-05,
|
498 |
+
"loss": 3.4169,
|
499 |
+
"step": 350
|
500 |
+
},
|
501 |
+
{
|
502 |
+
"epoch": 0.14,
|
503 |
+
"grad_norm": 13.389299392700195,
|
504 |
+
"learning_rate": 4.817291893365055e-05,
|
505 |
+
"loss": 3.1601,
|
506 |
+
"step": 355
|
507 |
+
},
|
508 |
+
{
|
509 |
+
"epoch": 0.14,
|
510 |
+
"grad_norm": 16.73059844970703,
|
511 |
+
"learning_rate": 4.8112293272502043e-05,
|
512 |
+
"loss": 3.2846,
|
513 |
+
"step": 360
|
514 |
+
},
|
515 |
+
{
|
516 |
+
"epoch": 0.15,
|
517 |
+
"grad_norm": 8.608989715576172,
|
518 |
+
"learning_rate": 4.805071755550177e-05,
|
519 |
+
"loss": 3.0091,
|
520 |
+
"step": 365
|
521 |
+
},
|
522 |
+
{
|
523 |
+
"epoch": 0.15,
|
524 |
+
"grad_norm": 5.731265544891357,
|
525 |
+
"learning_rate": 4.7988194313786275e-05,
|
526 |
+
"loss": 3.3688,
|
527 |
+
"step": 370
|
528 |
+
},
|
529 |
+
{
|
530 |
+
"epoch": 0.15,
|
531 |
+
"grad_norm": 11.152880668640137,
|
532 |
+
"learning_rate": 4.7924726117441135e-05,
|
533 |
+
"loss": 3.3177,
|
534 |
+
"step": 375
|
535 |
+
},
|
536 |
+
{
|
537 |
+
"epoch": 0.15,
|
538 |
+
"grad_norm": 8.759363174438477,
|
539 |
+
"learning_rate": 4.7860315575395316e-05,
|
540 |
+
"loss": 2.9184,
|
541 |
+
"step": 380
|
542 |
+
},
|
543 |
+
{
|
544 |
+
"epoch": 0.15,
|
545 |
+
"grad_norm": 37.01634216308594,
|
546 |
+
"learning_rate": 4.7794965335313926e-05,
|
547 |
+
"loss": 3.1494,
|
548 |
+
"step": 385
|
549 |
+
},
|
550 |
+
{
|
551 |
+
"epoch": 0.16,
|
552 |
+
"grad_norm": 5.318065643310547,
|
553 |
+
"learning_rate": 4.772867808348938e-05,
|
554 |
+
"loss": 3.7067,
|
555 |
+
"step": 390
|
556 |
+
},
|
557 |
+
{
|
558 |
+
"epoch": 0.16,
|
559 |
+
"grad_norm": 14.883944511413574,
|
560 |
+
"learning_rate": 4.766145654473095e-05,
|
561 |
+
"loss": 2.9832,
|
562 |
+
"step": 395
|
563 |
+
},
|
564 |
+
{
|
565 |
+
"epoch": 0.16,
|
566 |
+
"grad_norm": 35.67652130126953,
|
567 |
+
"learning_rate": 4.759330348225284e-05,
|
568 |
+
"loss": 3.6382,
|
569 |
+
"step": 400
|
570 |
+
},
|
571 |
+
{
|
572 |
+
"epoch": 0.16,
|
573 |
+
"grad_norm": 6.075866222381592,
|
574 |
+
"learning_rate": 4.752422169756048e-05,
|
575 |
+
"loss": 3.6166,
|
576 |
+
"step": 405
|
577 |
+
},
|
578 |
+
{
|
579 |
+
"epoch": 0.16,
|
580 |
+
"grad_norm": 10.807673454284668,
|
581 |
+
"learning_rate": 4.745421403033548e-05,
|
582 |
+
"loss": 2.896,
|
583 |
+
"step": 410
|
584 |
+
},
|
585 |
+
{
|
586 |
+
"epoch": 0.17,
|
587 |
+
"grad_norm": 17.647350311279297,
|
588 |
+
"learning_rate": 4.738328335831883e-05,
|
589 |
+
"loss": 2.9218,
|
590 |
+
"step": 415
|
591 |
+
},
|
592 |
+
{
|
593 |
+
"epoch": 0.17,
|
594 |
+
"grad_norm": 35.758544921875,
|
595 |
+
"learning_rate": 4.731143259719265e-05,
|
596 |
+
"loss": 3.0896,
|
597 |
+
"step": 420
|
598 |
+
},
|
599 |
+
{
|
600 |
+
"epoch": 0.17,
|
601 |
+
"grad_norm": 19.15152931213379,
|
602 |
+
"learning_rate": 4.72386647004603e-05,
|
603 |
+
"loss": 3.0875,
|
604 |
+
"step": 425
|
605 |
+
},
|
606 |
+
{
|
607 |
+
"epoch": 0.17,
|
608 |
+
"grad_norm": 10.721521377563477,
|
609 |
+
"learning_rate": 4.716498265932501e-05,
|
610 |
+
"loss": 2.833,
|
611 |
+
"step": 430
|
612 |
+
},
|
613 |
+
{
|
614 |
+
"epoch": 0.17,
|
615 |
+
"grad_norm": 36.97370147705078,
|
616 |
+
"learning_rate": 4.709038950256688e-05,
|
617 |
+
"loss": 3.3553,
|
618 |
+
"step": 435
|
619 |
+
},
|
620 |
+
{
|
621 |
+
"epoch": 0.18,
|
622 |
+
"grad_norm": 18.96316146850586,
|
623 |
+
"learning_rate": 4.701488829641845e-05,
|
624 |
+
"loss": 2.8189,
|
625 |
+
"step": 440
|
626 |
+
},
|
627 |
+
{
|
628 |
+
"epoch": 0.18,
|
629 |
+
"grad_norm": 8.662888526916504,
|
630 |
+
"learning_rate": 4.693848214443858e-05,
|
631 |
+
"loss": 3.4443,
|
632 |
+
"step": 445
|
633 |
+
},
|
634 |
+
{
|
635 |
+
"epoch": 0.18,
|
636 |
+
"grad_norm": 18.23220443725586,
|
637 |
+
"learning_rate": 4.686117418738489e-05,
|
638 |
+
"loss": 3.5717,
|
639 |
+
"step": 450
|
640 |
+
},
|
641 |
+
{
|
642 |
+
"epoch": 0.18,
|
643 |
+
"grad_norm": 39.92061996459961,
|
644 |
+
"learning_rate": 4.678296760308474e-05,
|
645 |
+
"loss": 3.2364,
|
646 |
+
"step": 455
|
647 |
+
},
|
648 |
+
{
|
649 |
+
"epoch": 0.18,
|
650 |
+
"grad_norm": 9.42786693572998,
|
651 |
+
"learning_rate": 4.6703865606304465e-05,
|
652 |
+
"loss": 2.9595,
|
653 |
+
"step": 460
|
654 |
+
},
|
655 |
+
{
|
656 |
+
"epoch": 0.19,
|
657 |
+
"grad_norm": 7.904202938079834,
|
658 |
+
"learning_rate": 4.662387144861734e-05,
|
659 |
+
"loss": 3.1582,
|
660 |
+
"step": 465
|
661 |
+
},
|
662 |
+
{
|
663 |
+
"epoch": 0.19,
|
664 |
+
"grad_norm": 26.168556213378906,
|
665 |
+
"learning_rate": 4.6542988418269876e-05,
|
666 |
+
"loss": 3.3465,
|
667 |
+
"step": 470
|
668 |
+
},
|
669 |
+
{
|
670 |
+
"epoch": 0.19,
|
671 |
+
"grad_norm": 10.387030601501465,
|
672 |
+
"learning_rate": 4.6461219840046654e-05,
|
673 |
+
"loss": 3.0016,
|
674 |
+
"step": 475
|
675 |
+
},
|
676 |
+
{
|
677 |
+
"epoch": 0.19,
|
678 |
+
"grad_norm": 9.973118782043457,
|
679 |
+
"learning_rate": 4.637856907513366e-05,
|
680 |
+
"loss": 3.2299,
|
681 |
+
"step": 480
|
682 |
+
},
|
683 |
+
{
|
684 |
+
"epoch": 0.19,
|
685 |
+
"grad_norm": 31.876556396484375,
|
686 |
+
"learning_rate": 4.629503952098011e-05,
|
687 |
+
"loss": 3.3682,
|
688 |
+
"step": 485
|
689 |
+
},
|
690 |
+
{
|
691 |
+
"epoch": 0.2,
|
692 |
+
"grad_norm": 23.903682708740234,
|
693 |
+
"learning_rate": 4.6210634611158816e-05,
|
694 |
+
"loss": 2.7361,
|
695 |
+
"step": 490
|
696 |
+
},
|
697 |
+
{
|
698 |
+
"epoch": 0.2,
|
699 |
+
"grad_norm": 7.048882007598877,
|
700 |
+
"learning_rate": 4.612535781522504e-05,
|
701 |
+
"loss": 2.8006,
|
702 |
+
"step": 495
|
703 |
+
},
|
704 |
+
{
|
705 |
+
"epoch": 0.2,
|
706 |
+
"grad_norm": 16.777807235717773,
|
707 |
+
"learning_rate": 4.6039212638573833e-05,
|
708 |
+
"loss": 2.7768,
|
709 |
+
"step": 500
|
710 |
+
},
|
711 |
+
{
|
712 |
+
"epoch": 0.2,
|
713 |
+
"grad_norm": 35.86275100708008,
|
714 |
+
"learning_rate": 4.595220262229601e-05,
|
715 |
+
"loss": 3.2371,
|
716 |
+
"step": 505
|
717 |
+
},
|
718 |
+
{
|
719 |
+
"epoch": 0.2,
|
720 |
+
"grad_norm": 18.5112361907959,
|
721 |
+
"learning_rate": 4.586433134303257e-05,
|
722 |
+
"loss": 3.9903,
|
723 |
+
"step": 510
|
724 |
+
},
|
725 |
+
{
|
726 |
+
"epoch": 0.21,
|
727 |
+
"grad_norm": 15.84546947479248,
|
728 |
+
"learning_rate": 4.5775602412827604e-05,
|
729 |
+
"loss": 3.2624,
|
730 |
+
"step": 515
|
731 |
+
},
|
732 |
+
{
|
733 |
+
"epoch": 0.21,
|
734 |
+
"grad_norm": 9.097062110900879,
|
735 |
+
"learning_rate": 4.5686019478979915e-05,
|
736 |
+
"loss": 3.036,
|
737 |
+
"step": 520
|
738 |
+
},
|
739 |
+
{
|
740 |
+
"epoch": 0.21,
|
741 |
+
"grad_norm": 12.135428428649902,
|
742 |
+
"learning_rate": 4.559558622389304e-05,
|
743 |
+
"loss": 3.0802,
|
744 |
+
"step": 525
|
745 |
+
},
|
746 |
+
{
|
747 |
+
"epoch": 0.21,
|
748 |
+
"grad_norm": 7.016368865966797,
|
749 |
+
"learning_rate": 4.55043063649239e-05,
|
750 |
+
"loss": 2.718,
|
751 |
+
"step": 530
|
752 |
+
},
|
753 |
+
{
|
754 |
+
"epoch": 0.21,
|
755 |
+
"grad_norm": 7.902190685272217,
|
756 |
+
"learning_rate": 4.5412183654229965e-05,
|
757 |
+
"loss": 2.9923,
|
758 |
+
"step": 535
|
759 |
+
},
|
760 |
+
{
|
761 |
+
"epoch": 0.22,
|
762 |
+
"grad_norm": 9.023144721984863,
|
763 |
+
"learning_rate": 4.531922187861507e-05,
|
764 |
+
"loss": 3.2273,
|
765 |
+
"step": 540
|
766 |
+
},
|
767 |
+
{
|
768 |
+
"epoch": 0.22,
|
769 |
+
"grad_norm": 12.821276664733887,
|
770 |
+
"learning_rate": 4.522542485937369e-05,
|
771 |
+
"loss": 3.0269,
|
772 |
+
"step": 545
|
773 |
+
},
|
774 |
+
{
|
775 |
+
"epoch": 0.22,
|
776 |
+
"grad_norm": 11.519277572631836,
|
777 |
+
"learning_rate": 4.51307964521339e-05,
|
778 |
+
"loss": 2.9414,
|
779 |
+
"step": 550
|
780 |
+
},
|
781 |
+
{
|
782 |
+
"epoch": 0.22,
|
783 |
+
"grad_norm": 6.272705554962158,
|
784 |
+
"learning_rate": 4.503534054669892e-05,
|
785 |
+
"loss": 3.1369,
|
786 |
+
"step": 555
|
787 |
+
},
|
788 |
+
{
|
789 |
+
"epoch": 0.22,
|
790 |
+
"grad_norm": 15.541712760925293,
|
791 |
+
"learning_rate": 4.493906106688712e-05,
|
792 |
+
"loss": 3.3594,
|
793 |
+
"step": 560
|
794 |
+
},
|
795 |
+
{
|
796 |
+
"epoch": 0.23,
|
797 |
+
"grad_norm": 29.176177978515625,
|
798 |
+
"learning_rate": 4.484196197037082e-05,
|
799 |
+
"loss": 3.7387,
|
800 |
+
"step": 565
|
801 |
+
},
|
802 |
+
{
|
803 |
+
"epoch": 0.23,
|
804 |
+
"grad_norm": 5.822291374206543,
|
805 |
+
"learning_rate": 4.474404724851356e-05,
|
806 |
+
"loss": 3.0015,
|
807 |
+
"step": 570
|
808 |
+
},
|
809 |
+
{
|
810 |
+
"epoch": 0.23,
|
811 |
+
"grad_norm": 13.846043586730957,
|
812 |
+
"learning_rate": 4.4645320926206064e-05,
|
813 |
+
"loss": 2.8736,
|
814 |
+
"step": 575
|
815 |
+
},
|
816 |
+
{
|
817 |
+
"epoch": 0.23,
|
818 |
+
"grad_norm": 6.161740303039551,
|
819 |
+
"learning_rate": 4.454578706170075e-05,
|
820 |
+
"loss": 3.1031,
|
821 |
+
"step": 580
|
822 |
+
},
|
823 |
+
{
|
824 |
+
"epoch": 0.23,
|
825 |
+
"grad_norm": 12.11698055267334,
|
826 |
+
"learning_rate": 4.444544974644493e-05,
|
827 |
+
"loss": 3.1832,
|
828 |
+
"step": 585
|
829 |
+
},
|
830 |
+
{
|
831 |
+
"epoch": 0.24,
|
832 |
+
"grad_norm": 5.144786834716797,
|
833 |
+
"learning_rate": 4.434431310491267e-05,
|
834 |
+
"loss": 2.7414,
|
835 |
+
"step": 590
|
836 |
+
},
|
837 |
+
{
|
838 |
+
"epoch": 0.24,
|
839 |
+
"grad_norm": 14.706486701965332,
|
840 |
+
"learning_rate": 4.4242381294435154e-05,
|
841 |
+
"loss": 2.7289,
|
842 |
+
"step": 595
|
843 |
+
},
|
844 |
+
{
|
845 |
+
"epoch": 0.24,
|
846 |
+
"grad_norm": 11.251163482666016,
|
847 |
+
"learning_rate": 4.413965850502987e-05,
|
848 |
+
"loss": 3.4787,
|
849 |
+
"step": 600
|
850 |
+
},
|
851 |
+
{
|
852 |
+
"epoch": 0.24,
|
853 |
+
"grad_norm": 11.005367279052734,
|
854 |
+
"learning_rate": 4.4036148959228365e-05,
|
855 |
+
"loss": 3.1987,
|
856 |
+
"step": 605
|
857 |
+
},
|
858 |
+
{
|
859 |
+
"epoch": 0.24,
|
860 |
+
"grad_norm": 13.336194038391113,
|
861 |
+
"learning_rate": 4.393185691190264e-05,
|
862 |
+
"loss": 2.9382,
|
863 |
+
"step": 610
|
864 |
+
},
|
865 |
+
{
|
866 |
+
"epoch": 0.25,
|
867 |
+
"grad_norm": 9.259475708007812,
|
868 |
+
"learning_rate": 4.382678665009028e-05,
|
869 |
+
"loss": 3.1349,
|
870 |
+
"step": 615
|
871 |
+
},
|
872 |
+
{
|
873 |
+
"epoch": 0.25,
|
874 |
+
"grad_norm": 20.902462005615234,
|
875 |
+
"learning_rate": 4.372094249281821e-05,
|
876 |
+
"loss": 3.1365,
|
877 |
+
"step": 620
|
878 |
+
},
|
879 |
+
{
|
880 |
+
"epoch": 0.25,
|
881 |
+
"grad_norm": 13.907161712646484,
|
882 |
+
"learning_rate": 4.3614328790925177e-05,
|
883 |
+
"loss": 3.3361,
|
884 |
+
"step": 625
|
885 |
+
},
|
886 |
+
{
|
887 |
+
"epoch": 0.25,
|
888 |
+
"grad_norm": 6.845438480377197,
|
889 |
+
"learning_rate": 4.350694992688289e-05,
|
890 |
+
"loss": 3.1903,
|
891 |
+
"step": 630
|
892 |
+
},
|
893 |
+
{
|
894 |
+
"epoch": 0.25,
|
895 |
+
"grad_norm": 14.879186630249023,
|
896 |
+
"learning_rate": 4.3398810314615876e-05,
|
897 |
+
"loss": 3.5133,
|
898 |
+
"step": 635
|
899 |
+
},
|
900 |
+
{
|
901 |
+
"epoch": 0.26,
|
902 |
+
"grad_norm": 10.454853057861328,
|
903 |
+
"learning_rate": 4.3289914399320034e-05,
|
904 |
+
"loss": 3.0928,
|
905 |
+
"step": 640
|
906 |
+
},
|
907 |
+
{
|
908 |
+
"epoch": 0.26,
|
909 |
+
"grad_norm": 10.692648887634277,
|
910 |
+
"learning_rate": 4.318026665727993e-05,
|
911 |
+
"loss": 3.1065,
|
912 |
+
"step": 645
|
913 |
+
},
|
914 |
+
{
|
915 |
+
"epoch": 0.26,
|
916 |
+
"grad_norm": 5.677508354187012,
|
917 |
+
"learning_rate": 4.306987159568479e-05,
|
918 |
+
"loss": 2.6484,
|
919 |
+
"step": 650
|
920 |
+
},
|
921 |
+
{
|
922 |
+
"epoch": 0.26,
|
923 |
+
"grad_norm": 7.058112621307373,
|
924 |
+
"learning_rate": 4.2958733752443195e-05,
|
925 |
+
"loss": 2.6531,
|
926 |
+
"step": 655
|
927 |
+
},
|
928 |
+
{
|
929 |
+
"epoch": 0.26,
|
930 |
+
"grad_norm": 7.029059886932373,
|
931 |
+
"learning_rate": 4.284685769599658e-05,
|
932 |
+
"loss": 3.3165,
|
933 |
+
"step": 660
|
934 |
+
},
|
935 |
+
{
|
936 |
+
"epoch": 0.27,
|
937 |
+
"grad_norm": 9.306894302368164,
|
938 |
+
"learning_rate": 4.273424802513145e-05,
|
939 |
+
"loss": 3.2707,
|
940 |
+
"step": 665
|
941 |
+
},
|
942 |
+
{
|
943 |
+
"epoch": 0.27,
|
944 |
+
"grad_norm": 6.9906392097473145,
|
945 |
+
"learning_rate": 4.262090936879029e-05,
|
946 |
+
"loss": 3.0822,
|
947 |
+
"step": 670
|
948 |
+
},
|
949 |
+
{
|
950 |
+
"epoch": 0.27,
|
951 |
+
"grad_norm": 10.101268768310547,
|
952 |
+
"learning_rate": 4.250684638588138e-05,
|
953 |
+
"loss": 3.3994,
|
954 |
+
"step": 675
|
955 |
+
},
|
956 |
+
{
|
957 |
+
"epoch": 0.27,
|
958 |
+
"grad_norm": 13.360941886901855,
|
959 |
+
"learning_rate": 4.239206376508717e-05,
|
960 |
+
"loss": 3.2358,
|
961 |
+
"step": 680
|
962 |
+
},
|
963 |
+
{
|
964 |
+
"epoch": 0.27,
|
965 |
+
"grad_norm": 25.73331642150879,
|
966 |
+
"learning_rate": 4.227656622467162e-05,
|
967 |
+
"loss": 3.1359,
|
968 |
+
"step": 685
|
969 |
+
},
|
970 |
+
{
|
971 |
+
"epoch": 0.28,
|
972 |
+
"grad_norm": 6.391608715057373,
|
973 |
+
"learning_rate": 4.216035851228626e-05,
|
974 |
+
"loss": 3.1343,
|
975 |
+
"step": 690
|
976 |
+
},
|
977 |
+
{
|
978 |
+
"epoch": 0.28,
|
979 |
+
"grad_norm": 21.201011657714844,
|
980 |
+
"learning_rate": 4.204344540477499e-05,
|
981 |
+
"loss": 3.0807,
|
982 |
+
"step": 695
|
983 |
+
},
|
984 |
+
{
|
985 |
+
"epoch": 0.28,
|
986 |
+
"grad_norm": 12.530668258666992,
|
987 |
+
"learning_rate": 4.192583170797774e-05,
|
988 |
+
"loss": 2.9729,
|
989 |
+
"step": 700
|
990 |
+
},
|
991 |
+
{
|
992 |
+
"epoch": 0.28,
|
993 |
+
"grad_norm": 17.209762573242188,
|
994 |
+
"learning_rate": 4.180752225653292e-05,
|
995 |
+
"loss": 2.9911,
|
996 |
+
"step": 705
|
997 |
+
},
|
998 |
+
{
|
999 |
+
"epoch": 0.28,
|
1000 |
+
"grad_norm": 16.309785842895508,
|
1001 |
+
"learning_rate": 4.16885219136787e-05,
|
1002 |
+
"loss": 3.0701,
|
1003 |
+
"step": 710
|
1004 |
+
},
|
1005 |
+
{
|
1006 |
+
"epoch": 0.29,
|
1007 |
+
"grad_norm": 7.244511127471924,
|
1008 |
+
"learning_rate": 4.1568835571053075e-05,
|
1009 |
+
"loss": 2.8169,
|
1010 |
+
"step": 715
|
1011 |
+
},
|
1012 |
+
{
|
1013 |
+
"epoch": 0.29,
|
1014 |
+
"grad_norm": 5.627272605895996,
|
1015 |
+
"learning_rate": 4.144846814849282e-05,
|
1016 |
+
"loss": 3.0518,
|
1017 |
+
"step": 720
|
1018 |
+
},
|
1019 |
+
{
|
1020 |
+
"epoch": 0.29,
|
1021 |
+
"grad_norm": 9.72758674621582,
|
1022 |
+
"learning_rate": 4.132742459383122e-05,
|
1023 |
+
"loss": 3.1135,
|
1024 |
+
"step": 725
|
1025 |
+
},
|
1026 |
+
{
|
1027 |
+
"epoch": 0.29,
|
1028 |
+
"grad_norm": 14.121967315673828,
|
1029 |
+
"learning_rate": 4.120570988269472e-05,
|
1030 |
+
"loss": 2.9481,
|
1031 |
+
"step": 730
|
1032 |
+
},
|
1033 |
+
{
|
1034 |
+
"epoch": 0.29,
|
1035 |
+
"grad_norm": 71.68965148925781,
|
1036 |
+
"learning_rate": 4.108332901829836e-05,
|
1037 |
+
"loss": 3.7509,
|
1038 |
+
"step": 735
|
1039 |
+
},
|
1040 |
+
{
|
1041 |
+
"epoch": 0.3,
|
1042 |
+
"grad_norm": 22.303184509277344,
|
1043 |
+
"learning_rate": 4.096028703124014e-05,
|
1044 |
+
"loss": 2.9864,
|
1045 |
+
"step": 740
|
1046 |
+
},
|
1047 |
+
{
|
1048 |
+
"epoch": 0.3,
|
1049 |
+
"grad_norm": 11.789464950561523,
|
1050 |
+
"learning_rate": 4.083658897929426e-05,
|
1051 |
+
"loss": 2.7573,
|
1052 |
+
"step": 745
|
1053 |
+
},
|
1054 |
+
{
|
1055 |
+
"epoch": 0.3,
|
1056 |
+
"grad_norm": 21.576683044433594,
|
1057 |
+
"learning_rate": 4.071223994720309e-05,
|
1058 |
+
"loss": 3.4365,
|
1059 |
+
"step": 750
|
1060 |
+
},
|
1061 |
+
{
|
1062 |
+
"epoch": 0.3,
|
1063 |
+
"grad_norm": 7.004244804382324,
|
1064 |
+
"learning_rate": 4.058724504646834e-05,
|
1065 |
+
"loss": 3.1543,
|
1066 |
+
"step": 755
|
1067 |
+
},
|
1068 |
+
{
|
1069 |
+
"epoch": 0.3,
|
1070 |
+
"grad_norm": 8.743468284606934,
|
1071 |
+
"learning_rate": 4.046160941514079e-05,
|
1072 |
+
"loss": 2.638,
|
1073 |
+
"step": 760
|
1074 |
+
},
|
1075 |
+
{
|
1076 |
+
"epoch": 0.31,
|
1077 |
+
"grad_norm": 9.981857299804688,
|
1078 |
+
"learning_rate": 4.033533821760917e-05,
|
1079 |
+
"loss": 2.4701,
|
1080 |
+
"step": 765
|
1081 |
+
},
|
1082 |
+
{
|
1083 |
+
"epoch": 0.31,
|
1084 |
+
"grad_norm": 11.454618453979492,
|
1085 |
+
"learning_rate": 4.0208436644387834e-05,
|
1086 |
+
"loss": 3.4152,
|
1087 |
+
"step": 770
|
1088 |
+
},
|
1089 |
+
{
|
1090 |
+
"epoch": 0.31,
|
1091 |
+
"grad_norm": 22.861881256103516,
|
1092 |
+
"learning_rate": 4.008090991190341e-05,
|
1093 |
+
"loss": 3.0526,
|
1094 |
+
"step": 775
|
1095 |
+
},
|
1096 |
+
{
|
1097 |
+
"epoch": 0.31,
|
1098 |
+
"grad_norm": 34.58321762084961,
|
1099 |
+
"learning_rate": 3.9952763262280405e-05,
|
1100 |
+
"loss": 3.1283,
|
1101 |
+
"step": 780
|
1102 |
+
},
|
1103 |
+
{
|
1104 |
+
"epoch": 0.31,
|
1105 |
+
"grad_norm": 12.163683891296387,
|
1106 |
+
"learning_rate": 3.982400196312564e-05,
|
1107 |
+
"loss": 2.8686,
|
1108 |
+
"step": 785
|
1109 |
+
},
|
1110 |
+
{
|
1111 |
+
"epoch": 0.32,
|
1112 |
+
"grad_norm": 34.78618240356445,
|
1113 |
+
"learning_rate": 3.969463130731183e-05,
|
1114 |
+
"loss": 2.782,
|
1115 |
+
"step": 790
|
1116 |
+
},
|
1117 |
+
{
|
1118 |
+
"epoch": 0.32,
|
1119 |
+
"grad_norm": 11.429190635681152,
|
1120 |
+
"learning_rate": 3.95646566127599e-05,
|
1121 |
+
"loss": 3.1518,
|
1122 |
+
"step": 795
|
1123 |
+
},
|
1124 |
+
{
|
1125 |
+
"epoch": 0.32,
|
1126 |
+
"grad_norm": 7.942728519439697,
|
1127 |
+
"learning_rate": 3.943408322222049e-05,
|
1128 |
+
"loss": 3.1183,
|
1129 |
+
"step": 800
|
1130 |
+
},
|
1131 |
+
{
|
1132 |
+
"epoch": 0.32,
|
1133 |
+
"grad_norm": 6.556036949157715,
|
1134 |
+
"learning_rate": 3.9302916503054246e-05,
|
1135 |
+
"loss": 2.9241,
|
1136 |
+
"step": 805
|
1137 |
+
},
|
1138 |
+
{
|
1139 |
+
"epoch": 0.32,
|
1140 |
+
"grad_norm": 7.161313533782959,
|
1141 |
+
"learning_rate": 3.917116184701125e-05,
|
1142 |
+
"loss": 2.8394,
|
1143 |
+
"step": 810
|
1144 |
+
},
|
1145 |
+
{
|
1146 |
+
"epoch": 0.33,
|
1147 |
+
"grad_norm": 17.556371688842773,
|
1148 |
+
"learning_rate": 3.903882467000937e-05,
|
1149 |
+
"loss": 2.667,
|
1150 |
+
"step": 815
|
1151 |
+
},
|
1152 |
+
{
|
1153 |
+
"epoch": 0.33,
|
1154 |
+
"grad_norm": 6.0064167976379395,
|
1155 |
+
"learning_rate": 3.8905910411911625e-05,
|
1156 |
+
"loss": 2.8605,
|
1157 |
+
"step": 820
|
1158 |
+
},
|
1159 |
+
{
|
1160 |
+
"epoch": 0.33,
|
1161 |
+
"grad_norm": 13.329277038574219,
|
1162 |
+
"learning_rate": 3.8772424536302564e-05,
|
1163 |
+
"loss": 2.8765,
|
1164 |
+
"step": 825
|
1165 |
+
},
|
1166 |
+
{
|
1167 |
+
"epoch": 0.33,
|
1168 |
+
"grad_norm": 7.7756876945495605,
|
1169 |
+
"learning_rate": 3.8638372530263715e-05,
|
1170 |
+
"loss": 2.6558,
|
1171 |
+
"step": 830
|
1172 |
+
},
|
1173 |
+
{
|
1174 |
+
"epoch": 0.33,
|
1175 |
+
"grad_norm": 6.505970478057861,
|
1176 |
+
"learning_rate": 3.850375990414801e-05,
|
1177 |
+
"loss": 2.6178,
|
1178 |
+
"step": 835
|
1179 |
+
},
|
1180 |
+
{
|
1181 |
+
"epoch": 0.34,
|
1182 |
+
"grad_norm": 29.54509925842285,
|
1183 |
+
"learning_rate": 3.836859219135324e-05,
|
1184 |
+
"loss": 2.9655,
|
1185 |
+
"step": 840
|
1186 |
+
},
|
1187 |
+
{
|
1188 |
+
"epoch": 0.34,
|
1189 |
+
"grad_norm": 12.92197322845459,
|
1190 |
+
"learning_rate": 3.823287494809469e-05,
|
1191 |
+
"loss": 2.7825,
|
1192 |
+
"step": 845
|
1193 |
+
},
|
1194 |
+
{
|
1195 |
+
"epoch": 0.34,
|
1196 |
+
"grad_norm": 10.073705673217773,
|
1197 |
+
"learning_rate": 3.8096613753176634e-05,
|
1198 |
+
"loss": 3.3055,
|
1199 |
+
"step": 850
|
1200 |
+
},
|
1201 |
+
{
|
1202 |
+
"epoch": 0.34,
|
1203 |
+
"grad_norm": 23.254878997802734,
|
1204 |
+
"learning_rate": 3.7959814207763135e-05,
|
1205 |
+
"loss": 2.906,
|
1206 |
+
"step": 855
|
1207 |
+
},
|
1208 |
+
{
|
1209 |
+
"epoch": 0.34,
|
1210 |
+
"grad_norm": 16.555452346801758,
|
1211 |
+
"learning_rate": 3.782248193514766e-05,
|
1212 |
+
"loss": 3.2928,
|
1213 |
+
"step": 860
|
1214 |
+
},
|
1215 |
+
{
|
1216 |
+
"epoch": 0.35,
|
1217 |
+
"grad_norm": 8.429189682006836,
|
1218 |
+
"learning_rate": 3.7684622580522055e-05,
|
1219 |
+
"loss": 2.6259,
|
1220 |
+
"step": 865
|
1221 |
+
},
|
1222 |
+
{
|
1223 |
+
"epoch": 0.35,
|
1224 |
+
"grad_norm": 9.762368202209473,
|
1225 |
+
"learning_rate": 3.7546241810744445e-05,
|
1226 |
+
"loss": 2.7576,
|
1227 |
+
"step": 870
|
1228 |
+
},
|
1229 |
+
{
|
1230 |
+
"epoch": 0.35,
|
1231 |
+
"grad_norm": 11.004511833190918,
|
1232 |
+
"learning_rate": 3.740734531410626e-05,
|
1233 |
+
"loss": 2.6714,
|
1234 |
+
"step": 875
|
1235 |
+
},
|
1236 |
+
{
|
1237 |
+
"epoch": 0.35,
|
1238 |
+
"grad_norm": 56.73052978515625,
|
1239 |
+
"learning_rate": 3.726793880009845e-05,
|
1240 |
+
"loss": 3.542,
|
1241 |
+
"step": 880
|
1242 |
+
},
|
1243 |
+
{
|
1244 |
+
"epoch": 0.35,
|
1245 |
+
"grad_norm": 10.330306053161621,
|
1246 |
+
"learning_rate": 3.7128027999176803e-05,
|
1247 |
+
"loss": 3.0011,
|
1248 |
+
"step": 885
|
1249 |
+
},
|
1250 |
+
{
|
1251 |
+
"epoch": 0.36,
|
1252 |
+
"grad_norm": 12.130693435668945,
|
1253 |
+
"learning_rate": 3.698761866252635e-05,
|
1254 |
+
"loss": 3.0935,
|
1255 |
+
"step": 890
|
1256 |
+
},
|
1257 |
+
{
|
1258 |
+
"epoch": 0.36,
|
1259 |
+
"grad_norm": 12.528786659240723,
|
1260 |
+
"learning_rate": 3.6846716561824965e-05,
|
1261 |
+
"loss": 3.3687,
|
1262 |
+
"step": 895
|
1263 |
+
},
|
1264 |
+
{
|
1265 |
+
"epoch": 0.36,
|
1266 |
+
"grad_norm": 14.122909545898438,
|
1267 |
+
"learning_rate": 3.670532748900615e-05,
|
1268 |
+
"loss": 2.5591,
|
1269 |
+
"step": 900
|
1270 |
+
},
|
1271 |
+
{
|
1272 |
+
"epoch": 0.36,
|
1273 |
+
"grad_norm": 8.046327590942383,
|
1274 |
+
"learning_rate": 3.656345725602089e-05,
|
1275 |
+
"loss": 2.3845,
|
1276 |
+
"step": 905
|
1277 |
+
},
|
1278 |
+
{
|
1279 |
+
"epoch": 0.36,
|
1280 |
+
"grad_norm": 22.46470832824707,
|
1281 |
+
"learning_rate": 3.642111169459879e-05,
|
1282 |
+
"loss": 2.8321,
|
1283 |
+
"step": 910
|
1284 |
+
},
|
1285 |
+
{
|
1286 |
+
"epoch": 0.37,
|
1287 |
+
"grad_norm": 10.554203987121582,
|
1288 |
+
"learning_rate": 3.6278296656008366e-05,
|
1289 |
+
"loss": 2.4067,
|
1290 |
+
"step": 915
|
1291 |
+
},
|
1292 |
+
{
|
1293 |
+
"epoch": 0.37,
|
1294 |
+
"grad_norm": 8.757329940795898,
|
1295 |
+
"learning_rate": 3.6135018010816477e-05,
|
1296 |
+
"loss": 2.7003,
|
1297 |
+
"step": 920
|
1298 |
+
},
|
1299 |
+
{
|
1300 |
+
"epoch": 0.37,
|
1301 |
+
"grad_norm": 8.956037521362305,
|
1302 |
+
"learning_rate": 3.599128164864706e-05,
|
1303 |
+
"loss": 2.7975,
|
1304 |
+
"step": 925
|
1305 |
+
},
|
1306 |
+
{
|
1307 |
+
"epoch": 0.37,
|
1308 |
+
"grad_norm": 15.920394897460938,
|
1309 |
+
"learning_rate": 3.5847093477938956e-05,
|
1310 |
+
"loss": 2.7154,
|
1311 |
+
"step": 930
|
1312 |
+
},
|
1313 |
+
{
|
1314 |
+
"epoch": 0.37,
|
1315 |
+
"grad_norm": 15.14090633392334,
|
1316 |
+
"learning_rate": 3.570245942570315e-05,
|
1317 |
+
"loss": 3.3547,
|
1318 |
+
"step": 935
|
1319 |
+
},
|
1320 |
+
{
|
1321 |
+
"epoch": 0.38,
|
1322 |
+
"grad_norm": 17.72629165649414,
|
1323 |
+
"learning_rate": 3.5557385437279e-05,
|
1324 |
+
"loss": 2.7852,
|
1325 |
+
"step": 940
|
1326 |
+
},
|
1327 |
+
{
|
1328 |
+
"epoch": 0.38,
|
1329 |
+
"grad_norm": 5.857475757598877,
|
1330 |
+
"learning_rate": 3.5411877476089975e-05,
|
1331 |
+
"loss": 2.73,
|
1332 |
+
"step": 945
|
1333 |
+
},
|
1334 |
+
{
|
1335 |
+
"epoch": 0.38,
|
1336 |
+
"grad_norm": 22.337295532226562,
|
1337 |
+
"learning_rate": 3.526594152339845e-05,
|
1338 |
+
"loss": 3.1076,
|
1339 |
+
"step": 950
|
1340 |
+
},
|
1341 |
+
{
|
1342 |
+
"epoch": 0.38,
|
1343 |
+
"grad_norm": 10.523070335388184,
|
1344 |
+
"learning_rate": 3.5119583578059846e-05,
|
1345 |
+
"loss": 3.1857,
|
1346 |
+
"step": 955
|
1347 |
+
},
|
1348 |
+
{
|
1349 |
+
"epoch": 0.38,
|
1350 |
+
"grad_norm": 11.635588645935059,
|
1351 |
+
"learning_rate": 3.497280965627605e-05,
|
1352 |
+
"loss": 2.8584,
|
1353 |
+
"step": 960
|
1354 |
+
},
|
1355 |
+
{
|
1356 |
+
"epoch": 0.39,
|
1357 |
+
"grad_norm": 27.29017448425293,
|
1358 |
+
"learning_rate": 3.4825625791348096e-05,
|
1359 |
+
"loss": 3.3567,
|
1360 |
+
"step": 965
|
1361 |
+
},
|
1362 |
+
{
|
1363 |
+
"epoch": 0.39,
|
1364 |
+
"grad_norm": 25.24538230895996,
|
1365 |
+
"learning_rate": 3.467803803342821e-05,
|
1366 |
+
"loss": 3.1713,
|
1367 |
+
"step": 970
|
1368 |
+
},
|
1369 |
+
{
|
1370 |
+
"epoch": 0.39,
|
1371 |
+
"grad_norm": 11.2062406539917,
|
1372 |
+
"learning_rate": 3.4530052449271044e-05,
|
1373 |
+
"loss": 2.8941,
|
1374 |
+
"step": 975
|
1375 |
+
},
|
1376 |
+
{
|
1377 |
+
"epoch": 0.39,
|
1378 |
+
"grad_norm": 20.25462532043457,
|
1379 |
+
"learning_rate": 3.438167512198436e-05,
|
1380 |
+
"loss": 3.1291,
|
1381 |
+
"step": 980
|
1382 |
+
},
|
1383 |
+
{
|
1384 |
+
"epoch": 0.39,
|
1385 |
+
"grad_norm": 13.0928373336792,
|
1386 |
+
"learning_rate": 3.4232912150778914e-05,
|
1387 |
+
"loss": 2.4672,
|
1388 |
+
"step": 985
|
1389 |
+
},
|
1390 |
+
{
|
1391 |
+
"epoch": 0.4,
|
1392 |
+
"grad_norm": 21.326560974121094,
|
1393 |
+
"learning_rate": 3.408376965071779e-05,
|
1394 |
+
"loss": 3.0489,
|
1395 |
+
"step": 990
|
1396 |
+
},
|
1397 |
+
{
|
1398 |
+
"epoch": 0.4,
|
1399 |
+
"grad_norm": 54.281578063964844,
|
1400 |
+
"learning_rate": 3.393425375246503e-05,
|
1401 |
+
"loss": 3.0251,
|
1402 |
+
"step": 995
|
1403 |
+
},
|
1404 |
+
{
|
1405 |
+
"epoch": 0.4,
|
1406 |
+
"grad_norm": 18.1798038482666,
|
1407 |
+
"learning_rate": 3.378437060203357e-05,
|
1408 |
+
"loss": 3.0152,
|
1409 |
+
"step": 1000
|
1410 |
+
},
|
1411 |
+
{
|
1412 |
+
"epoch": 0.4,
|
1413 |
+
"grad_norm": 8.757742881774902,
|
1414 |
+
"learning_rate": 3.363412636053269e-05,
|
1415 |
+
"loss": 2.5223,
|
1416 |
+
"step": 1005
|
1417 |
+
},
|
1418 |
+
{
|
1419 |
+
"epoch": 0.4,
|
1420 |
+
"grad_norm": 7.601003170013428,
|
1421 |
+
"learning_rate": 3.348352720391469e-05,
|
1422 |
+
"loss": 2.8158,
|
1423 |
+
"step": 1010
|
1424 |
+
},
|
1425 |
+
{
|
1426 |
+
"epoch": 0.41,
|
1427 |
+
"grad_norm": 16.87445068359375,
|
1428 |
+
"learning_rate": 3.3332579322721046e-05,
|
1429 |
+
"loss": 3.3828,
|
1430 |
+
"step": 1015
|
1431 |
+
},
|
1432 |
+
{
|
1433 |
+
"epoch": 0.41,
|
1434 |
+
"grad_norm": 11.726461410522461,
|
1435 |
+
"learning_rate": 3.318128892182792e-05,
|
1436 |
+
"loss": 2.5411,
|
1437 |
+
"step": 1020
|
1438 |
+
},
|
1439 |
+
{
|
1440 |
+
"epoch": 0.41,
|
1441 |
+
"grad_norm": 11.669478416442871,
|
1442 |
+
"learning_rate": 3.3029662220191144e-05,
|
1443 |
+
"loss": 2.6686,
|
1444 |
+
"step": 1025
|
1445 |
+
},
|
1446 |
+
{
|
1447 |
+
"epoch": 0.41,
|
1448 |
+
"grad_norm": 58.972007751464844,
|
1449 |
+
"learning_rate": 3.2877705450590526e-05,
|
1450 |
+
"loss": 3.3026,
|
1451 |
+
"step": 1030
|
1452 |
+
},
|
1453 |
+
{
|
1454 |
+
"epoch": 0.41,
|
1455 |
+
"grad_norm": 7.434502124786377,
|
1456 |
+
"learning_rate": 3.272542485937369e-05,
|
1457 |
+
"loss": 2.6652,
|
1458 |
+
"step": 1035
|
1459 |
+
},
|
1460 |
+
{
|
1461 |
+
"epoch": 0.42,
|
1462 |
+
"grad_norm": 22.020599365234375,
|
1463 |
+
"learning_rate": 3.2572826706199305e-05,
|
1464 |
+
"loss": 3.1553,
|
1465 |
+
"step": 1040
|
1466 |
+
},
|
1467 |
+
{
|
1468 |
+
"epoch": 0.42,
|
1469 |
+
"grad_norm": 17.60317611694336,
|
1470 |
+
"learning_rate": 3.2419917263779766e-05,
|
1471 |
+
"loss": 3.2016,
|
1472 |
+
"step": 1045
|
1473 |
+
},
|
1474 |
+
{
|
1475 |
+
"epoch": 0.42,
|
1476 |
+
"grad_norm": 15.09745979309082,
|
1477 |
+
"learning_rate": 3.2266702817623346e-05,
|
1478 |
+
"loss": 2.6242,
|
1479 |
+
"step": 1050
|
1480 |
+
},
|
1481 |
+
{
|
1482 |
+
"epoch": 0.42,
|
1483 |
+
"grad_norm": 11.325228691101074,
|
1484 |
+
"learning_rate": 3.211318966577581e-05,
|
1485 |
+
"loss": 2.9649,
|
1486 |
+
"step": 1055
|
1487 |
+
},
|
1488 |
+
{
|
1489 |
+
"epoch": 0.42,
|
1490 |
+
"grad_norm": 16.898090362548828,
|
1491 |
+
"learning_rate": 3.195938411856159e-05,
|
1492 |
+
"loss": 2.4623,
|
1493 |
+
"step": 1060
|
1494 |
+
},
|
1495 |
+
{
|
1496 |
+
"epoch": 0.43,
|
1497 |
+
"grad_norm": 8.761795043945312,
|
1498 |
+
"learning_rate": 3.180529249832428e-05,
|
1499 |
+
"loss": 2.4152,
|
1500 |
+
"step": 1065
|
1501 |
+
},
|
1502 |
+
{
|
1503 |
+
"epoch": 0.43,
|
1504 |
+
"grad_norm": 11.047560691833496,
|
1505 |
+
"learning_rate": 3.165092113916688e-05,
|
1506 |
+
"loss": 3.0119,
|
1507 |
+
"step": 1070
|
1508 |
+
},
|
1509 |
+
{
|
1510 |
+
"epoch": 0.43,
|
1511 |
+
"grad_norm": 11.61601734161377,
|
1512 |
+
"learning_rate": 3.149627638669132e-05,
|
1513 |
+
"loss": 2.4781,
|
1514 |
+
"step": 1075
|
1515 |
+
},
|
1516 |
+
{
|
1517 |
+
"epoch": 0.43,
|
1518 |
+
"grad_norm": 21.56399154663086,
|
1519 |
+
"learning_rate": 3.1341364597737686e-05,
|
1520 |
+
"loss": 3.4005,
|
1521 |
+
"step": 1080
|
1522 |
+
},
|
1523 |
+
{
|
1524 |
+
"epoch": 0.43,
|
1525 |
+
"grad_norm": 15.666382789611816,
|
1526 |
+
"learning_rate": 3.118619214012286e-05,
|
1527 |
+
"loss": 2.9385,
|
1528 |
+
"step": 1085
|
1529 |
+
},
|
1530 |
+
{
|
1531 |
+
"epoch": 0.44,
|
1532 |
+
"grad_norm": 11.866064071655273,
|
1533 |
+
"learning_rate": 3.1030765392378816e-05,
|
1534 |
+
"loss": 3.0096,
|
1535 |
+
"step": 1090
|
1536 |
+
},
|
1537 |
+
{
|
1538 |
+
"epoch": 0.44,
|
1539 |
+
"grad_norm": 11.25349235534668,
|
1540 |
+
"learning_rate": 3.0875090743490384e-05,
|
1541 |
+
"loss": 2.0702,
|
1542 |
+
"step": 1095
|
1543 |
+
},
|
1544 |
+
{
|
1545 |
+
"epoch": 0.44,
|
1546 |
+
"grad_norm": 10.070486068725586,
|
1547 |
+
"learning_rate": 3.071917459263264e-05,
|
1548 |
+
"loss": 3.0652,
|
1549 |
+
"step": 1100
|
1550 |
+
},
|
1551 |
+
{
|
1552 |
+
"epoch": 0.44,
|
1553 |
+
"grad_norm": 7.047021389007568,
|
1554 |
+
"learning_rate": 3.056302334890786e-05,
|
1555 |
+
"loss": 2.8617,
|
1556 |
+
"step": 1105
|
1557 |
+
},
|
1558 |
+
{
|
1559 |
+
"epoch": 0.44,
|
1560 |
+
"grad_norm": 11.59208869934082,
|
1561 |
+
"learning_rate": 3.040664343108209e-05,
|
1562 |
+
"loss": 2.8697,
|
1563 |
+
"step": 1110
|
1564 |
+
},
|
1565 |
+
{
|
1566 |
+
"epoch": 0.45,
|
1567 |
+
"grad_norm": 9.361946105957031,
|
1568 |
+
"learning_rate": 3.0250041267321232e-05,
|
1569 |
+
"loss": 2.7646,
|
1570 |
+
"step": 1115
|
1571 |
+
},
|
1572 |
+
{
|
1573 |
+
"epoch": 0.45,
|
1574 |
+
"grad_norm": 11.60400390625,
|
1575 |
+
"learning_rate": 3.0093223294926892e-05,
|
1576 |
+
"loss": 2.8238,
|
1577 |
+
"step": 1120
|
1578 |
+
},
|
1579 |
+
{
|
1580 |
+
"epoch": 0.45,
|
1581 |
+
"grad_norm": 19.50318145751953,
|
1582 |
+
"learning_rate": 2.993619596007168e-05,
|
1583 |
+
"loss": 2.8857,
|
1584 |
+
"step": 1125
|
1585 |
+
},
|
1586 |
+
{
|
1587 |
+
"epoch": 0.45,
|
1588 |
+
"grad_norm": 24.245025634765625,
|
1589 |
+
"learning_rate": 2.9778965717534313e-05,
|
1590 |
+
"loss": 2.949,
|
1591 |
+
"step": 1130
|
1592 |
+
},
|
1593 |
+
{
|
1594 |
+
"epoch": 0.45,
|
1595 |
+
"grad_norm": 14.323025703430176,
|
1596 |
+
"learning_rate": 2.962153903043422e-05,
|
1597 |
+
"loss": 2.6237,
|
1598 |
+
"step": 1135
|
1599 |
+
},
|
1600 |
+
{
|
1601 |
+
"epoch": 0.46,
|
1602 |
+
"grad_norm": 13.869138717651367,
|
1603 |
+
"learning_rate": 2.9463922369965917e-05,
|
1604 |
+
"loss": 2.5021,
|
1605 |
+
"step": 1140
|
1606 |
+
},
|
1607 |
+
{
|
1608 |
+
"epoch": 0.46,
|
1609 |
+
"grad_norm": 9.667675018310547,
|
1610 |
+
"learning_rate": 2.9306122215132976e-05,
|
1611 |
+
"loss": 2.9346,
|
1612 |
+
"step": 1145
|
1613 |
+
},
|
1614 |
+
{
|
1615 |
+
"epoch": 0.46,
|
1616 |
+
"grad_norm": 22.312841415405273,
|
1617 |
+
"learning_rate": 2.91481450524817e-05,
|
1618 |
+
"loss": 2.7172,
|
1619 |
+
"step": 1150
|
1620 |
+
},
|
1621 |
+
{
|
1622 |
+
"epoch": 0.46,
|
1623 |
+
"grad_norm": 10.218385696411133,
|
1624 |
+
"learning_rate": 2.8989997375834482e-05,
|
1625 |
+
"loss": 2.4864,
|
1626 |
+
"step": 1155
|
1627 |
+
},
|
1628 |
+
{
|
1629 |
+
"epoch": 0.46,
|
1630 |
+
"grad_norm": 7.388011455535889,
|
1631 |
+
"learning_rate": 2.8831685686022897e-05,
|
1632 |
+
"loss": 2.7519,
|
1633 |
+
"step": 1160
|
1634 |
+
},
|
1635 |
+
{
|
1636 |
+
"epoch": 0.47,
|
1637 |
+
"grad_norm": 8.172524452209473,
|
1638 |
+
"learning_rate": 2.8673216490620452e-05,
|
1639 |
+
"loss": 2.701,
|
1640 |
+
"step": 1165
|
1641 |
+
},
|
1642 |
+
{
|
1643 |
+
"epoch": 0.47,
|
1644 |
+
"grad_norm": 11.60141372680664,
|
1645 |
+
"learning_rate": 2.8514596303675073e-05,
|
1646 |
+
"loss": 3.1608,
|
1647 |
+
"step": 1170
|
1648 |
+
},
|
1649 |
+
{
|
1650 |
+
"epoch": 0.47,
|
1651 |
+
"grad_norm": 21.537139892578125,
|
1652 |
+
"learning_rate": 2.8355831645441388e-05,
|
1653 |
+
"loss": 2.9307,
|
1654 |
+
"step": 1175
|
1655 |
+
},
|
1656 |
+
{
|
1657 |
+
"epoch": 0.47,
|
1658 |
+
"grad_norm": 12.06334400177002,
|
1659 |
+
"learning_rate": 2.8196929042112652e-05,
|
1660 |
+
"loss": 2.9414,
|
1661 |
+
"step": 1180
|
1662 |
+
},
|
1663 |
+
{
|
1664 |
+
"epoch": 0.47,
|
1665 |
+
"grad_norm": 8.39703369140625,
|
1666 |
+
"learning_rate": 2.8037895025552512e-05,
|
1667 |
+
"loss": 2.938,
|
1668 |
+
"step": 1185
|
1669 |
+
},
|
1670 |
+
{
|
1671 |
+
"epoch": 0.48,
|
1672 |
+
"grad_norm": 8.78774642944336,
|
1673 |
+
"learning_rate": 2.787873613302649e-05,
|
1674 |
+
"loss": 2.5751,
|
1675 |
+
"step": 1190
|
1676 |
+
},
|
1677 |
+
{
|
1678 |
+
"epoch": 0.48,
|
1679 |
+
"grad_norm": 24.190916061401367,
|
1680 |
+
"learning_rate": 2.7719458906933277e-05,
|
1681 |
+
"loss": 2.328,
|
1682 |
+
"step": 1195
|
1683 |
+
},
|
1684 |
+
{
|
1685 |
+
"epoch": 0.48,
|
1686 |
+
"grad_norm": 17.80084228515625,
|
1687 |
+
"learning_rate": 2.7560069894535784e-05,
|
1688 |
+
"loss": 2.7258,
|
1689 |
+
"step": 1200
|
1690 |
+
},
|
1691 |
+
{
|
1692 |
+
"epoch": 0.48,
|
1693 |
+
"grad_norm": 17.03345489501953,
|
1694 |
+
"learning_rate": 2.7400575647692046e-05,
|
1695 |
+
"loss": 2.8104,
|
1696 |
+
"step": 1205
|
1697 |
+
},
|
1698 |
+
{
|
1699 |
+
"epoch": 0.48,
|
1700 |
+
"grad_norm": 36.040897369384766,
|
1701 |
+
"learning_rate": 2.724098272258584e-05,
|
1702 |
+
"loss": 2.9138,
|
1703 |
+
"step": 1210
|
1704 |
+
},
|
1705 |
+
{
|
1706 |
+
"epoch": 0.49,
|
1707 |
+
"grad_norm": 9.431939125061035,
|
1708 |
+
"learning_rate": 2.7081297679457236e-05,
|
1709 |
+
"loss": 2.3052,
|
1710 |
+
"step": 1215
|
1711 |
+
},
|
1712 |
+
{
|
1713 |
+
"epoch": 0.49,
|
1714 |
+
"grad_norm": 10.720209121704102,
|
1715 |
+
"learning_rate": 2.692152708233292e-05,
|
1716 |
+
"loss": 2.3984,
|
1717 |
+
"step": 1220
|
1718 |
+
},
|
1719 |
+
{
|
1720 |
+
"epoch": 0.49,
|
1721 |
+
"grad_norm": 13.637066841125488,
|
1722 |
+
"learning_rate": 2.676167749875635e-05,
|
1723 |
+
"loss": 2.8592,
|
1724 |
+
"step": 1225
|
1725 |
+
},
|
1726 |
+
{
|
1727 |
+
"epoch": 0.49,
|
1728 |
+
"grad_norm": 8.968180656433105,
|
1729 |
+
"learning_rate": 2.6601755499517826e-05,
|
1730 |
+
"loss": 2.4064,
|
1731 |
+
"step": 1230
|
1732 |
+
},
|
1733 |
+
{
|
1734 |
+
"epoch": 0.49,
|
1735 |
+
"grad_norm": 9.793657302856445,
|
1736 |
+
"learning_rate": 2.6441767658384366e-05,
|
1737 |
+
"loss": 3.0291,
|
1738 |
+
"step": 1235
|
1739 |
+
},
|
1740 |
+
{
|
1741 |
+
"epoch": 0.5,
|
1742 |
+
"grad_norm": 8.77736759185791,
|
1743 |
+
"learning_rate": 2.628172055182948e-05,
|
1744 |
+
"loss": 2.3978,
|
1745 |
+
"step": 1240
|
1746 |
+
},
|
1747 |
+
{
|
1748 |
+
"epoch": 0.5,
|
1749 |
+
"grad_norm": 9.85341739654541,
|
1750 |
+
"learning_rate": 2.6121620758762877e-05,
|
1751 |
+
"loss": 2.5431,
|
1752 |
+
"step": 1245
|
1753 |
+
},
|
1754 |
+
{
|
1755 |
+
"epoch": 0.5,
|
1756 |
+
"grad_norm": 11.554092407226562,
|
1757 |
+
"learning_rate": 2.596147486025996e-05,
|
1758 |
+
"loss": 2.8736,
|
1759 |
+
"step": 1250
|
1760 |
+
},
|
1761 |
+
{
|
1762 |
+
"epoch": 0.5,
|
1763 |
+
"grad_norm": 8.345808029174805,
|
1764 |
+
"learning_rate": 2.5801289439291388e-05,
|
1765 |
+
"loss": 2.6837,
|
1766 |
+
"step": 1255
|
1767 |
+
},
|
1768 |
+
{
|
1769 |
+
"epoch": 0.5,
|
1770 |
+
"grad_norm": 9.449942588806152,
|
1771 |
+
"learning_rate": 2.564107108045239e-05,
|
1772 |
+
"loss": 3.1686,
|
1773 |
+
"step": 1260
|
1774 |
+
},
|
1775 |
+
{
|
1776 |
+
"epoch": 0.51,
|
1777 |
+
"grad_norm": 32.239585876464844,
|
1778 |
+
"learning_rate": 2.5480826369692178e-05,
|
1779 |
+
"loss": 3.0723,
|
1780 |
+
"step": 1265
|
1781 |
+
},
|
1782 |
+
{
|
1783 |
+
"epoch": 0.51,
|
1784 |
+
"grad_norm": 36.09556579589844,
|
1785 |
+
"learning_rate": 2.5320561894043184e-05,
|
1786 |
+
"loss": 2.6983,
|
1787 |
+
"step": 1270
|
1788 |
+
},
|
1789 |
+
{
|
1790 |
+
"epoch": 0.51,
|
1791 |
+
"grad_norm": 42.91587829589844,
|
1792 |
+
"learning_rate": 2.5160284241350278e-05,
|
1793 |
+
"loss": 2.6336,
|
1794 |
+
"step": 1275
|
1795 |
+
},
|
1796 |
+
{
|
1797 |
+
"epoch": 0.51,
|
1798 |
+
"grad_norm": 20.82798194885254,
|
1799 |
+
"learning_rate": 2.5e-05,
|
1800 |
+
"loss": 2.8635,
|
1801 |
+
"step": 1280
|
1802 |
+
},
|
1803 |
+
{
|
1804 |
+
"epoch": 0.51,
|
1805 |
+
"grad_norm": 7.868647575378418,
|
1806 |
+
"learning_rate": 2.4839715758649724e-05,
|
1807 |
+
"loss": 2.8232,
|
1808 |
+
"step": 1285
|
1809 |
+
},
|
1810 |
+
{
|
1811 |
+
"epoch": 0.52,
|
1812 |
+
"grad_norm": 12.662830352783203,
|
1813 |
+
"learning_rate": 2.467943810595682e-05,
|
1814 |
+
"loss": 2.8228,
|
1815 |
+
"step": 1290
|
1816 |
+
},
|
1817 |
+
{
|
1818 |
+
"epoch": 0.52,
|
1819 |
+
"grad_norm": 15.0082368850708,
|
1820 |
+
"learning_rate": 2.4519173630307825e-05,
|
1821 |
+
"loss": 2.6371,
|
1822 |
+
"step": 1295
|
1823 |
+
},
|
1824 |
+
{
|
1825 |
+
"epoch": 0.52,
|
1826 |
+
"grad_norm": 15.185354232788086,
|
1827 |
+
"learning_rate": 2.4358928919547616e-05,
|
1828 |
+
"loss": 2.6336,
|
1829 |
+
"step": 1300
|
1830 |
+
},
|
1831 |
+
{
|
1832 |
+
"epoch": 0.52,
|
1833 |
+
"grad_norm": 7.559114456176758,
|
1834 |
+
"learning_rate": 2.419871056070862e-05,
|
1835 |
+
"loss": 2.4209,
|
1836 |
+
"step": 1305
|
1837 |
+
},
|
1838 |
+
{
|
1839 |
+
"epoch": 0.52,
|
1840 |
+
"grad_norm": 13.428425788879395,
|
1841 |
+
"learning_rate": 2.403852513974004e-05,
|
1842 |
+
"loss": 2.6897,
|
1843 |
+
"step": 1310
|
1844 |
+
},
|
1845 |
+
{
|
1846 |
+
"epoch": 0.53,
|
1847 |
+
"grad_norm": 38.91737747192383,
|
1848 |
+
"learning_rate": 2.3878379241237136e-05,
|
1849 |
+
"loss": 2.569,
|
1850 |
+
"step": 1315
|
1851 |
+
},
|
1852 |
+
{
|
1853 |
+
"epoch": 0.53,
|
1854 |
+
"grad_norm": 15.211456298828125,
|
1855 |
+
"learning_rate": 2.3718279448170525e-05,
|
1856 |
+
"loss": 2.4856,
|
1857 |
+
"step": 1320
|
1858 |
+
},
|
1859 |
+
{
|
1860 |
+
"epoch": 0.53,
|
1861 |
+
"grad_norm": 11.391417503356934,
|
1862 |
+
"learning_rate": 2.3558232341615643e-05,
|
1863 |
+
"loss": 2.5527,
|
1864 |
+
"step": 1325
|
1865 |
+
},
|
1866 |
+
{
|
1867 |
+
"epoch": 0.53,
|
1868 |
+
"grad_norm": 21.21587371826172,
|
1869 |
+
"learning_rate": 2.339824450048218e-05,
|
1870 |
+
"loss": 3.5899,
|
1871 |
+
"step": 1330
|
1872 |
+
},
|
1873 |
+
{
|
1874 |
+
"epoch": 0.53,
|
1875 |
+
"grad_norm": 17.66887664794922,
|
1876 |
+
"learning_rate": 2.323832250124365e-05,
|
1877 |
+
"loss": 2.4669,
|
1878 |
+
"step": 1335
|
1879 |
+
},
|
1880 |
+
{
|
1881 |
+
"epoch": 0.54,
|
1882 |
+
"grad_norm": 10.46285629272461,
|
1883 |
+
"learning_rate": 2.3078472917667092e-05,
|
1884 |
+
"loss": 2.4984,
|
1885 |
+
"step": 1340
|
1886 |
+
},
|
1887 |
+
{
|
1888 |
+
"epoch": 0.54,
|
1889 |
+
"grad_norm": 14.3226318359375,
|
1890 |
+
"learning_rate": 2.291870232054277e-05,
|
1891 |
+
"loss": 2.8999,
|
1892 |
+
"step": 1345
|
1893 |
+
},
|
1894 |
+
{
|
1895 |
+
"epoch": 0.54,
|
1896 |
+
"grad_norm": 47.95698547363281,
|
1897 |
+
"learning_rate": 2.2759017277414166e-05,
|
1898 |
+
"loss": 2.7827,
|
1899 |
+
"step": 1350
|
1900 |
+
},
|
1901 |
+
{
|
1902 |
+
"epoch": 0.54,
|
1903 |
+
"grad_norm": 10.691919326782227,
|
1904 |
+
"learning_rate": 2.2599424352307957e-05,
|
1905 |
+
"loss": 2.9201,
|
1906 |
+
"step": 1355
|
1907 |
+
},
|
1908 |
+
{
|
1909 |
+
"epoch": 0.54,
|
1910 |
+
"grad_norm": 15.748466491699219,
|
1911 |
+
"learning_rate": 2.243993010546422e-05,
|
1912 |
+
"loss": 2.3433,
|
1913 |
+
"step": 1360
|
1914 |
+
},
|
1915 |
+
{
|
1916 |
+
"epoch": 0.55,
|
1917 |
+
"grad_norm": 18.585126876831055,
|
1918 |
+
"learning_rate": 2.2280541093066732e-05,
|
1919 |
+
"loss": 2.7896,
|
1920 |
+
"step": 1365
|
1921 |
+
},
|
1922 |
+
{
|
1923 |
+
"epoch": 0.55,
|
1924 |
+
"grad_norm": 9.304043769836426,
|
1925 |
+
"learning_rate": 2.212126386697352e-05,
|
1926 |
+
"loss": 2.781,
|
1927 |
+
"step": 1370
|
1928 |
+
},
|
1929 |
+
{
|
1930 |
+
"epoch": 0.55,
|
1931 |
+
"grad_norm": 14.388626098632812,
|
1932 |
+
"learning_rate": 2.196210497444749e-05,
|
1933 |
+
"loss": 2.9678,
|
1934 |
+
"step": 1375
|
1935 |
+
},
|
1936 |
+
{
|
1937 |
+
"epoch": 0.55,
|
1938 |
+
"grad_norm": 9.608166694641113,
|
1939 |
+
"learning_rate": 2.1803070957887347e-05,
|
1940 |
+
"loss": 3.0607,
|
1941 |
+
"step": 1380
|
1942 |
+
},
|
1943 |
+
{
|
1944 |
+
"epoch": 0.55,
|
1945 |
+
"grad_norm": 6.090882301330566,
|
1946 |
+
"learning_rate": 2.164416835455862e-05,
|
1947 |
+
"loss": 3.6207,
|
1948 |
+
"step": 1385
|
1949 |
+
},
|
1950 |
+
{
|
1951 |
+
"epoch": 0.56,
|
1952 |
+
"grad_norm": 13.572842597961426,
|
1953 |
+
"learning_rate": 2.1485403696324936e-05,
|
1954 |
+
"loss": 2.5546,
|
1955 |
+
"step": 1390
|
1956 |
+
},
|
1957 |
+
{
|
1958 |
+
"epoch": 0.56,
|
1959 |
+
"grad_norm": 7.186910629272461,
|
1960 |
+
"learning_rate": 2.1326783509379554e-05,
|
1961 |
+
"loss": 2.9678,
|
1962 |
+
"step": 1395
|
1963 |
+
},
|
1964 |
+
{
|
1965 |
+
"epoch": 0.56,
|
1966 |
+
"grad_norm": 50.014766693115234,
|
1967 |
+
"learning_rate": 2.11683143139771e-05,
|
1968 |
+
"loss": 3.1836,
|
1969 |
+
"step": 1400
|
1970 |
+
},
|
1971 |
+
{
|
1972 |
+
"epoch": 0.56,
|
1973 |
+
"grad_norm": 6.37592077255249,
|
1974 |
+
"learning_rate": 2.1010002624165527e-05,
|
1975 |
+
"loss": 2.659,
|
1976 |
+
"step": 1405
|
1977 |
+
},
|
1978 |
+
{
|
1979 |
+
"epoch": 0.56,
|
1980 |
+
"grad_norm": 22.413625717163086,
|
1981 |
+
"learning_rate": 2.0851854947518313e-05,
|
1982 |
+
"loss": 2.6929,
|
1983 |
+
"step": 1410
|
1984 |
+
},
|
1985 |
+
{
|
1986 |
+
"epoch": 0.57,
|
1987 |
+
"grad_norm": 11.021233558654785,
|
1988 |
+
"learning_rate": 2.069387778486703e-05,
|
1989 |
+
"loss": 2.7117,
|
1990 |
+
"step": 1415
|
1991 |
+
},
|
1992 |
+
{
|
1993 |
+
"epoch": 0.57,
|
1994 |
+
"grad_norm": 56.379844665527344,
|
1995 |
+
"learning_rate": 2.0536077630034086e-05,
|
1996 |
+
"loss": 3.4033,
|
1997 |
+
"step": 1420
|
1998 |
+
},
|
1999 |
+
{
|
2000 |
+
"epoch": 0.57,
|
2001 |
+
"grad_norm": 10.377793312072754,
|
2002 |
+
"learning_rate": 2.0378460969565782e-05,
|
2003 |
+
"loss": 2.4798,
|
2004 |
+
"step": 1425
|
2005 |
+
},
|
2006 |
+
{
|
2007 |
+
"epoch": 0.57,
|
2008 |
+
"grad_norm": 6.888087749481201,
|
2009 |
+
"learning_rate": 2.02210342824657e-05,
|
2010 |
+
"loss": 2.6521,
|
2011 |
+
"step": 1430
|
2012 |
+
},
|
2013 |
+
{
|
2014 |
+
"epoch": 0.57,
|
2015 |
+
"grad_norm": 6.764181613922119,
|
2016 |
+
"learning_rate": 2.0063804039928324e-05,
|
2017 |
+
"loss": 2.6824,
|
2018 |
+
"step": 1435
|
2019 |
+
},
|
2020 |
+
{
|
2021 |
+
"epoch": 0.58,
|
2022 |
+
"grad_norm": 55.408729553222656,
|
2023 |
+
"learning_rate": 1.9906776705073114e-05,
|
2024 |
+
"loss": 3.5755,
|
2025 |
+
"step": 1440
|
2026 |
+
},
|
2027 |
+
{
|
2028 |
+
"epoch": 0.58,
|
2029 |
+
"grad_norm": 8.327556610107422,
|
2030 |
+
"learning_rate": 1.9749958732678767e-05,
|
2031 |
+
"loss": 2.3398,
|
2032 |
+
"step": 1445
|
2033 |
+
},
|
2034 |
+
{
|
2035 |
+
"epoch": 0.58,
|
2036 |
+
"grad_norm": 9.360859870910645,
|
2037 |
+
"learning_rate": 1.9593356568917913e-05,
|
2038 |
+
"loss": 2.5018,
|
2039 |
+
"step": 1450
|
2040 |
+
},
|
2041 |
+
{
|
2042 |
+
"epoch": 0.58,
|
2043 |
+
"grad_norm": 32.50775146484375,
|
2044 |
+
"learning_rate": 1.9436976651092144e-05,
|
2045 |
+
"loss": 2.8295,
|
2046 |
+
"step": 1455
|
2047 |
+
},
|
2048 |
+
{
|
2049 |
+
"epoch": 0.58,
|
2050 |
+
"grad_norm": 23.467443466186523,
|
2051 |
+
"learning_rate": 1.928082540736737e-05,
|
2052 |
+
"loss": 2.6034,
|
2053 |
+
"step": 1460
|
2054 |
+
},
|
2055 |
+
{
|
2056 |
+
"epoch": 0.59,
|
2057 |
+
"grad_norm": 10.357468605041504,
|
2058 |
+
"learning_rate": 1.9124909256509622e-05,
|
2059 |
+
"loss": 3.0481,
|
2060 |
+
"step": 1465
|
2061 |
+
},
|
2062 |
+
{
|
2063 |
+
"epoch": 0.59,
|
2064 |
+
"grad_norm": 16.312400817871094,
|
2065 |
+
"learning_rate": 1.8969234607621186e-05,
|
2066 |
+
"loss": 2.9409,
|
2067 |
+
"step": 1470
|
2068 |
+
},
|
2069 |
+
{
|
2070 |
+
"epoch": 0.59,
|
2071 |
+
"grad_norm": 15.405855178833008,
|
2072 |
+
"learning_rate": 1.8813807859877147e-05,
|
2073 |
+
"loss": 2.6141,
|
2074 |
+
"step": 1475
|
2075 |
+
},
|
2076 |
+
{
|
2077 |
+
"epoch": 0.59,
|
2078 |
+
"grad_norm": 10.774744987487793,
|
2079 |
+
"learning_rate": 1.865863540226232e-05,
|
2080 |
+
"loss": 2.5919,
|
2081 |
+
"step": 1480
|
2082 |
+
},
|
2083 |
+
{
|
2084 |
+
"epoch": 0.59,
|
2085 |
+
"grad_norm": 7.001652717590332,
|
2086 |
+
"learning_rate": 1.8503723613308683e-05,
|
2087 |
+
"loss": 3.043,
|
2088 |
+
"step": 1485
|
2089 |
+
},
|
2090 |
+
{
|
2091 |
+
"epoch": 0.6,
|
2092 |
+
"grad_norm": 32.264766693115234,
|
2093 |
+
"learning_rate": 1.8349078860833123e-05,
|
2094 |
+
"loss": 2.711,
|
2095 |
+
"step": 1490
|
2096 |
+
},
|
2097 |
+
{
|
2098 |
+
"epoch": 0.6,
|
2099 |
+
"grad_norm": 15.870736122131348,
|
2100 |
+
"learning_rate": 1.8194707501675724e-05,
|
2101 |
+
"loss": 2.4717,
|
2102 |
+
"step": 1495
|
2103 |
+
},
|
2104 |
+
{
|
2105 |
+
"epoch": 0.6,
|
2106 |
+
"grad_norm": 15.47707748413086,
|
2107 |
+
"learning_rate": 1.8040615881438425e-05,
|
2108 |
+
"loss": 2.9706,
|
2109 |
+
"step": 1500
|
2110 |
+
},
|
2111 |
+
{
|
2112 |
+
"epoch": 0.6,
|
2113 |
+
"grad_norm": 12.465653419494629,
|
2114 |
+
"learning_rate": 1.7886810334224192e-05,
|
2115 |
+
"loss": 3.0024,
|
2116 |
+
"step": 1505
|
2117 |
+
},
|
2118 |
+
{
|
2119 |
+
"epoch": 0.6,
|
2120 |
+
"grad_norm": 20.99846839904785,
|
2121 |
+
"learning_rate": 1.7733297182376663e-05,
|
2122 |
+
"loss": 3.3546,
|
2123 |
+
"step": 1510
|
2124 |
+
},
|
2125 |
+
{
|
2126 |
+
"epoch": 0.61,
|
2127 |
+
"grad_norm": 37.41334533691406,
|
2128 |
+
"learning_rate": 1.7580082736220237e-05,
|
2129 |
+
"loss": 2.924,
|
2130 |
+
"step": 1515
|
2131 |
+
},
|
2132 |
+
{
|
2133 |
+
"epoch": 0.61,
|
2134 |
+
"grad_norm": 6.230709552764893,
|
2135 |
+
"learning_rate": 1.74271732938007e-05,
|
2136 |
+
"loss": 2.7253,
|
2137 |
+
"step": 1520
|
2138 |
+
},
|
2139 |
+
{
|
2140 |
+
"epoch": 0.61,
|
2141 |
+
"grad_norm": 15.803852081298828,
|
2142 |
+
"learning_rate": 1.7274575140626318e-05,
|
2143 |
+
"loss": 2.8601,
|
2144 |
+
"step": 1525
|
2145 |
+
},
|
2146 |
+
{
|
2147 |
+
"epoch": 0.61,
|
2148 |
+
"grad_norm": 6.981109142303467,
|
2149 |
+
"learning_rate": 1.7122294549409484e-05,
|
2150 |
+
"loss": 2.6754,
|
2151 |
+
"step": 1530
|
2152 |
+
},
|
2153 |
+
{
|
2154 |
+
"epoch": 0.61,
|
2155 |
+
"grad_norm": 21.247787475585938,
|
2156 |
+
"learning_rate": 1.6970337779808862e-05,
|
2157 |
+
"loss": 2.5148,
|
2158 |
+
"step": 1535
|
2159 |
+
},
|
2160 |
+
{
|
2161 |
+
"epoch": 0.62,
|
2162 |
+
"grad_norm": 13.401627540588379,
|
2163 |
+
"learning_rate": 1.6818711078172077e-05,
|
2164 |
+
"loss": 2.5783,
|
2165 |
+
"step": 1540
|
2166 |
+
},
|
2167 |
+
{
|
2168 |
+
"epoch": 0.62,
|
2169 |
+
"grad_norm": 8.62679386138916,
|
2170 |
+
"learning_rate": 1.666742067727896e-05,
|
2171 |
+
"loss": 2.5611,
|
2172 |
+
"step": 1545
|
2173 |
+
},
|
2174 |
+
{
|
2175 |
+
"epoch": 0.62,
|
2176 |
+
"grad_norm": 6.567328929901123,
|
2177 |
+
"learning_rate": 1.6516472796085315e-05,
|
2178 |
+
"loss": 2.3179,
|
2179 |
+
"step": 1550
|
2180 |
+
},
|
2181 |
+
{
|
2182 |
+
"epoch": 0.62,
|
2183 |
+
"grad_norm": 9.707379341125488,
|
2184 |
+
"learning_rate": 1.6365873639467315e-05,
|
2185 |
+
"loss": 2.3441,
|
2186 |
+
"step": 1555
|
2187 |
+
},
|
2188 |
+
{
|
2189 |
+
"epoch": 0.62,
|
2190 |
+
"grad_norm": 38.05426788330078,
|
2191 |
+
"learning_rate": 1.621562939796643e-05,
|
2192 |
+
"loss": 2.562,
|
2193 |
+
"step": 1560
|
2194 |
+
},
|
2195 |
+
{
|
2196 |
+
"epoch": 0.63,
|
2197 |
+
"grad_norm": 18.448406219482422,
|
2198 |
+
"learning_rate": 1.6065746247534984e-05,
|
2199 |
+
"loss": 2.6616,
|
2200 |
+
"step": 1565
|
2201 |
+
},
|
2202 |
+
{
|
2203 |
+
"epoch": 0.63,
|
2204 |
+
"grad_norm": 12.998920440673828,
|
2205 |
+
"learning_rate": 1.5916230349282215e-05,
|
2206 |
+
"loss": 2.9305,
|
2207 |
+
"step": 1570
|
2208 |
+
},
|
2209 |
+
{
|
2210 |
+
"epoch": 0.63,
|
2211 |
+
"grad_norm": 7.341577053070068,
|
2212 |
+
"learning_rate": 1.5767087849221096e-05,
|
2213 |
+
"loss": 2.6997,
|
2214 |
+
"step": 1575
|
2215 |
+
},
|
2216 |
+
{
|
2217 |
+
"epoch": 0.63,
|
2218 |
+
"grad_norm": 11.726909637451172,
|
2219 |
+
"learning_rate": 1.561832487801565e-05,
|
2220 |
+
"loss": 2.9281,
|
2221 |
+
"step": 1580
|
2222 |
+
},
|
2223 |
+
{
|
2224 |
+
"epoch": 0.63,
|
2225 |
+
"grad_norm": 10.877985000610352,
|
2226 |
+
"learning_rate": 1.5469947550728958e-05,
|
2227 |
+
"loss": 3.7181,
|
2228 |
+
"step": 1585
|
2229 |
+
},
|
2230 |
+
{
|
2231 |
+
"epoch": 0.64,
|
2232 |
+
"grad_norm": 11.845093727111816,
|
2233 |
+
"learning_rate": 1.53219619665718e-05,
|
2234 |
+
"loss": 3.1277,
|
2235 |
+
"step": 1590
|
2236 |
+
},
|
2237 |
+
{
|
2238 |
+
"epoch": 0.64,
|
2239 |
+
"grad_norm": 11.064024925231934,
|
2240 |
+
"learning_rate": 1.5174374208651912e-05,
|
2241 |
+
"loss": 2.8159,
|
2242 |
+
"step": 1595
|
2243 |
+
},
|
2244 |
+
{
|
2245 |
+
"epoch": 0.64,
|
2246 |
+
"grad_norm": 12.611387252807617,
|
2247 |
+
"learning_rate": 1.502719034372396e-05,
|
2248 |
+
"loss": 2.9906,
|
2249 |
+
"step": 1600
|
2250 |
+
},
|
2251 |
+
{
|
2252 |
+
"epoch": 0.64,
|
2253 |
+
"grad_norm": 8.44251537322998,
|
2254 |
+
"learning_rate": 1.4880416421940155e-05,
|
2255 |
+
"loss": 2.7055,
|
2256 |
+
"step": 1605
|
2257 |
+
},
|
2258 |
+
{
|
2259 |
+
"epoch": 0.64,
|
2260 |
+
"grad_norm": 11.020173072814941,
|
2261 |
+
"learning_rate": 1.4734058476601553e-05,
|
2262 |
+
"loss": 2.7852,
|
2263 |
+
"step": 1610
|
2264 |
+
},
|
2265 |
+
{
|
2266 |
+
"epoch": 0.65,
|
2267 |
+
"grad_norm": 18.893342971801758,
|
2268 |
+
"learning_rate": 1.458812252391003e-05,
|
2269 |
+
"loss": 2.7677,
|
2270 |
+
"step": 1615
|
2271 |
+
},
|
2272 |
+
{
|
2273 |
+
"epoch": 0.65,
|
2274 |
+
"grad_norm": 9.38496208190918,
|
2275 |
+
"learning_rate": 1.444261456272101e-05,
|
2276 |
+
"loss": 2.7732,
|
2277 |
+
"step": 1620
|
2278 |
+
},
|
2279 |
+
{
|
2280 |
+
"epoch": 0.65,
|
2281 |
+
"grad_norm": 6.147873401641846,
|
2282 |
+
"learning_rate": 1.4297540574296869e-05,
|
2283 |
+
"loss": 2.2796,
|
2284 |
+
"step": 1625
|
2285 |
+
},
|
2286 |
+
{
|
2287 |
+
"epoch": 0.65,
|
2288 |
+
"grad_norm": 18.142108917236328,
|
2289 |
+
"learning_rate": 1.4152906522061048e-05,
|
2290 |
+
"loss": 3.1047,
|
2291 |
+
"step": 1630
|
2292 |
+
},
|
2293 |
+
{
|
2294 |
+
"epoch": 0.65,
|
2295 |
+
"grad_norm": 16.73578643798828,
|
2296 |
+
"learning_rate": 1.400871835135295e-05,
|
2297 |
+
"loss": 2.9558,
|
2298 |
+
"step": 1635
|
2299 |
+
},
|
2300 |
+
{
|
2301 |
+
"epoch": 0.66,
|
2302 |
+
"grad_norm": 8.667417526245117,
|
2303 |
+
"learning_rate": 1.386498198918352e-05,
|
2304 |
+
"loss": 2.5587,
|
2305 |
+
"step": 1640
|
2306 |
+
},
|
2307 |
+
{
|
2308 |
+
"epoch": 0.66,
|
2309 |
+
"grad_norm": 29.769210815429688,
|
2310 |
+
"learning_rate": 1.3721703343991633e-05,
|
2311 |
+
"loss": 3.2146,
|
2312 |
+
"step": 1645
|
2313 |
+
},
|
2314 |
+
{
|
2315 |
+
"epoch": 0.66,
|
2316 |
+
"grad_norm": 27.924272537231445,
|
2317 |
+
"learning_rate": 1.3578888305401207e-05,
|
2318 |
+
"loss": 2.8355,
|
2319 |
+
"step": 1650
|
2320 |
+
},
|
2321 |
+
{
|
2322 |
+
"epoch": 0.66,
|
2323 |
+
"grad_norm": 8.13482666015625,
|
2324 |
+
"learning_rate": 1.3436542743979125e-05,
|
2325 |
+
"loss": 2.8218,
|
2326 |
+
"step": 1655
|
2327 |
+
},
|
2328 |
+
{
|
2329 |
+
"epoch": 0.66,
|
2330 |
+
"grad_norm": 16.816307067871094,
|
2331 |
+
"learning_rate": 1.329467251099386e-05,
|
2332 |
+
"loss": 2.5636,
|
2333 |
+
"step": 1660
|
2334 |
+
},
|
2335 |
+
{
|
2336 |
+
"epoch": 0.67,
|
2337 |
+
"grad_norm": 10.0423583984375,
|
2338 |
+
"learning_rate": 1.3153283438175034e-05,
|
2339 |
+
"loss": 2.3591,
|
2340 |
+
"step": 1665
|
2341 |
+
},
|
2342 |
+
{
|
2343 |
+
"epoch": 0.67,
|
2344 |
+
"grad_norm": 10.223281860351562,
|
2345 |
+
"learning_rate": 1.3012381337473656e-05,
|
2346 |
+
"loss": 3.0024,
|
2347 |
+
"step": 1670
|
2348 |
+
},
|
2349 |
+
{
|
2350 |
+
"epoch": 0.67,
|
2351 |
+
"grad_norm": 16.49390983581543,
|
2352 |
+
"learning_rate": 1.2871972000823196e-05,
|
2353 |
+
"loss": 2.9023,
|
2354 |
+
"step": 1675
|
2355 |
+
},
|
2356 |
+
{
|
2357 |
+
"epoch": 0.67,
|
2358 |
+
"grad_norm": 7.545689582824707,
|
2359 |
+
"learning_rate": 1.2732061199901562e-05,
|
2360 |
+
"loss": 2.8308,
|
2361 |
+
"step": 1680
|
2362 |
+
},
|
2363 |
+
{
|
2364 |
+
"epoch": 0.67,
|
2365 |
+
"grad_norm": 16.10993003845215,
|
2366 |
+
"learning_rate": 1.2592654685893757e-05,
|
2367 |
+
"loss": 2.6319,
|
2368 |
+
"step": 1685
|
2369 |
+
},
|
2370 |
+
{
|
2371 |
+
"epoch": 0.68,
|
2372 |
+
"grad_norm": 17.161388397216797,
|
2373 |
+
"learning_rate": 1.2453758189255568e-05,
|
2374 |
+
"loss": 2.6535,
|
2375 |
+
"step": 1690
|
2376 |
+
},
|
2377 |
+
{
|
2378 |
+
"epoch": 0.68,
|
2379 |
+
"grad_norm": 9.898162841796875,
|
2380 |
+
"learning_rate": 1.231537741947795e-05,
|
2381 |
+
"loss": 3.3956,
|
2382 |
+
"step": 1695
|
2383 |
+
},
|
2384 |
+
{
|
2385 |
+
"epoch": 0.68,
|
2386 |
+
"grad_norm": 8.259682655334473,
|
2387 |
+
"learning_rate": 1.217751806485235e-05,
|
2388 |
+
"loss": 2.5027,
|
2389 |
+
"step": 1700
|
2390 |
+
},
|
2391 |
+
{
|
2392 |
+
"epoch": 0.68,
|
2393 |
+
"grad_norm": 13.10878849029541,
|
2394 |
+
"learning_rate": 1.2040185792236874e-05,
|
2395 |
+
"loss": 2.4352,
|
2396 |
+
"step": 1705
|
2397 |
+
},
|
2398 |
+
{
|
2399 |
+
"epoch": 0.68,
|
2400 |
+
"grad_norm": 5.2070722579956055,
|
2401 |
+
"learning_rate": 1.1903386246823361e-05,
|
2402 |
+
"loss": 2.8799,
|
2403 |
+
"step": 1710
|
2404 |
+
},
|
2405 |
+
{
|
2406 |
+
"epoch": 0.69,
|
2407 |
+
"grad_norm": 12.991480827331543,
|
2408 |
+
"learning_rate": 1.1767125051905315e-05,
|
2409 |
+
"loss": 3.0858,
|
2410 |
+
"step": 1715
|
2411 |
+
},
|
2412 |
+
{
|
2413 |
+
"epoch": 0.69,
|
2414 |
+
"grad_norm": 10.183439254760742,
|
2415 |
+
"learning_rate": 1.1631407808646758e-05,
|
2416 |
+
"loss": 2.767,
|
2417 |
+
"step": 1720
|
2418 |
+
},
|
2419 |
+
{
|
2420 |
+
"epoch": 0.69,
|
2421 |
+
"grad_norm": 44.539459228515625,
|
2422 |
+
"learning_rate": 1.1496240095852001e-05,
|
2423 |
+
"loss": 2.9168,
|
2424 |
+
"step": 1725
|
2425 |
+
},
|
2426 |
+
{
|
2427 |
+
"epoch": 0.69,
|
2428 |
+
"grad_norm": 9.825093269348145,
|
2429 |
+
"learning_rate": 1.1361627469736285e-05,
|
2430 |
+
"loss": 2.6493,
|
2431 |
+
"step": 1730
|
2432 |
+
},
|
2433 |
+
{
|
2434 |
+
"epoch": 0.69,
|
2435 |
+
"grad_norm": 13.887325286865234,
|
2436 |
+
"learning_rate": 1.122757546369744e-05,
|
2437 |
+
"loss": 2.8159,
|
2438 |
+
"step": 1735
|
2439 |
+
},
|
2440 |
+
{
|
2441 |
+
"epoch": 0.7,
|
2442 |
+
"grad_norm": 6.512905120849609,
|
2443 |
+
"learning_rate": 1.1094089588088383e-05,
|
2444 |
+
"loss": 3.4574,
|
2445 |
+
"step": 1740
|
2446 |
+
},
|
2447 |
+
{
|
2448 |
+
"epoch": 0.7,
|
2449 |
+
"grad_norm": 8.677343368530273,
|
2450 |
+
"learning_rate": 1.096117532999063e-05,
|
2451 |
+
"loss": 2.7936,
|
2452 |
+
"step": 1745
|
2453 |
+
},
|
2454 |
+
{
|
2455 |
+
"epoch": 0.7,
|
2456 |
+
"grad_norm": 29.443038940429688,
|
2457 |
+
"learning_rate": 1.082883815298876e-05,
|
2458 |
+
"loss": 2.7603,
|
2459 |
+
"step": 1750
|
2460 |
+
},
|
2461 |
+
{
|
2462 |
+
"epoch": 0.7,
|
2463 |
+
"grad_norm": 6.62708044052124,
|
2464 |
+
"learning_rate": 1.0697083496945765e-05,
|
2465 |
+
"loss": 2.6984,
|
2466 |
+
"step": 1755
|
2467 |
+
},
|
2468 |
+
{
|
2469 |
+
"epoch": 0.7,
|
2470 |
+
"grad_norm": 11.575033187866211,
|
2471 |
+
"learning_rate": 1.0565916777779519e-05,
|
2472 |
+
"loss": 2.7648,
|
2473 |
+
"step": 1760
|
2474 |
+
},
|
2475 |
+
{
|
2476 |
+
"epoch": 0.71,
|
2477 |
+
"grad_norm": 19.933650970458984,
|
2478 |
+
"learning_rate": 1.0435343387240098e-05,
|
2479 |
+
"loss": 2.9776,
|
2480 |
+
"step": 1765
|
2481 |
+
},
|
2482 |
+
{
|
2483 |
+
"epoch": 0.71,
|
2484 |
+
"grad_norm": 12.603243827819824,
|
2485 |
+
"learning_rate": 1.0305368692688174e-05,
|
2486 |
+
"loss": 2.4419,
|
2487 |
+
"step": 1770
|
2488 |
+
},
|
2489 |
+
{
|
2490 |
+
"epoch": 0.71,
|
2491 |
+
"grad_norm": 8.084098815917969,
|
2492 |
+
"learning_rate": 1.0175998036874356e-05,
|
2493 |
+
"loss": 2.7489,
|
2494 |
+
"step": 1775
|
2495 |
+
},
|
2496 |
+
{
|
2497 |
+
"epoch": 0.71,
|
2498 |
+
"grad_norm": 20.572294235229492,
|
2499 |
+
"learning_rate": 1.0047236737719601e-05,
|
2500 |
+
"loss": 3.0075,
|
2501 |
+
"step": 1780
|
2502 |
+
},
|
2503 |
+
{
|
2504 |
+
"epoch": 0.71,
|
2505 |
+
"grad_norm": 8.42055606842041,
|
2506 |
+
"learning_rate": 9.919090088096589e-06,
|
2507 |
+
"loss": 2.7706,
|
2508 |
+
"step": 1785
|
2509 |
+
},
|
2510 |
+
{
|
2511 |
+
"epoch": 0.72,
|
2512 |
+
"grad_norm": 33.90031051635742,
|
2513 |
+
"learning_rate": 9.791563355612172e-06,
|
2514 |
+
"loss": 3.6592,
|
2515 |
+
"step": 1790
|
2516 |
+
},
|
2517 |
+
{
|
2518 |
+
"epoch": 0.72,
|
2519 |
+
"grad_norm": 8.972926139831543,
|
2520 |
+
"learning_rate": 9.664661782390841e-06,
|
2521 |
+
"loss": 2.4741,
|
2522 |
+
"step": 1795
|
2523 |
+
},
|
2524 |
+
{
|
2525 |
+
"epoch": 0.72,
|
2526 |
+
"grad_norm": 34.49540328979492,
|
2527 |
+
"learning_rate": 9.538390584859214e-06,
|
2528 |
+
"loss": 2.8379,
|
2529 |
+
"step": 1800
|
2530 |
+
},
|
2531 |
+
{
|
2532 |
+
"epoch": 0.72,
|
2533 |
+
"grad_norm": 17.665390014648438,
|
2534 |
+
"learning_rate": 9.412754953531663e-06,
|
2535 |
+
"loss": 3.0774,
|
2536 |
+
"step": 1805
|
2537 |
+
},
|
2538 |
+
{
|
2539 |
+
"epoch": 0.72,
|
2540 |
+
"grad_norm": 24.203062057495117,
|
2541 |
+
"learning_rate": 9.287760052796909e-06,
|
2542 |
+
"loss": 2.6639,
|
2543 |
+
"step": 1810
|
2544 |
+
},
|
2545 |
+
{
|
2546 |
+
"epoch": 0.73,
|
2547 |
+
"grad_norm": 15.584200859069824,
|
2548 |
+
"learning_rate": 9.163411020705762e-06,
|
2549 |
+
"loss": 2.75,
|
2550 |
+
"step": 1815
|
2551 |
+
},
|
2552 |
+
{
|
2553 |
+
"epoch": 0.73,
|
2554 |
+
"grad_norm": 9.29335880279541,
|
2555 |
+
"learning_rate": 9.039712968759864e-06,
|
2556 |
+
"loss": 2.9486,
|
2557 |
+
"step": 1820
|
2558 |
+
},
|
2559 |
+
{
|
2560 |
+
"epoch": 0.73,
|
2561 |
+
"grad_norm": 10.088394165039062,
|
2562 |
+
"learning_rate": 8.916670981701655e-06,
|
2563 |
+
"loss": 2.251,
|
2564 |
+
"step": 1825
|
2565 |
+
},
|
2566 |
+
{
|
2567 |
+
"epoch": 0.73,
|
2568 |
+
"grad_norm": 12.336640357971191,
|
2569 |
+
"learning_rate": 8.794290117305296e-06,
|
2570 |
+
"loss": 2.9527,
|
2571 |
+
"step": 1830
|
2572 |
+
},
|
2573 |
+
{
|
2574 |
+
"epoch": 0.73,
|
2575 |
+
"grad_norm": 8.988898277282715,
|
2576 |
+
"learning_rate": 8.672575406168782e-06,
|
2577 |
+
"loss": 2.6327,
|
2578 |
+
"step": 1835
|
2579 |
+
},
|
2580 |
+
{
|
2581 |
+
"epoch": 0.74,
|
2582 |
+
"grad_norm": 11.5975980758667,
|
2583 |
+
"learning_rate": 8.551531851507186e-06,
|
2584 |
+
"loss": 2.6512,
|
2585 |
+
"step": 1840
|
2586 |
+
},
|
2587 |
+
{
|
2588 |
+
"epoch": 0.74,
|
2589 |
+
"grad_norm": 23.88221549987793,
|
2590 |
+
"learning_rate": 8.431164428946927e-06,
|
2591 |
+
"loss": 2.6542,
|
2592 |
+
"step": 1845
|
2593 |
+
},
|
2594 |
+
{
|
2595 |
+
"epoch": 0.74,
|
2596 |
+
"grad_norm": 8.243393898010254,
|
2597 |
+
"learning_rate": 8.3114780863213e-06,
|
2598 |
+
"loss": 3.0325,
|
2599 |
+
"step": 1850
|
2600 |
+
},
|
2601 |
+
{
|
2602 |
+
"epoch": 0.74,
|
2603 |
+
"grad_norm": 13.184893608093262,
|
2604 |
+
"learning_rate": 8.192477743467078e-06,
|
2605 |
+
"loss": 2.7061,
|
2606 |
+
"step": 1855
|
2607 |
+
},
|
2608 |
+
{
|
2609 |
+
"epoch": 0.74,
|
2610 |
+
"grad_norm": 7.766141414642334,
|
2611 |
+
"learning_rate": 8.07416829202227e-06,
|
2612 |
+
"loss": 2.2925,
|
2613 |
+
"step": 1860
|
2614 |
+
},
|
2615 |
+
{
|
2616 |
+
"epoch": 0.75,
|
2617 |
+
"grad_norm": 30.14096450805664,
|
2618 |
+
"learning_rate": 7.956554595225016e-06,
|
2619 |
+
"loss": 3.4068,
|
2620 |
+
"step": 1865
|
2621 |
+
},
|
2622 |
+
{
|
2623 |
+
"epoch": 0.75,
|
2624 |
+
"grad_norm": 16.042314529418945,
|
2625 |
+
"learning_rate": 7.839641487713745e-06,
|
2626 |
+
"loss": 2.388,
|
2627 |
+
"step": 1870
|
2628 |
+
},
|
2629 |
+
{
|
2630 |
+
"epoch": 0.75,
|
2631 |
+
"grad_norm": 16.486698150634766,
|
2632 |
+
"learning_rate": 7.723433775328384e-06,
|
2633 |
+
"loss": 3.549,
|
2634 |
+
"step": 1875
|
2635 |
+
},
|
2636 |
+
{
|
2637 |
+
"epoch": 0.75,
|
2638 |
+
"grad_norm": 10.384564399719238,
|
2639 |
+
"learning_rate": 7.607936234912841e-06,
|
2640 |
+
"loss": 3.186,
|
2641 |
+
"step": 1880
|
2642 |
+
},
|
2643 |
+
{
|
2644 |
+
"epoch": 0.75,
|
2645 |
+
"grad_norm": 13.724809646606445,
|
2646 |
+
"learning_rate": 7.493153614118634e-06,
|
2647 |
+
"loss": 2.6417,
|
2648 |
+
"step": 1885
|
2649 |
+
},
|
2650 |
+
{
|
2651 |
+
"epoch": 0.76,
|
2652 |
+
"grad_norm": 8.79148006439209,
|
2653 |
+
"learning_rate": 7.379090631209712e-06,
|
2654 |
+
"loss": 2.6296,
|
2655 |
+
"step": 1890
|
2656 |
+
},
|
2657 |
+
{
|
2658 |
+
"epoch": 0.76,
|
2659 |
+
"grad_norm": 14.533072471618652,
|
2660 |
+
"learning_rate": 7.265751974868554e-06,
|
2661 |
+
"loss": 2.5174,
|
2662 |
+
"step": 1895
|
2663 |
+
},
|
2664 |
+
{
|
2665 |
+
"epoch": 0.76,
|
2666 |
+
"grad_norm": 11.014676094055176,
|
2667 |
+
"learning_rate": 7.153142304003418e-06,
|
2668 |
+
"loss": 2.5473,
|
2669 |
+
"step": 1900
|
2670 |
+
},
|
2671 |
+
{
|
2672 |
+
"epoch": 0.76,
|
2673 |
+
"grad_norm": 14.28827953338623,
|
2674 |
+
"learning_rate": 7.041266247556813e-06,
|
2675 |
+
"loss": 2.8404,
|
2676 |
+
"step": 1905
|
2677 |
+
},
|
2678 |
+
{
|
2679 |
+
"epoch": 0.76,
|
2680 |
+
"grad_norm": 13.483999252319336,
|
2681 |
+
"learning_rate": 6.930128404315214e-06,
|
2682 |
+
"loss": 2.6145,
|
2683 |
+
"step": 1910
|
2684 |
+
},
|
2685 |
+
{
|
2686 |
+
"epoch": 0.77,
|
2687 |
+
"grad_norm": 22.0334529876709,
|
2688 |
+
"learning_rate": 6.819733342720066e-06,
|
2689 |
+
"loss": 2.7059,
|
2690 |
+
"step": 1915
|
2691 |
+
},
|
2692 |
+
{
|
2693 |
+
"epoch": 0.77,
|
2694 |
+
"grad_norm": 42.69429397583008,
|
2695 |
+
"learning_rate": 6.7100856006799665e-06,
|
2696 |
+
"loss": 3.0099,
|
2697 |
+
"step": 1920
|
2698 |
+
},
|
2699 |
+
{
|
2700 |
+
"epoch": 0.77,
|
2701 |
+
"grad_norm": 9.883504867553711,
|
2702 |
+
"learning_rate": 6.601189685384126e-06,
|
2703 |
+
"loss": 2.9242,
|
2704 |
+
"step": 1925
|
2705 |
+
},
|
2706 |
+
{
|
2707 |
+
"epoch": 0.77,
|
2708 |
+
"grad_norm": 17.86440658569336,
|
2709 |
+
"learning_rate": 6.493050073117116e-06,
|
2710 |
+
"loss": 3.022,
|
2711 |
+
"step": 1930
|
2712 |
+
},
|
2713 |
+
{
|
2714 |
+
"epoch": 0.77,
|
2715 |
+
"grad_norm": 27.365087509155273,
|
2716 |
+
"learning_rate": 6.385671209074828e-06,
|
2717 |
+
"loss": 2.9742,
|
2718 |
+
"step": 1935
|
2719 |
+
},
|
2720 |
+
{
|
2721 |
+
"epoch": 0.78,
|
2722 |
+
"grad_norm": 13.641215324401855,
|
2723 |
+
"learning_rate": 6.279057507181796e-06,
|
2724 |
+
"loss": 2.4529,
|
2725 |
+
"step": 1940
|
2726 |
+
},
|
2727 |
+
{
|
2728 |
+
"epoch": 0.78,
|
2729 |
+
"grad_norm": 8.5147123336792,
|
2730 |
+
"learning_rate": 6.173213349909729e-06,
|
2731 |
+
"loss": 2.5926,
|
2732 |
+
"step": 1945
|
2733 |
+
},
|
2734 |
+
{
|
2735 |
+
"epoch": 0.78,
|
2736 |
+
"grad_norm": 13.009632110595703,
|
2737 |
+
"learning_rate": 6.068143088097372e-06,
|
2738 |
+
"loss": 2.4284,
|
2739 |
+
"step": 1950
|
2740 |
+
},
|
2741 |
+
{
|
2742 |
+
"epoch": 0.78,
|
2743 |
+
"grad_norm": 11.745450973510742,
|
2744 |
+
"learning_rate": 5.9638510407716394e-06,
|
2745 |
+
"loss": 2.7295,
|
2746 |
+
"step": 1955
|
2747 |
+
},
|
2748 |
+
{
|
2749 |
+
"epoch": 0.78,
|
2750 |
+
"grad_norm": 7.102616310119629,
|
2751 |
+
"learning_rate": 5.860341494970131e-06,
|
2752 |
+
"loss": 2.3844,
|
2753 |
+
"step": 1960
|
2754 |
+
},
|
2755 |
+
{
|
2756 |
+
"epoch": 0.79,
|
2757 |
+
"grad_norm": 19.79552459716797,
|
2758 |
+
"learning_rate": 5.757618705564849e-06,
|
2759 |
+
"loss": 2.6637,
|
2760 |
+
"step": 1965
|
2761 |
+
},
|
2762 |
+
{
|
2763 |
+
"epoch": 0.79,
|
2764 |
+
"grad_norm": 10.288678169250488,
|
2765 |
+
"learning_rate": 5.655686895087329e-06,
|
2766 |
+
"loss": 2.6376,
|
2767 |
+
"step": 1970
|
2768 |
+
},
|
2769 |
+
{
|
2770 |
+
"epoch": 0.79,
|
2771 |
+
"grad_norm": 5.515965938568115,
|
2772 |
+
"learning_rate": 5.554550253555066e-06,
|
2773 |
+
"loss": 2.7672,
|
2774 |
+
"step": 1975
|
2775 |
+
},
|
2776 |
+
{
|
2777 |
+
"epoch": 0.79,
|
2778 |
+
"grad_norm": 22.45775604248047,
|
2779 |
+
"learning_rate": 5.454212938299255e-06,
|
2780 |
+
"loss": 2.9637,
|
2781 |
+
"step": 1980
|
2782 |
+
},
|
2783 |
+
{
|
2784 |
+
"epoch": 0.79,
|
2785 |
+
"grad_norm": 8.804770469665527,
|
2786 |
+
"learning_rate": 5.354679073793942e-06,
|
2787 |
+
"loss": 2.4708,
|
2788 |
+
"step": 1985
|
2789 |
+
},
|
2790 |
+
{
|
2791 |
+
"epoch": 0.8,
|
2792 |
+
"grad_norm": 14.247252464294434,
|
2793 |
+
"learning_rate": 5.255952751486443e-06,
|
2794 |
+
"loss": 2.6551,
|
2795 |
+
"step": 1990
|
2796 |
+
},
|
2797 |
+
{
|
2798 |
+
"epoch": 0.8,
|
2799 |
+
"grad_norm": 7.674140930175781,
|
2800 |
+
"learning_rate": 5.158038029629195e-06,
|
2801 |
+
"loss": 2.7633,
|
2802 |
+
"step": 1995
|
2803 |
+
},
|
2804 |
+
{
|
2805 |
+
"epoch": 0.8,
|
2806 |
+
"grad_norm": 9.93137264251709,
|
2807 |
+
"learning_rate": 5.060938933112891e-06,
|
2808 |
+
"loss": 2.7034,
|
2809 |
+
"step": 2000
|
2810 |
+
},
|
2811 |
+
{
|
2812 |
+
"epoch": 0.8,
|
2813 |
+
"grad_norm": 11.792275428771973,
|
2814 |
+
"learning_rate": 4.9646594533010875e-06,
|
2815 |
+
"loss": 2.8564,
|
2816 |
+
"step": 2005
|
2817 |
+
},
|
2818 |
+
{
|
2819 |
+
"epoch": 0.8,
|
2820 |
+
"grad_norm": 13.885851860046387,
|
2821 |
+
"learning_rate": 4.869203547866097e-06,
|
2822 |
+
"loss": 2.6809,
|
2823 |
+
"step": 2010
|
2824 |
+
},
|
2825 |
+
{
|
2826 |
+
"epoch": 0.81,
|
2827 |
+
"grad_norm": 21.813217163085938,
|
2828 |
+
"learning_rate": 4.7745751406263165e-06,
|
2829 |
+
"loss": 3.0319,
|
2830 |
+
"step": 2015
|
2831 |
+
},
|
2832 |
+
{
|
2833 |
+
"epoch": 0.81,
|
2834 |
+
"grad_norm": 12.183605194091797,
|
2835 |
+
"learning_rate": 4.680778121384935e-06,
|
2836 |
+
"loss": 2.5685,
|
2837 |
+
"step": 2020
|
2838 |
+
},
|
2839 |
+
{
|
2840 |
+
"epoch": 0.81,
|
2841 |
+
"grad_norm": 26.5369930267334,
|
2842 |
+
"learning_rate": 4.587816345770032e-06,
|
2843 |
+
"loss": 2.9539,
|
2844 |
+
"step": 2025
|
2845 |
+
},
|
2846 |
+
{
|
2847 |
+
"epoch": 0.81,
|
2848 |
+
"grad_norm": 5.17782735824585,
|
2849 |
+
"learning_rate": 4.495693635076101e-06,
|
2850 |
+
"loss": 2.3644,
|
2851 |
+
"step": 2030
|
2852 |
+
},
|
2853 |
+
{
|
2854 |
+
"epoch": 0.81,
|
2855 |
+
"grad_norm": 21.22880744934082,
|
2856 |
+
"learning_rate": 4.404413776106958e-06,
|
2857 |
+
"loss": 3.0965,
|
2858 |
+
"step": 2035
|
2859 |
+
},
|
2860 |
+
{
|
2861 |
+
"epoch": 0.82,
|
2862 |
+
"grad_norm": 5.561256408691406,
|
2863 |
+
"learning_rate": 4.313980521020092e-06,
|
2864 |
+
"loss": 3.032,
|
2865 |
+
"step": 2040
|
2866 |
+
},
|
2867 |
+
{
|
2868 |
+
"epoch": 0.82,
|
2869 |
+
"grad_norm": 7.701003074645996,
|
2870 |
+
"learning_rate": 4.224397587172402e-06,
|
2871 |
+
"loss": 2.7757,
|
2872 |
+
"step": 2045
|
2873 |
+
},
|
2874 |
+
{
|
2875 |
+
"epoch": 0.82,
|
2876 |
+
"grad_norm": 17.345849990844727,
|
2877 |
+
"learning_rate": 4.135668656967434e-06,
|
2878 |
+
"loss": 2.3689,
|
2879 |
+
"step": 2050
|
2880 |
+
},
|
2881 |
+
{
|
2882 |
+
"epoch": 0.82,
|
2883 |
+
"grad_norm": 8.013896942138672,
|
2884 |
+
"learning_rate": 4.047797377703985e-06,
|
2885 |
+
"loss": 2.512,
|
2886 |
+
"step": 2055
|
2887 |
+
},
|
2888 |
+
{
|
2889 |
+
"epoch": 0.82,
|
2890 |
+
"grad_norm": 6.709656715393066,
|
2891 |
+
"learning_rate": 3.9607873614261715e-06,
|
2892 |
+
"loss": 3.0421,
|
2893 |
+
"step": 2060
|
2894 |
+
},
|
2895 |
+
{
|
2896 |
+
"epoch": 0.83,
|
2897 |
+
"grad_norm": 13.18295669555664,
|
2898 |
+
"learning_rate": 3.8746421847749765e-06,
|
2899 |
+
"loss": 2.4467,
|
2900 |
+
"step": 2065
|
2901 |
+
},
|
2902 |
+
{
|
2903 |
+
"epoch": 0.83,
|
2904 |
+
"grad_norm": 9.89919662475586,
|
2905 |
+
"learning_rate": 3.789365388841193e-06,
|
2906 |
+
"loss": 2.322,
|
2907 |
+
"step": 2070
|
2908 |
+
},
|
2909 |
+
{
|
2910 |
+
"epoch": 0.83,
|
2911 |
+
"grad_norm": 20.111215591430664,
|
2912 |
+
"learning_rate": 3.7049604790198976e-06,
|
2913 |
+
"loss": 3.2308,
|
2914 |
+
"step": 2075
|
2915 |
+
},
|
2916 |
+
{
|
2917 |
+
"epoch": 0.83,
|
2918 |
+
"grad_norm": 9.88414192199707,
|
2919 |
+
"learning_rate": 3.621430924866348e-06,
|
2920 |
+
"loss": 2.5939,
|
2921 |
+
"step": 2080
|
2922 |
+
},
|
2923 |
+
{
|
2924 |
+
"epoch": 0.83,
|
2925 |
+
"grad_norm": 8.166717529296875,
|
2926 |
+
"learning_rate": 3.5387801599533475e-06,
|
2927 |
+
"loss": 2.9439,
|
2928 |
+
"step": 2085
|
2929 |
+
},
|
2930 |
+
{
|
2931 |
+
"epoch": 0.84,
|
2932 |
+
"grad_norm": 31.1898193359375,
|
2933 |
+
"learning_rate": 3.4570115817301243e-06,
|
2934 |
+
"loss": 3.2292,
|
2935 |
+
"step": 2090
|
2936 |
+
},
|
2937 |
+
{
|
2938 |
+
"epoch": 0.84,
|
2939 |
+
"grad_norm": 26.66686248779297,
|
2940 |
+
"learning_rate": 3.3761285513826625e-06,
|
2941 |
+
"loss": 2.9332,
|
2942 |
+
"step": 2095
|
2943 |
+
},
|
2944 |
+
{
|
2945 |
+
"epoch": 0.84,
|
2946 |
+
"grad_norm": 4.934303283691406,
|
2947 |
+
"learning_rate": 3.296134393695538e-06,
|
2948 |
+
"loss": 2.7759,
|
2949 |
+
"step": 2100
|
2950 |
+
},
|
2951 |
+
{
|
2952 |
+
"epoch": 0.84,
|
2953 |
+
"grad_norm": 11.340413093566895,
|
2954 |
+
"learning_rate": 3.217032396915265e-06,
|
2955 |
+
"loss": 2.3886,
|
2956 |
+
"step": 2105
|
2957 |
+
},
|
2958 |
+
{
|
2959 |
+
"epoch": 0.84,
|
2960 |
+
"grad_norm": 10.123784065246582,
|
2961 |
+
"learning_rate": 3.1388258126151093e-06,
|
2962 |
+
"loss": 2.6871,
|
2963 |
+
"step": 2110
|
2964 |
+
},
|
2965 |
+
{
|
2966 |
+
"epoch": 0.85,
|
2967 |
+
"grad_norm": 9.508092880249023,
|
2968 |
+
"learning_rate": 3.06151785556143e-06,
|
2969 |
+
"loss": 3.0332,
|
2970 |
+
"step": 2115
|
2971 |
+
},
|
2972 |
+
{
|
2973 |
+
"epoch": 0.85,
|
2974 |
+
"grad_norm": 6.981659412384033,
|
2975 |
+
"learning_rate": 2.98511170358155e-06,
|
2976 |
+
"loss": 2.6556,
|
2977 |
+
"step": 2120
|
2978 |
+
},
|
2979 |
+
{
|
2980 |
+
"epoch": 0.85,
|
2981 |
+
"grad_norm": 7.004157066345215,
|
2982 |
+
"learning_rate": 2.9096104974331184e-06,
|
2983 |
+
"loss": 2.7863,
|
2984 |
+
"step": 2125
|
2985 |
+
},
|
2986 |
+
{
|
2987 |
+
"epoch": 0.85,
|
2988 |
+
"grad_norm": 15.677950859069824,
|
2989 |
+
"learning_rate": 2.8350173406749973e-06,
|
2990 |
+
"loss": 2.7697,
|
2991 |
+
"step": 2130
|
2992 |
+
},
|
2993 |
+
{
|
2994 |
+
"epoch": 0.85,
|
2995 |
+
"grad_norm": 7.333318710327148,
|
2996 |
+
"learning_rate": 2.7613352995397078e-06,
|
2997 |
+
"loss": 2.6726,
|
2998 |
+
"step": 2135
|
2999 |
+
},
|
3000 |
+
{
|
3001 |
+
"epoch": 0.86,
|
3002 |
+
"grad_norm": 5.832099437713623,
|
3003 |
+
"learning_rate": 2.688567402807357e-06,
|
3004 |
+
"loss": 2.8863,
|
3005 |
+
"step": 2140
|
3006 |
+
},
|
3007 |
+
{
|
3008 |
+
"epoch": 0.86,
|
3009 |
+
"grad_norm": 10.78770637512207,
|
3010 |
+
"learning_rate": 2.6167166416811746e-06,
|
3011 |
+
"loss": 2.8584,
|
3012 |
+
"step": 2145
|
3013 |
+
},
|
3014 |
+
{
|
3015 |
+
"epoch": 0.86,
|
3016 |
+
"grad_norm": 16.921499252319336,
|
3017 |
+
"learning_rate": 2.545785969664524e-06,
|
3018 |
+
"loss": 2.9597,
|
3019 |
+
"step": 2150
|
3020 |
+
},
|
3021 |
+
{
|
3022 |
+
"epoch": 0.86,
|
3023 |
+
"grad_norm": 8.275824546813965,
|
3024 |
+
"learning_rate": 2.475778302439524e-06,
|
3025 |
+
"loss": 2.1545,
|
3026 |
+
"step": 2155
|
3027 |
+
},
|
3028 |
+
{
|
3029 |
+
"epoch": 0.86,
|
3030 |
+
"grad_norm": 20.255998611450195,
|
3031 |
+
"learning_rate": 2.4066965177471645e-06,
|
3032 |
+
"loss": 2.5138,
|
3033 |
+
"step": 2160
|
3034 |
+
},
|
3035 |
+
{
|
3036 |
+
"epoch": 0.87,
|
3037 |
+
"grad_norm": 6.863336086273193,
|
3038 |
+
"learning_rate": 2.338543455269046e-06,
|
3039 |
+
"loss": 2.5736,
|
3040 |
+
"step": 2165
|
3041 |
+
},
|
3042 |
+
{
|
3043 |
+
"epoch": 0.87,
|
3044 |
+
"grad_norm": 10.660894393920898,
|
3045 |
+
"learning_rate": 2.271321916510627e-06,
|
3046 |
+
"loss": 2.5524,
|
3047 |
+
"step": 2170
|
3048 |
+
},
|
3049 |
+
{
|
3050 |
+
"epoch": 0.87,
|
3051 |
+
"grad_norm": 13.18586254119873,
|
3052 |
+
"learning_rate": 2.205034664686076e-06,
|
3053 |
+
"loss": 2.5716,
|
3054 |
+
"step": 2175
|
3055 |
+
},
|
3056 |
+
{
|
3057 |
+
"epoch": 0.87,
|
3058 |
+
"grad_norm": 11.016261100769043,
|
3059 |
+
"learning_rate": 2.1396844246046903e-06,
|
3060 |
+
"loss": 3.3971,
|
3061 |
+
"step": 2180
|
3062 |
+
},
|
3063 |
+
{
|
3064 |
+
"epoch": 0.87,
|
3065 |
+
"grad_norm": 23.928808212280273,
|
3066 |
+
"learning_rate": 2.075273882558873e-06,
|
3067 |
+
"loss": 2.7627,
|
3068 |
+
"step": 2185
|
3069 |
+
},
|
3070 |
+
{
|
3071 |
+
"epoch": 0.88,
|
3072 |
+
"grad_norm": 8.036683082580566,
|
3073 |
+
"learning_rate": 2.0118056862137357e-06,
|
3074 |
+
"loss": 2.7452,
|
3075 |
+
"step": 2190
|
3076 |
+
},
|
3077 |
+
{
|
3078 |
+
"epoch": 0.88,
|
3079 |
+
"grad_norm": 15.620336532592773,
|
3080 |
+
"learning_rate": 1.949282444498238e-06,
|
3081 |
+
"loss": 2.7892,
|
3082 |
+
"step": 2195
|
3083 |
+
},
|
3084 |
+
{
|
3085 |
+
"epoch": 0.88,
|
3086 |
+
"grad_norm": 5.829482078552246,
|
3087 |
+
"learning_rate": 1.8877067274979648e-06,
|
3088 |
+
"loss": 2.7427,
|
3089 |
+
"step": 2200
|
3090 |
+
},
|
3091 |
+
{
|
3092 |
+
"epoch": 0.88,
|
3093 |
+
"grad_norm": 24.735841751098633,
|
3094 |
+
"learning_rate": 1.827081066349459e-06,
|
3095 |
+
"loss": 2.3911,
|
3096 |
+
"step": 2205
|
3097 |
+
},
|
3098 |
+
{
|
3099 |
+
"epoch": 0.88,
|
3100 |
+
"grad_norm": 15.522488594055176,
|
3101 |
+
"learning_rate": 1.767407953136202e-06,
|
3102 |
+
"loss": 2.7896,
|
3103 |
+
"step": 2210
|
3104 |
+
},
|
3105 |
+
{
|
3106 |
+
"epoch": 0.89,
|
3107 |
+
"grad_norm": 10.789400100708008,
|
3108 |
+
"learning_rate": 1.7086898407861485e-06,
|
3109 |
+
"loss": 2.6058,
|
3110 |
+
"step": 2215
|
3111 |
+
},
|
3112 |
+
{
|
3113 |
+
"epoch": 0.89,
|
3114 |
+
"grad_norm": 12.961084365844727,
|
3115 |
+
"learning_rate": 1.6509291429709223e-06,
|
3116 |
+
"loss": 2.1545,
|
3117 |
+
"step": 2220
|
3118 |
+
},
|
3119 |
+
{
|
3120 |
+
"epoch": 0.89,
|
3121 |
+
"grad_norm": 42.19524002075195,
|
3122 |
+
"learning_rate": 1.59412823400657e-06,
|
3123 |
+
"loss": 2.7638,
|
3124 |
+
"step": 2225
|
3125 |
+
},
|
3126 |
+
{
|
3127 |
+
"epoch": 0.89,
|
3128 |
+
"grad_norm": 11.88717269897461,
|
3129 |
+
"learning_rate": 1.538289448755989e-06,
|
3130 |
+
"loss": 3.258,
|
3131 |
+
"step": 2230
|
3132 |
+
},
|
3133 |
+
{
|
3134 |
+
"epoch": 0.89,
|
3135 |
+
"grad_norm": 9.746241569519043,
|
3136 |
+
"learning_rate": 1.483415082532938e-06,
|
3137 |
+
"loss": 2.4654,
|
3138 |
+
"step": 2235
|
3139 |
+
},
|
3140 |
+
{
|
3141 |
+
"epoch": 0.9,
|
3142 |
+
"grad_norm": 8.284168243408203,
|
3143 |
+
"learning_rate": 1.4295073910076757e-06,
|
3144 |
+
"loss": 2.8934,
|
3145 |
+
"step": 2240
|
3146 |
+
},
|
3147 |
+
{
|
3148 |
+
"epoch": 0.9,
|
3149 |
+
"grad_norm": 6.290571212768555,
|
3150 |
+
"learning_rate": 1.3765685901142716e-06,
|
3151 |
+
"loss": 2.8615,
|
3152 |
+
"step": 2245
|
3153 |
+
},
|
3154 |
+
{
|
3155 |
+
"epoch": 0.9,
|
3156 |
+
"grad_norm": 9.758255004882812,
|
3157 |
+
"learning_rate": 1.3246008559594709e-06,
|
3158 |
+
"loss": 2.4915,
|
3159 |
+
"step": 2250
|
3160 |
+
},
|
3161 |
+
{
|
3162 |
+
"epoch": 0.9,
|
3163 |
+
"grad_norm": 64.08142852783203,
|
3164 |
+
"learning_rate": 1.273606324733284e-06,
|
3165 |
+
"loss": 2.6644,
|
3166 |
+
"step": 2255
|
3167 |
+
},
|
3168 |
+
{
|
3169 |
+
"epoch": 0.9,
|
3170 |
+
"grad_norm": 6.796569347381592,
|
3171 |
+
"learning_rate": 1.2235870926211619e-06,
|
3172 |
+
"loss": 3.0118,
|
3173 |
+
"step": 2260
|
3174 |
+
},
|
3175 |
+
{
|
3176 |
+
"epoch": 0.91,
|
3177 |
+
"grad_norm": 48.14067459106445,
|
3178 |
+
"learning_rate": 1.1745452157178206e-06,
|
3179 |
+
"loss": 2.5428,
|
3180 |
+
"step": 2265
|
3181 |
+
},
|
3182 |
+
{
|
3183 |
+
"epoch": 0.91,
|
3184 |
+
"grad_norm": 78.56604766845703,
|
3185 |
+
"learning_rate": 1.1264827099427417e-06,
|
3186 |
+
"loss": 3.2121,
|
3187 |
+
"step": 2270
|
3188 |
+
},
|
3189 |
+
{
|
3190 |
+
"epoch": 0.91,
|
3191 |
+
"grad_norm": 8.145052909851074,
|
3192 |
+
"learning_rate": 1.0794015509572818e-06,
|
3193 |
+
"loss": 2.7715,
|
3194 |
+
"step": 2275
|
3195 |
+
},
|
3196 |
+
{
|
3197 |
+
"epoch": 0.91,
|
3198 |
+
"grad_norm": 9.410415649414062,
|
3199 |
+
"learning_rate": 1.0333036740834856e-06,
|
3200 |
+
"loss": 2.9311,
|
3201 |
+
"step": 2280
|
3202 |
+
},
|
3203 |
+
{
|
3204 |
+
"epoch": 0.91,
|
3205 |
+
"grad_norm": 26.370338439941406,
|
3206 |
+
"learning_rate": 9.881909742245177e-07,
|
3207 |
+
"loss": 2.8867,
|
3208 |
+
"step": 2285
|
3209 |
+
},
|
3210 |
+
{
|
3211 |
+
"epoch": 0.92,
|
3212 |
+
"grad_norm": 19.1849365234375,
|
3213 |
+
"learning_rate": 9.440653057867815e-07,
|
3214 |
+
"loss": 2.559,
|
3215 |
+
"step": 2290
|
3216 |
+
},
|
3217 |
+
{
|
3218 |
+
"epoch": 0.92,
|
3219 |
+
"grad_norm": 25.568355560302734,
|
3220 |
+
"learning_rate": 9.009284826036691e-07,
|
3221 |
+
"loss": 2.5787,
|
3222 |
+
"step": 2295
|
3223 |
+
},
|
3224 |
+
{
|
3225 |
+
"epoch": 0.92,
|
3226 |
+
"grad_norm": 14.307110786437988,
|
3227 |
+
"learning_rate": 8.587822778610283e-07,
|
3228 |
+
"loss": 2.8058,
|
3229 |
+
"step": 2300
|
3230 |
+
},
|
3231 |
+
{
|
3232 |
+
"epoch": 0.92,
|
3233 |
+
"grad_norm": 25.297014236450195,
|
3234 |
+
"learning_rate": 8.176284240242638e-07,
|
3235 |
+
"loss": 2.8921,
|
3236 |
+
"step": 2305
|
3237 |
+
},
|
3238 |
+
{
|
3239 |
+
"epoch": 0.92,
|
3240 |
+
"grad_norm": 19.239059448242188,
|
3241 |
+
"learning_rate": 7.774686127671183e-07,
|
3242 |
+
"loss": 2.6122,
|
3243 |
+
"step": 2310
|
3244 |
+
},
|
3245 |
+
{
|
3246 |
+
"epoch": 0.93,
|
3247 |
+
"grad_norm": 11.51448917388916,
|
3248 |
+
"learning_rate": 7.383044949021339e-07,
|
3249 |
+
"loss": 2.2361,
|
3250 |
+
"step": 2315
|
3251 |
+
},
|
3252 |
+
{
|
3253 |
+
"epoch": 0.93,
|
3254 |
+
"grad_norm": 6.253133773803711,
|
3255 |
+
"learning_rate": 7.00137680312804e-07,
|
3256 |
+
"loss": 2.5475,
|
3257 |
+
"step": 2320
|
3258 |
+
},
|
3259 |
+
{
|
3260 |
+
"epoch": 0.93,
|
3261 |
+
"grad_norm": 17.918170928955078,
|
3262 |
+
"learning_rate": 6.62969737887384e-07,
|
3263 |
+
"loss": 2.8589,
|
3264 |
+
"step": 2325
|
3265 |
+
},
|
3266 |
+
{
|
3267 |
+
"epoch": 0.93,
|
3268 |
+
"grad_norm": 15.065885543823242,
|
3269 |
+
"learning_rate": 6.268021954544096e-07,
|
3270 |
+
"loss": 2.9153,
|
3271 |
+
"step": 2330
|
3272 |
+
},
|
3273 |
+
{
|
3274 |
+
"epoch": 0.93,
|
3275 |
+
"grad_norm": 13.842984199523926,
|
3276 |
+
"learning_rate": 5.916365397198975e-07,
|
3277 |
+
"loss": 2.894,
|
3278 |
+
"step": 2335
|
3279 |
+
},
|
3280 |
+
{
|
3281 |
+
"epoch": 0.94,
|
3282 |
+
"grad_norm": 10.621698379516602,
|
3283 |
+
"learning_rate": 5.574742162062163e-07,
|
3284 |
+
"loss": 3.1294,
|
3285 |
+
"step": 2340
|
3286 |
+
},
|
3287 |
+
{
|
3288 |
+
"epoch": 0.94,
|
3289 |
+
"grad_norm": 13.219593048095703,
|
3290 |
+
"learning_rate": 5.243166291926782e-07,
|
3291 |
+
"loss": 2.6543,
|
3292 |
+
"step": 2345
|
3293 |
+
},
|
3294 |
+
{
|
3295 |
+
"epoch": 0.94,
|
3296 |
+
"grad_norm": 11.599324226379395,
|
3297 |
+
"learning_rate": 4.921651416578188e-07,
|
3298 |
+
"loss": 2.3677,
|
3299 |
+
"step": 2350
|
3300 |
+
},
|
3301 |
+
{
|
3302 |
+
"epoch": 0.94,
|
3303 |
+
"grad_norm": 8.731278419494629,
|
3304 |
+
"learning_rate": 4.6102107522336403e-07,
|
3305 |
+
"loss": 2.7634,
|
3306 |
+
"step": 2355
|
3307 |
+
},
|
3308 |
+
{
|
3309 |
+
"epoch": 0.94,
|
3310 |
+
"grad_norm": 43.908077239990234,
|
3311 |
+
"learning_rate": 4.308857100999042e-07,
|
3312 |
+
"loss": 2.8106,
|
3313 |
+
"step": 2360
|
3314 |
+
},
|
3315 |
+
{
|
3316 |
+
"epoch": 0.95,
|
3317 |
+
"grad_norm": 9.233963966369629,
|
3318 |
+
"learning_rate": 4.0176028503425835e-07,
|
3319 |
+
"loss": 2.5252,
|
3320 |
+
"step": 2365
|
3321 |
+
},
|
3322 |
+
{
|
3323 |
+
"epoch": 0.95,
|
3324 |
+
"grad_norm": 13.496291160583496,
|
3325 |
+
"learning_rate": 3.7364599725858153e-07,
|
3326 |
+
"loss": 2.2908,
|
3327 |
+
"step": 2370
|
3328 |
+
},
|
3329 |
+
{
|
3330 |
+
"epoch": 0.95,
|
3331 |
+
"grad_norm": 11.278969764709473,
|
3332 |
+
"learning_rate": 3.465440024411265e-07,
|
3333 |
+
"loss": 2.5045,
|
3334 |
+
"step": 2375
|
3335 |
+
},
|
3336 |
+
{
|
3337 |
+
"epoch": 0.95,
|
3338 |
+
"grad_norm": 13.121251106262207,
|
3339 |
+
"learning_rate": 3.204554146387456e-07,
|
3340 |
+
"loss": 2.4661,
|
3341 |
+
"step": 2380
|
3342 |
+
},
|
3343 |
+
{
|
3344 |
+
"epoch": 0.95,
|
3345 |
+
"grad_norm": 17.475078582763672,
|
3346 |
+
"learning_rate": 2.9538130625110796e-07,
|
3347 |
+
"loss": 2.9907,
|
3348 |
+
"step": 2385
|
3349 |
+
},
|
3350 |
+
{
|
3351 |
+
"epoch": 0.96,
|
3352 |
+
"grad_norm": 6.753970146179199,
|
3353 |
+
"learning_rate": 2.7132270797659563e-07,
|
3354 |
+
"loss": 2.4326,
|
3355 |
+
"step": 2390
|
3356 |
+
},
|
3357 |
+
{
|
3358 |
+
"epoch": 0.96,
|
3359 |
+
"grad_norm": 18.052387237548828,
|
3360 |
+
"learning_rate": 2.482806087699546e-07,
|
3361 |
+
"loss": 3.0056,
|
3362 |
+
"step": 2395
|
3363 |
+
},
|
3364 |
+
{
|
3365 |
+
"epoch": 0.96,
|
3366 |
+
"grad_norm": 11.094266891479492,
|
3367 |
+
"learning_rate": 2.262559558016325e-07,
|
3368 |
+
"loss": 2.7058,
|
3369 |
+
"step": 2400
|
3370 |
+
}
|
3371 |
+
],
|
3372 |
+
"logging_steps": 5,
|
3373 |
+
"max_steps": 2500,
|
3374 |
+
"num_input_tokens_seen": 0,
|
3375 |
+
"num_train_epochs": 1,
|
3376 |
+
"save_steps": 200,
|
3377 |
+
"total_flos": 1.3202552777146368e+16,
|
3378 |
+
"train_batch_size": 1,
|
3379 |
+
"trial_name": null,
|
3380 |
+
"trial_params": null
|
3381 |
+
}
|
training_args.bin
ADDED
@@ -0,0 +1,3 @@
|
|
|
|
|
|
|
|
|
1 |
+
version https://git-lfs.github.com/spec/v1
|
2 |
+
oid sha256:da208503ba553e055a114829f3f9b1801f58a080e541dba7b2d0db27a99b7ad4
|
3 |
+
size 5048
|