TheBloke commited on
Commit
ed3e7e0
1 Parent(s): d1f6f27

Initial GPTQ model commit, of compat file.

Browse files
README.md ADDED
@@ -0,0 +1,92 @@
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
1
+ ---
2
+ language:
3
+ - en
4
+ tags:
5
+ - causal-lm
6
+ - llama
7
+ inference: false
8
+ ---
9
+ # Wizard-Vicuna-13B-GPTQ
10
+
11
+ This is GPTQ format quantised 4bit models of [Eric Hartford's 'uncensored' training of Wizard-Vicuna 13B](https://huggingface.co/ehartford/Wizard-Vicuna-13B-Uncensored).
12
+
13
+ It is the result of quantising to 4bit using [GPTQ-for-LLaMa](https://github.com/qwopqwop200/GPTQ-for-LLaMa).
14
+
15
+ ## Repositories available
16
+
17
+ * [4bit GPTQ models for GPU inference](https://huggingface.co/TheBloke/Wizard-Vicuna-13B-Uncensored-GPTQ).
18
+ * [4bit and 5bit GGML models for CPU inference](https://huggingface.co/TheBloke/Wizard-Vicuna-13B-Uncensored-GGML).
19
+ * [float16 HF format model for GPU inference and further conversions](https://huggingface.co/TheBloke/Wizard-Vicuna-13B-Uncensored-HF).
20
+
21
+ ## How to easily download and use this model in text-generation-webui
22
+
23
+ Open the text-generation-webui UI as normal.
24
+
25
+ 1. Click the **Model tab**.
26
+ 2. Under **Download custom model or LoRA**, enter `TheBloke/Wizard-Vicuna-13B-Uncensored-GPTQ`.
27
+ 3. Click **Download**.
28
+ 4. Wait until it says it's finished downloading.
29
+ 5. Click the **Refresh** icon next to **Model** in the top left.
30
+ 6. In the **Model drop-down**: choose the model you just downloaded, `Wizard-Vicuna-13B-Uncensored-GPTQ`.
31
+ 7. If you see an error in the bottom right, ignore it - it's temporary.
32
+ 8. Fill out the `GPTQ parameters` on the right: `Bits = 4`, `Groupsize = 128`, `model_type = Llama`
33
+ 9. Click **Save settings for this model** in the top right.
34
+ 10. Click **Reload the Model** in the top right.
35
+ 11. Once it says it's loaded, click the **Text Generation tab** and enter a prompt!
36
+
37
+ ## Provided files
38
+
39
+ **Compatible file - Wizard-Vicuna-13B-Uncensored-GPTQ-4bit-128g.compat.no-act-order.safetensors**
40
+
41
+ In the `main` branch - the default one - you will find `Wizard-Vicuna-13B-Uncensored-GPTQ-4bit-128g.compat.no-act-order.safetensors`
42
+
43
+ This will work with all versions of GPTQ-for-LLaMa. It has maximum compatibility
44
+
45
+ It was created without the `--act-order` parameter. It may have slightly lower inference quality compared to the other file, but is guaranteed to work on all versions of GPTQ-for-LLaMa and text-generation-webui.
46
+
47
+ * `Wizard-Vicuna-13B-Uncensored-GPTQ-4bit-128g.compat.no-act-order.safetensors`
48
+ * Works with all versions of GPTQ-for-LLaMa code, both Triton and CUDA branches
49
+ * Works with AutoGPTQ.
50
+ * Works with text-generation-webui one-click-installers
51
+ * Parameters: Groupsize = 128g. No act-order.
52
+ * Command used to create the GPTQ:
53
+ ```
54
+ python llama.py ehartford_Wizard-Vicuna-13B-Uncensored c4 --wbits 4 --groupsize 128 --true-sequential --save_safetensors Wizard-Vicuna-13B-Uncensored-GPTQ-4bit-128g.compat.no-act-order.safetensors
55
+ ```
56
+
57
+ **Latest file with act-order - Wizard-Vicuna-13B-Uncensored-GPTQ-4bit-128g.latest.act-order.safetensors**
58
+
59
+ In the `latest` branch you will find `Wizard-Vicuna-13B-Uncensored-GPTQ-4bit-128g.latest.act-order.safetensors`
60
+
61
+ This requires recent GPTQ-for-LLaMa code. It will not work with ooba's fork.
62
+
63
+ It was created with the `--act-order` parameter to maximise inference quality.
64
+
65
+
66
+ To download this branch in text-generation-webui, enter `TheBloke/Wizard-Vicuna-13B-Uncensored-GPTQ:latest` in the Download Model box.
67
+
68
+ * `Wizard-Vicuna-13B-Uncensored-GPTQ-4bit-128g.latest.act-order.safetensors`
69
+ * Works only with recent GPTQ-for-LLaMa.
70
+ * Works with AutoGPTQ.
71
+ * Will not work with text-generation-webui one-click-installers
72
+ * Parameters: Groupsize = 128g. Act order.
73
+ * Command used to create the GPTQ:
74
+ ```
75
+ python llama.py ehartford_Wizard-Vicuna-13B-Uncensored c4 --wbits 4 --groupsize 128 --true-sequential --act-order --save_safetensors Wizard-Vicuna-13B-Uncensored-GPTQ-4bit-128g.latest.act-order.safetensors
76
+ ```
77
+
78
+ # Original model card
79
+
80
+ This is [wizard-vicuna-13b](https://huggingface.co/junelee/wizard-vicuna-13b) trained with a subset of the dataset - responses that contained alignment / moralizing were removed. The intent is to train a WizardLM that doesn't have alignment built-in, so that alignment (of any sort) can be added separately with for example with a RLHF LoRA.
81
+
82
+ Shout out to the open source AI/ML community, and everyone who helped me out.
83
+
84
+ Note:
85
+
86
+ An uncensored model has no guardrails.
87
+
88
+ You are responsible for anything you do with the model, just as you are responsible for anything you do with any dangerous object such as a knife, gun, lighter, or car.
89
+
90
+ Publishing anything this model generates is the same as publishing it yourself.
91
+
92
+ You are responsible for the content you publish, and you cannot blame the model any more than you can blame the knife, gun, lighter, or car for what you do with it.
Wizard-Vicuna-13B-Uncensored-GPTQ-4bit-128g.compat.no-act-order.safetensors ADDED
@@ -0,0 +1,3 @@
 
 
 
 
1
+ version https://git-lfs.github.com/spec/v1
2
+ oid sha256:5acd7195dbe3826d44e77d5184d3f0b29ac0add2f77a4b3f6f848d301fc797c5
3
+ size 8110988216
config.json ADDED
@@ -0,0 +1,24 @@
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
1
+ {
2
+ "_name_or_path": "/workspace/models/llama-13b",
3
+ "architectures": [
4
+ "LlamaForCausalLM"
5
+ ],
6
+ "bos_token_id": 1,
7
+ "eos_token_id": 2,
8
+ "hidden_act": "silu",
9
+ "hidden_size": 5120,
10
+ "initializer_range": 0.02,
11
+ "intermediate_size": 13824,
12
+ "max_position_embeddings": 2048,
13
+ "max_sequence_length": 2048,
14
+ "model_type": "llama",
15
+ "num_attention_heads": 40,
16
+ "num_hidden_layers": 40,
17
+ "pad_token_id": 0,
18
+ "rms_norm_eps": 1e-06,
19
+ "tie_word_embeddings": false,
20
+ "torch_dtype": "float32",
21
+ "transformers_version": "4.28.1",
22
+ "use_cache": true,
23
+ "vocab_size": 32000
24
+ }
generation_config.json ADDED
@@ -0,0 +1,7 @@
 
 
 
 
 
 
 
 
1
+ {
2
+ "_from_model_config": true,
3
+ "bos_token_id": 1,
4
+ "eos_token_id": 2,
5
+ "pad_token_id": 0,
6
+ "transformers_version": "4.28.1"
7
+ }
special_tokens_map.json ADDED
@@ -0,0 +1,24 @@
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
1
+ {
2
+ "bos_token": {
3
+ "content": "<s>",
4
+ "lstrip": false,
5
+ "normalized": true,
6
+ "rstrip": false,
7
+ "single_word": false
8
+ },
9
+ "eos_token": {
10
+ "content": "</s>",
11
+ "lstrip": false,
12
+ "normalized": true,
13
+ "rstrip": false,
14
+ "single_word": false
15
+ },
16
+ "pad_token": "<unk>",
17
+ "unk_token": {
18
+ "content": "<unk>",
19
+ "lstrip": false,
20
+ "normalized": true,
21
+ "rstrip": false,
22
+ "single_word": false
23
+ }
24
+ }
tokenizer.json ADDED
The diff for this file is too large to render. See raw diff
 
tokenizer.model ADDED
@@ -0,0 +1,3 @@
 
 
 
 
1
+ version https://git-lfs.github.com/spec/v1
2
+ oid sha256:9e556afd44213b6bd1be2b850ebbbd98f5481437a8021afaf58ee7fb1818d347
3
+ size 499723
tokenizer_config.json ADDED
@@ -0,0 +1,34 @@
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
1
+ {
2
+ "add_bos_token": true,
3
+ "add_eos_token": false,
4
+ "bos_token": {
5
+ "__type": "AddedToken",
6
+ "content": "<s>",
7
+ "lstrip": false,
8
+ "normalized": true,
9
+ "rstrip": false,
10
+ "single_word": false
11
+ },
12
+ "clean_up_tokenization_spaces": false,
13
+ "eos_token": {
14
+ "__type": "AddedToken",
15
+ "content": "</s>",
16
+ "lstrip": false,
17
+ "normalized": true,
18
+ "rstrip": false,
19
+ "single_word": false
20
+ },
21
+ "model_max_length": 2048,
22
+ "pad_token": null,
23
+ "padding_side": "right",
24
+ "sp_model_kwargs": {},
25
+ "tokenizer_class": "LlamaTokenizer",
26
+ "unk_token": {
27
+ "__type": "AddedToken",
28
+ "content": "<unk>",
29
+ "lstrip": false,
30
+ "normalized": true,
31
+ "rstrip": false,
32
+ "single_word": false
33
+ }
34
+ }
trainer_state.json ADDED
@@ -0,0 +1,2401 @@
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
1
+ {
2
+ "best_metric": null,
3
+ "best_model_checkpoint": null,
4
+ "epoch": 2.9924528301886792,
5
+ "global_step": 396,
6
+ "is_hyper_param_search": false,
7
+ "is_local_process_zero": true,
8
+ "is_world_process_zero": true,
9
+ "log_history": [
10
+ {
11
+ "epoch": 0.01,
12
+ "learning_rate": 1.25e-06,
13
+ "loss": 0.8661,
14
+ "step": 1
15
+ },
16
+ {
17
+ "epoch": 0.02,
18
+ "learning_rate": 2.5e-06,
19
+ "loss": 0.8758,
20
+ "step": 2
21
+ },
22
+ {
23
+ "epoch": 0.02,
24
+ "learning_rate": 3.7500000000000005e-06,
25
+ "loss": 0.7376,
26
+ "step": 3
27
+ },
28
+ {
29
+ "epoch": 0.03,
30
+ "learning_rate": 5e-06,
31
+ "loss": 0.7434,
32
+ "step": 4
33
+ },
34
+ {
35
+ "epoch": 0.04,
36
+ "learning_rate": 6.25e-06,
37
+ "loss": 0.6945,
38
+ "step": 5
39
+ },
40
+ {
41
+ "epoch": 0.05,
42
+ "learning_rate": 7.500000000000001e-06,
43
+ "loss": 0.6896,
44
+ "step": 6
45
+ },
46
+ {
47
+ "epoch": 0.05,
48
+ "learning_rate": 8.750000000000001e-06,
49
+ "loss": 0.6587,
50
+ "step": 7
51
+ },
52
+ {
53
+ "epoch": 0.06,
54
+ "learning_rate": 1e-05,
55
+ "loss": 0.6568,
56
+ "step": 8
57
+ },
58
+ {
59
+ "epoch": 0.07,
60
+ "learning_rate": 1.125e-05,
61
+ "loss": 0.6368,
62
+ "step": 9
63
+ },
64
+ {
65
+ "epoch": 0.08,
66
+ "learning_rate": 1.25e-05,
67
+ "loss": 0.6348,
68
+ "step": 10
69
+ },
70
+ {
71
+ "epoch": 0.08,
72
+ "learning_rate": 1.375e-05,
73
+ "loss": 0.6196,
74
+ "step": 11
75
+ },
76
+ {
77
+ "epoch": 0.09,
78
+ "learning_rate": 1.5000000000000002e-05,
79
+ "loss": 0.6126,
80
+ "step": 12
81
+ },
82
+ {
83
+ "epoch": 0.1,
84
+ "learning_rate": 1.6250000000000002e-05,
85
+ "loss": 0.6079,
86
+ "step": 13
87
+ },
88
+ {
89
+ "epoch": 0.11,
90
+ "learning_rate": 1.7500000000000002e-05,
91
+ "loss": 0.5893,
92
+ "step": 14
93
+ },
94
+ {
95
+ "epoch": 0.11,
96
+ "learning_rate": 1.8750000000000002e-05,
97
+ "loss": 0.567,
98
+ "step": 15
99
+ },
100
+ {
101
+ "epoch": 0.12,
102
+ "learning_rate": 2e-05,
103
+ "loss": 0.5891,
104
+ "step": 16
105
+ },
106
+ {
107
+ "epoch": 0.13,
108
+ "learning_rate": 1.9999658256641746e-05,
109
+ "loss": 0.5882,
110
+ "step": 17
111
+ },
112
+ {
113
+ "epoch": 0.14,
114
+ "learning_rate": 1.9998633049924693e-05,
115
+ "loss": 0.559,
116
+ "step": 18
117
+ },
118
+ {
119
+ "epoch": 0.14,
120
+ "learning_rate": 1.999692444992035e-05,
121
+ "loss": 0.5827,
122
+ "step": 19
123
+ },
124
+ {
125
+ "epoch": 0.15,
126
+ "learning_rate": 1.999453257340926e-05,
127
+ "loss": 0.5694,
128
+ "step": 20
129
+ },
130
+ {
131
+ "epoch": 0.16,
132
+ "learning_rate": 1.999145758387301e-05,
133
+ "loss": 0.5581,
134
+ "step": 21
135
+ },
136
+ {
137
+ "epoch": 0.17,
138
+ "learning_rate": 1.998769969148305e-05,
139
+ "loss": 0.5604,
140
+ "step": 22
141
+ },
142
+ {
143
+ "epoch": 0.17,
144
+ "learning_rate": 1.9983259153086328e-05,
145
+ "loss": 0.5556,
146
+ "step": 23
147
+ },
148
+ {
149
+ "epoch": 0.18,
150
+ "learning_rate": 1.9978136272187745e-05,
151
+ "loss": 0.5674,
152
+ "step": 24
153
+ },
154
+ {
155
+ "epoch": 0.19,
156
+ "learning_rate": 1.997233139892941e-05,
157
+ "loss": 0.5265,
158
+ "step": 25
159
+ },
160
+ {
161
+ "epoch": 0.2,
162
+ "learning_rate": 1.99658449300667e-05,
163
+ "loss": 0.5486,
164
+ "step": 26
165
+ },
166
+ {
167
+ "epoch": 0.2,
168
+ "learning_rate": 1.995867730894114e-05,
169
+ "loss": 0.5432,
170
+ "step": 27
171
+ },
172
+ {
173
+ "epoch": 0.21,
174
+ "learning_rate": 1.9950829025450116e-05,
175
+ "loss": 0.5616,
176
+ "step": 28
177
+ },
178
+ {
179
+ "epoch": 0.22,
180
+ "learning_rate": 1.9942300616013378e-05,
181
+ "loss": 0.5415,
182
+ "step": 29
183
+ },
184
+ {
185
+ "epoch": 0.23,
186
+ "learning_rate": 1.9933092663536384e-05,
187
+ "loss": 0.5493,
188
+ "step": 30
189
+ },
190
+ {
191
+ "epoch": 0.23,
192
+ "learning_rate": 1.992320579737045e-05,
193
+ "loss": 0.5608,
194
+ "step": 31
195
+ },
196
+ {
197
+ "epoch": 0.24,
198
+ "learning_rate": 1.9912640693269754e-05,
199
+ "loss": 0.5353,
200
+ "step": 32
201
+ },
202
+ {
203
+ "epoch": 0.25,
204
+ "learning_rate": 1.990139807334512e-05,
205
+ "loss": 0.5511,
206
+ "step": 33
207
+ },
208
+ {
209
+ "epoch": 0.26,
210
+ "learning_rate": 1.9889478706014687e-05,
211
+ "loss": 0.5331,
212
+ "step": 34
213
+ },
214
+ {
215
+ "epoch": 0.26,
216
+ "learning_rate": 1.9876883405951378e-05,
217
+ "loss": 0.5402,
218
+ "step": 35
219
+ },
220
+ {
221
+ "epoch": 0.27,
222
+ "learning_rate": 1.9863613034027224e-05,
223
+ "loss": 0.5244,
224
+ "step": 36
225
+ },
226
+ {
227
+ "epoch": 0.28,
228
+ "learning_rate": 1.984966849725452e-05,
229
+ "loss": 0.5238,
230
+ "step": 37
231
+ },
232
+ {
233
+ "epoch": 0.29,
234
+ "learning_rate": 1.9835050748723826e-05,
235
+ "loss": 0.5306,
236
+ "step": 38
237
+ },
238
+ {
239
+ "epoch": 0.29,
240
+ "learning_rate": 1.981976078753884e-05,
241
+ "loss": 0.5177,
242
+ "step": 39
243
+ },
244
+ {
245
+ "epoch": 0.3,
246
+ "learning_rate": 1.9803799658748096e-05,
247
+ "loss": 0.5193,
248
+ "step": 40
249
+ },
250
+ {
251
+ "epoch": 0.31,
252
+ "learning_rate": 1.9787168453273546e-05,
253
+ "loss": 0.4995,
254
+ "step": 41
255
+ },
256
+ {
257
+ "epoch": 0.32,
258
+ "learning_rate": 1.9769868307835996e-05,
259
+ "loss": 0.517,
260
+ "step": 42
261
+ },
262
+ {
263
+ "epoch": 0.32,
264
+ "learning_rate": 1.97519004048774e-05,
265
+ "loss": 0.5225,
266
+ "step": 43
267
+ },
268
+ {
269
+ "epoch": 0.33,
270
+ "learning_rate": 1.973326597248006e-05,
271
+ "loss": 0.5065,
272
+ "step": 44
273
+ },
274
+ {
275
+ "epoch": 0.34,
276
+ "learning_rate": 1.9713966284282677e-05,
277
+ "loss": 0.5198,
278
+ "step": 45
279
+ },
280
+ {
281
+ "epoch": 0.35,
282
+ "learning_rate": 1.9694002659393306e-05,
283
+ "loss": 0.5226,
284
+ "step": 46
285
+ },
286
+ {
287
+ "epoch": 0.35,
288
+ "learning_rate": 1.9673376462299186e-05,
289
+ "loss": 0.5022,
290
+ "step": 47
291
+ },
292
+ {
293
+ "epoch": 0.36,
294
+ "learning_rate": 1.9652089102773487e-05,
295
+ "loss": 0.513,
296
+ "step": 48
297
+ },
298
+ {
299
+ "epoch": 0.37,
300
+ "learning_rate": 1.963014203577896e-05,
301
+ "loss": 0.4919,
302
+ "step": 49
303
+ },
304
+ {
305
+ "epoch": 0.38,
306
+ "learning_rate": 1.9607536761368484e-05,
307
+ "loss": 0.5182,
308
+ "step": 50
309
+ },
310
+ {
311
+ "epoch": 0.38,
312
+ "learning_rate": 1.958427482458253e-05,
313
+ "loss": 0.5067,
314
+ "step": 51
315
+ },
316
+ {
317
+ "epoch": 0.39,
318
+ "learning_rate": 1.9560357815343577e-05,
319
+ "loss": 0.5196,
320
+ "step": 52
321
+ },
322
+ {
323
+ "epoch": 0.4,
324
+ "learning_rate": 1.9535787368347444e-05,
325
+ "loss": 0.5092,
326
+ "step": 53
327
+ },
328
+ {
329
+ "epoch": 0.41,
330
+ "learning_rate": 1.9510565162951538e-05,
331
+ "loss": 0.5075,
332
+ "step": 54
333
+ },
334
+ {
335
+ "epoch": 0.42,
336
+ "learning_rate": 1.9484692923060095e-05,
337
+ "loss": 0.4976,
338
+ "step": 55
339
+ },
340
+ {
341
+ "epoch": 0.42,
342
+ "learning_rate": 1.9458172417006347e-05,
343
+ "loss": 0.5013,
344
+ "step": 56
345
+ },
346
+ {
347
+ "epoch": 0.43,
348
+ "learning_rate": 1.9431005457431654e-05,
349
+ "loss": 0.4874,
350
+ "step": 57
351
+ },
352
+ {
353
+ "epoch": 0.44,
354
+ "learning_rate": 1.9403193901161614e-05,
355
+ "loss": 0.5119,
356
+ "step": 58
357
+ },
358
+ {
359
+ "epoch": 0.45,
360
+ "learning_rate": 1.9374739649079155e-05,
361
+ "loss": 0.4872,
362
+ "step": 59
363
+ },
364
+ {
365
+ "epoch": 0.45,
366
+ "learning_rate": 1.934564464599461e-05,
367
+ "loss": 0.4822,
368
+ "step": 60
369
+ },
370
+ {
371
+ "epoch": 0.46,
372
+ "learning_rate": 1.9315910880512792e-05,
373
+ "loss": 0.4976,
374
+ "step": 61
375
+ },
376
+ {
377
+ "epoch": 0.47,
378
+ "learning_rate": 1.9285540384897073e-05,
379
+ "loss": 0.495,
380
+ "step": 62
381
+ },
382
+ {
383
+ "epoch": 0.48,
384
+ "learning_rate": 1.9254535234930486e-05,
385
+ "loss": 0.4942,
386
+ "step": 63
387
+ },
388
+ {
389
+ "epoch": 0.48,
390
+ "learning_rate": 1.922289754977385e-05,
391
+ "loss": 0.4924,
392
+ "step": 64
393
+ },
394
+ {
395
+ "epoch": 0.49,
396
+ "learning_rate": 1.919062949182091e-05,
397
+ "loss": 0.478,
398
+ "step": 65
399
+ },
400
+ {
401
+ "epoch": 0.5,
402
+ "learning_rate": 1.9157733266550577e-05,
403
+ "loss": 0.4918,
404
+ "step": 66
405
+ },
406
+ {
407
+ "epoch": 0.51,
408
+ "learning_rate": 1.9124211122376138e-05,
409
+ "loss": 0.4908,
410
+ "step": 67
411
+ },
412
+ {
413
+ "epoch": 0.51,
414
+ "learning_rate": 1.909006535049163e-05,
415
+ "loss": 0.4915,
416
+ "step": 68
417
+ },
418
+ {
419
+ "epoch": 0.52,
420
+ "learning_rate": 1.9055298284715192e-05,
421
+ "loss": 0.5002,
422
+ "step": 69
423
+ },
424
+ {
425
+ "epoch": 0.53,
426
+ "learning_rate": 1.9019912301329593e-05,
427
+ "loss": 0.4956,
428
+ "step": 70
429
+ },
430
+ {
431
+ "epoch": 0.54,
432
+ "learning_rate": 1.898390981891979e-05,
433
+ "loss": 0.4728,
434
+ "step": 71
435
+ },
436
+ {
437
+ "epoch": 0.54,
438
+ "learning_rate": 1.8947293298207637e-05,
439
+ "loss": 0.471,
440
+ "step": 72
441
+ },
442
+ {
443
+ "epoch": 0.55,
444
+ "learning_rate": 1.891006524188368e-05,
445
+ "loss": 0.5036,
446
+ "step": 73
447
+ },
448
+ {
449
+ "epoch": 0.56,
450
+ "learning_rate": 1.887222819443612e-05,
451
+ "loss": 0.4837,
452
+ "step": 74
453
+ },
454
+ {
455
+ "epoch": 0.57,
456
+ "learning_rate": 1.883378474197689e-05,
457
+ "loss": 0.4758,
458
+ "step": 75
459
+ },
460
+ {
461
+ "epoch": 0.57,
462
+ "learning_rate": 1.879473751206489e-05,
463
+ "loss": 0.4789,
464
+ "step": 76
465
+ },
466
+ {
467
+ "epoch": 0.58,
468
+ "learning_rate": 1.875508917352643e-05,
469
+ "loss": 0.4798,
470
+ "step": 77
471
+ },
472
+ {
473
+ "epoch": 0.59,
474
+ "learning_rate": 1.8714842436272774e-05,
475
+ "loss": 0.4737,
476
+ "step": 78
477
+ },
478
+ {
479
+ "epoch": 0.6,
480
+ "learning_rate": 1.8674000051114953e-05,
481
+ "loss": 0.4618,
482
+ "step": 79
483
+ },
484
+ {
485
+ "epoch": 0.6,
486
+ "learning_rate": 1.863256480957574e-05,
487
+ "loss": 0.479,
488
+ "step": 80
489
+ },
490
+ {
491
+ "epoch": 0.61,
492
+ "learning_rate": 1.8590539543698852e-05,
493
+ "loss": 0.4616,
494
+ "step": 81
495
+ },
496
+ {
497
+ "epoch": 0.62,
498
+ "learning_rate": 1.854792712585539e-05,
499
+ "loss": 0.4825,
500
+ "step": 82
501
+ },
502
+ {
503
+ "epoch": 0.63,
504
+ "learning_rate": 1.8504730468547508e-05,
505
+ "loss": 0.478,
506
+ "step": 83
507
+ },
508
+ {
509
+ "epoch": 0.63,
510
+ "learning_rate": 1.8460952524209355e-05,
511
+ "loss": 0.4773,
512
+ "step": 84
513
+ },
514
+ {
515
+ "epoch": 0.64,
516
+ "learning_rate": 1.8416596285005274e-05,
517
+ "loss": 0.4626,
518
+ "step": 85
519
+ },
520
+ {
521
+ "epoch": 0.65,
522
+ "learning_rate": 1.8371664782625287e-05,
523
+ "loss": 0.498,
524
+ "step": 86
525
+ },
526
+ {
527
+ "epoch": 0.66,
528
+ "learning_rate": 1.8326161088077905e-05,
529
+ "loss": 0.4772,
530
+ "step": 87
531
+ },
532
+ {
533
+ "epoch": 0.66,
534
+ "learning_rate": 1.8280088311480203e-05,
535
+ "loss": 0.4749,
536
+ "step": 88
537
+ },
538
+ {
539
+ "epoch": 0.67,
540
+ "learning_rate": 1.823344960184526e-05,
541
+ "loss": 0.4575,
542
+ "step": 89
543
+ },
544
+ {
545
+ "epoch": 0.68,
546
+ "learning_rate": 1.8186248146866928e-05,
547
+ "loss": 0.4649,
548
+ "step": 90
549
+ },
550
+ {
551
+ "epoch": 0.69,
552
+ "learning_rate": 1.813848717270195e-05,
553
+ "loss": 0.4666,
554
+ "step": 91
555
+ },
556
+ {
557
+ "epoch": 0.69,
558
+ "learning_rate": 1.8090169943749477e-05,
559
+ "loss": 0.4606,
560
+ "step": 92
561
+ },
562
+ {
563
+ "epoch": 0.7,
564
+ "learning_rate": 1.804129976242792e-05,
565
+ "loss": 0.4658,
566
+ "step": 93
567
+ },
568
+ {
569
+ "epoch": 0.71,
570
+ "learning_rate": 1.7991879968949248e-05,
571
+ "loss": 0.4434,
572
+ "step": 94
573
+ },
574
+ {
575
+ "epoch": 0.72,
576
+ "learning_rate": 1.7941913941090712e-05,
577
+ "loss": 0.4583,
578
+ "step": 95
579
+ },
580
+ {
581
+ "epoch": 0.72,
582
+ "learning_rate": 1.789140509396394e-05,
583
+ "loss": 0.4769,
584
+ "step": 96
585
+ },
586
+ {
587
+ "epoch": 0.73,
588
+ "learning_rate": 1.784035687978153e-05,
589
+ "loss": 0.4477,
590
+ "step": 97
591
+ },
592
+ {
593
+ "epoch": 0.74,
594
+ "learning_rate": 1.7788772787621126e-05,
595
+ "loss": 0.4576,
596
+ "step": 98
597
+ },
598
+ {
599
+ "epoch": 0.75,
600
+ "learning_rate": 1.7736656343186897e-05,
601
+ "loss": 0.4457,
602
+ "step": 99
603
+ },
604
+ {
605
+ "epoch": 0.75,
606
+ "learning_rate": 1.7684011108568593e-05,
607
+ "loss": 0.4774,
608
+ "step": 100
609
+ },
610
+ {
611
+ "epoch": 0.76,
612
+ "learning_rate": 1.7630840681998068e-05,
613
+ "loss": 0.4716,
614
+ "step": 101
615
+ },
616
+ {
617
+ "epoch": 0.77,
618
+ "learning_rate": 1.757714869760335e-05,
619
+ "loss": 0.4352,
620
+ "step": 102
621
+ },
622
+ {
623
+ "epoch": 0.78,
624
+ "learning_rate": 1.752293882516025e-05,
625
+ "loss": 0.457,
626
+ "step": 103
627
+ },
628
+ {
629
+ "epoch": 0.78,
630
+ "learning_rate": 1.7468214769841542e-05,
631
+ "loss": 0.4525,
632
+ "step": 104
633
+ },
634
+ {
635
+ "epoch": 0.79,
636
+ "learning_rate": 1.7412980271963712e-05,
637
+ "loss": 0.4518,
638
+ "step": 105
639
+ },
640
+ {
641
+ "epoch": 0.8,
642
+ "learning_rate": 1.735723910673132e-05,
643
+ "loss": 0.478,
644
+ "step": 106
645
+ },
646
+ {
647
+ "epoch": 0.81,
648
+ "learning_rate": 1.7300995083978965e-05,
649
+ "loss": 0.4472,
650
+ "step": 107
651
+ },
652
+ {
653
+ "epoch": 0.82,
654
+ "learning_rate": 1.7244252047910893e-05,
655
+ "loss": 0.4443,
656
+ "step": 108
657
+ },
658
+ {
659
+ "epoch": 0.82,
660
+ "learning_rate": 1.718701387683824e-05,
661
+ "loss": 0.4216,
662
+ "step": 109
663
+ },
664
+ {
665
+ "epoch": 0.83,
666
+ "learning_rate": 1.7129284482913973e-05,
667
+ "loss": 0.4337,
668
+ "step": 110
669
+ },
670
+ {
671
+ "epoch": 0.84,
672
+ "learning_rate": 1.7071067811865477e-05,
673
+ "loss": 0.4494,
674
+ "step": 111
675
+ },
676
+ {
677
+ "epoch": 0.85,
678
+ "learning_rate": 1.7012367842724887e-05,
679
+ "loss": 0.4231,
680
+ "step": 112
681
+ },
682
+ {
683
+ "epoch": 0.85,
684
+ "learning_rate": 1.6953188587557122e-05,
685
+ "loss": 0.4385,
686
+ "step": 113
687
+ },
688
+ {
689
+ "epoch": 0.86,
690
+ "learning_rate": 1.6893534091185658e-05,
691
+ "loss": 0.4513,
692
+ "step": 114
693
+ },
694
+ {
695
+ "epoch": 0.87,
696
+ "learning_rate": 1.6833408430916085e-05,
697
+ "loss": 0.4532,
698
+ "step": 115
699
+ },
700
+ {
701
+ "epoch": 0.88,
702
+ "learning_rate": 1.6772815716257414e-05,
703
+ "loss": 0.4427,
704
+ "step": 116
705
+ },
706
+ {
707
+ "epoch": 0.88,
708
+ "learning_rate": 1.6711760088641197e-05,
709
+ "loss": 0.449,
710
+ "step": 117
711
+ },
712
+ {
713
+ "epoch": 0.89,
714
+ "learning_rate": 1.6650245721138483e-05,
715
+ "loss": 0.4229,
716
+ "step": 118
717
+ },
718
+ {
719
+ "epoch": 0.9,
720
+ "learning_rate": 1.658827681817458e-05,
721
+ "loss": 0.428,
722
+ "step": 119
723
+ },
724
+ {
725
+ "epoch": 0.91,
726
+ "learning_rate": 1.6525857615241686e-05,
727
+ "loss": 0.4522,
728
+ "step": 120
729
+ },
730
+ {
731
+ "epoch": 0.91,
732
+ "learning_rate": 1.646299237860941e-05,
733
+ "loss": 0.4357,
734
+ "step": 121
735
+ },
736
+ {
737
+ "epoch": 0.92,
738
+ "learning_rate": 1.6399685405033168e-05,
739
+ "loss": 0.4428,
740
+ "step": 122
741
+ },
742
+ {
743
+ "epoch": 0.93,
744
+ "learning_rate": 1.6335941021460507e-05,
745
+ "loss": 0.4299,
746
+ "step": 123
747
+ },
748
+ {
749
+ "epoch": 0.94,
750
+ "learning_rate": 1.6271763584735373e-05,
751
+ "loss": 0.4459,
752
+ "step": 124
753
+ },
754
+ {
755
+ "epoch": 0.94,
756
+ "learning_rate": 1.6207157481300315e-05,
757
+ "loss": 0.446,
758
+ "step": 125
759
+ },
760
+ {
761
+ "epoch": 0.95,
762
+ "learning_rate": 1.6142127126896682e-05,
763
+ "loss": 0.433,
764
+ "step": 126
765
+ },
766
+ {
767
+ "epoch": 0.96,
768
+ "learning_rate": 1.6076676966262815e-05,
769
+ "loss": 0.4322,
770
+ "step": 127
771
+ },
772
+ {
773
+ "epoch": 0.97,
774
+ "learning_rate": 1.6010811472830253e-05,
775
+ "loss": 0.4187,
776
+ "step": 128
777
+ },
778
+ {
779
+ "epoch": 0.97,
780
+ "learning_rate": 1.5944535148417982e-05,
781
+ "loss": 0.4231,
782
+ "step": 129
783
+ },
784
+ {
785
+ "epoch": 0.98,
786
+ "learning_rate": 1.5877852522924733e-05,
787
+ "loss": 0.4323,
788
+ "step": 130
789
+ },
790
+ {
791
+ "epoch": 0.99,
792
+ "learning_rate": 1.5810768154019386e-05,
793
+ "loss": 0.4322,
794
+ "step": 131
795
+ },
796
+ {
797
+ "epoch": 1.0,
798
+ "learning_rate": 1.5743286626829437e-05,
799
+ "loss": 0.4322,
800
+ "step": 132
801
+ },
802
+ {
803
+ "epoch": 1.0,
804
+ "learning_rate": 1.5675412553627638e-05,
805
+ "loss": 0.4192,
806
+ "step": 133
807
+ },
808
+ {
809
+ "epoch": 1.01,
810
+ "learning_rate": 1.560715057351673e-05,
811
+ "loss": 0.3991,
812
+ "step": 134
813
+ },
814
+ {
815
+ "epoch": 1.02,
816
+ "learning_rate": 1.5538505352112373e-05,
817
+ "loss": 0.4059,
818
+ "step": 135
819
+ },
820
+ {
821
+ "epoch": 1.03,
822
+ "learning_rate": 1.5469481581224274e-05,
823
+ "loss": 0.3876,
824
+ "step": 136
825
+ },
826
+ {
827
+ "epoch": 1.03,
828
+ "learning_rate": 1.5400083978535475e-05,
829
+ "loss": 0.4075,
830
+ "step": 137
831
+ },
832
+ {
833
+ "epoch": 1.04,
834
+ "learning_rate": 1.533031728727994e-05,
835
+ "loss": 0.4278,
836
+ "step": 138
837
+ },
838
+ {
839
+ "epoch": 1.05,
840
+ "learning_rate": 1.526018627591834e-05,
841
+ "loss": 0.4178,
842
+ "step": 139
843
+ },
844
+ {
845
+ "epoch": 1.06,
846
+ "learning_rate": 1.5189695737812153e-05,
847
+ "loss": 0.4035,
848
+ "step": 140
849
+ },
850
+ {
851
+ "epoch": 1.06,
852
+ "learning_rate": 1.5118850490896012e-05,
853
+ "loss": 0.3904,
854
+ "step": 141
855
+ },
856
+ {
857
+ "epoch": 1.07,
858
+ "learning_rate": 1.504765537734844e-05,
859
+ "loss": 0.3911,
860
+ "step": 142
861
+ },
862
+ {
863
+ "epoch": 1.08,
864
+ "learning_rate": 1.4976115263260876e-05,
865
+ "loss": 0.3752,
866
+ "step": 143
867
+ },
868
+ {
869
+ "epoch": 1.09,
870
+ "learning_rate": 1.4904235038305084e-05,
871
+ "loss": 0.4027,
872
+ "step": 144
873
+ },
874
+ {
875
+ "epoch": 1.09,
876
+ "learning_rate": 1.4832019615398962e-05,
877
+ "loss": 0.3846,
878
+ "step": 145
879
+ },
880
+ {
881
+ "epoch": 1.1,
882
+ "learning_rate": 1.4759473930370738e-05,
883
+ "loss": 0.3962,
884
+ "step": 146
885
+ },
886
+ {
887
+ "epoch": 1.11,
888
+ "learning_rate": 1.4686602941621618e-05,
889
+ "loss": 0.3967,
890
+ "step": 147
891
+ },
892
+ {
893
+ "epoch": 1.12,
894
+ "learning_rate": 1.461341162978688e-05,
895
+ "loss": 0.392,
896
+ "step": 148
897
+ },
898
+ {
899
+ "epoch": 1.12,
900
+ "learning_rate": 1.4539904997395468e-05,
901
+ "loss": 0.4096,
902
+ "step": 149
903
+ },
904
+ {
905
+ "epoch": 1.13,
906
+ "learning_rate": 1.4466088068528068e-05,
907
+ "loss": 0.4014,
908
+ "step": 150
909
+ },
910
+ {
911
+ "epoch": 1.14,
912
+ "learning_rate": 1.4391965888473705e-05,
913
+ "loss": 0.3856,
914
+ "step": 151
915
+ },
916
+ {
917
+ "epoch": 1.15,
918
+ "learning_rate": 1.4317543523384928e-05,
919
+ "loss": 0.3677,
920
+ "step": 152
921
+ },
922
+ {
923
+ "epoch": 1.15,
924
+ "learning_rate": 1.4242826059931538e-05,
925
+ "loss": 0.3858,
926
+ "step": 153
927
+ },
928
+ {
929
+ "epoch": 1.16,
930
+ "learning_rate": 1.4167818604952906e-05,
931
+ "loss": 0.3814,
932
+ "step": 154
933
+ },
934
+ {
935
+ "epoch": 1.17,
936
+ "learning_rate": 1.409252628510894e-05,
937
+ "loss": 0.4017,
938
+ "step": 155
939
+ },
940
+ {
941
+ "epoch": 1.18,
942
+ "learning_rate": 1.4016954246529697e-05,
943
+ "loss": 0.3981,
944
+ "step": 156
945
+ },
946
+ {
947
+ "epoch": 1.18,
948
+ "learning_rate": 1.3941107654463619e-05,
949
+ "loss": 0.3867,
950
+ "step": 157
951
+ },
952
+ {
953
+ "epoch": 1.19,
954
+ "learning_rate": 1.3864991692924524e-05,
955
+ "loss": 0.3948,
956
+ "step": 158
957
+ },
958
+ {
959
+ "epoch": 1.2,
960
+ "learning_rate": 1.3788611564337277e-05,
961
+ "loss": 0.3747,
962
+ "step": 159
963
+ },
964
+ {
965
+ "epoch": 1.21,
966
+ "learning_rate": 1.3711972489182208e-05,
967
+ "loss": 0.3866,
968
+ "step": 160
969
+ },
970
+ {
971
+ "epoch": 1.22,
972
+ "learning_rate": 1.3635079705638298e-05,
973
+ "loss": 0.3982,
974
+ "step": 161
975
+ },
976
+ {
977
+ "epoch": 1.22,
978
+ "learning_rate": 1.3557938469225167e-05,
979
+ "loss": 0.3835,
980
+ "step": 162
981
+ },
982
+ {
983
+ "epoch": 1.23,
984
+ "learning_rate": 1.3480554052443847e-05,
985
+ "loss": 0.3955,
986
+ "step": 163
987
+ },
988
+ {
989
+ "epoch": 1.24,
990
+ "learning_rate": 1.3402931744416432e-05,
991
+ "loss": 0.396,
992
+ "step": 164
993
+ },
994
+ {
995
+ "epoch": 1.25,
996
+ "learning_rate": 1.332507685052457e-05,
997
+ "loss": 0.3615,
998
+ "step": 165
999
+ },
1000
+ {
1001
+ "epoch": 1.25,
1002
+ "learning_rate": 1.3246994692046837e-05,
1003
+ "loss": 0.3688,
1004
+ "step": 166
1005
+ },
1006
+ {
1007
+ "epoch": 1.26,
1008
+ "learning_rate": 1.3168690605795044e-05,
1009
+ "loss": 0.379,
1010
+ "step": 167
1011
+ },
1012
+ {
1013
+ "epoch": 1.27,
1014
+ "learning_rate": 1.3090169943749475e-05,
1015
+ "loss": 0.3919,
1016
+ "step": 168
1017
+ },
1018
+ {
1019
+ "epoch": 1.28,
1020
+ "learning_rate": 1.3011438072693077e-05,
1021
+ "loss": 0.379,
1022
+ "step": 169
1023
+ },
1024
+ {
1025
+ "epoch": 1.28,
1026
+ "learning_rate": 1.293250037384465e-05,
1027
+ "loss": 0.3981,
1028
+ "step": 170
1029
+ },
1030
+ {
1031
+ "epoch": 1.29,
1032
+ "learning_rate": 1.2853362242491054e-05,
1033
+ "loss": 0.3692,
1034
+ "step": 171
1035
+ },
1036
+ {
1037
+ "epoch": 1.3,
1038
+ "learning_rate": 1.2774029087618448e-05,
1039
+ "loss": 0.3874,
1040
+ "step": 172
1041
+ },
1042
+ {
1043
+ "epoch": 1.31,
1044
+ "learning_rate": 1.269450633154258e-05,
1045
+ "loss": 0.38,
1046
+ "step": 173
1047
+ },
1048
+ {
1049
+ "epoch": 1.31,
1050
+ "learning_rate": 1.26147994095382e-05,
1051
+ "loss": 0.3664,
1052
+ "step": 174
1053
+ },
1054
+ {
1055
+ "epoch": 1.32,
1056
+ "learning_rate": 1.253491376946754e-05,
1057
+ "loss": 0.3683,
1058
+ "step": 175
1059
+ },
1060
+ {
1061
+ "epoch": 1.33,
1062
+ "learning_rate": 1.2454854871407993e-05,
1063
+ "loss": 0.3811,
1064
+ "step": 176
1065
+ },
1066
+ {
1067
+ "epoch": 1.34,
1068
+ "learning_rate": 1.2374628187278888e-05,
1069
+ "loss": 0.3797,
1070
+ "step": 177
1071
+ },
1072
+ {
1073
+ "epoch": 1.34,
1074
+ "learning_rate": 1.2294239200467516e-05,
1075
+ "loss": 0.3807,
1076
+ "step": 178
1077
+ },
1078
+ {
1079
+ "epoch": 1.35,
1080
+ "learning_rate": 1.2213693405454345e-05,
1081
+ "loss": 0.3739,
1082
+ "step": 179
1083
+ },
1084
+ {
1085
+ "epoch": 1.36,
1086
+ "learning_rate": 1.213299630743747e-05,
1087
+ "loss": 0.368,
1088
+ "step": 180
1089
+ },
1090
+ {
1091
+ "epoch": 1.37,
1092
+ "learning_rate": 1.2052153421956343e-05,
1093
+ "loss": 0.3832,
1094
+ "step": 181
1095
+ },
1096
+ {
1097
+ "epoch": 1.37,
1098
+ "learning_rate": 1.1971170274514802e-05,
1099
+ "loss": 0.3755,
1100
+ "step": 182
1101
+ },
1102
+ {
1103
+ "epoch": 1.38,
1104
+ "learning_rate": 1.1890052400203405e-05,
1105
+ "loss": 0.3721,
1106
+ "step": 183
1107
+ },
1108
+ {
1109
+ "epoch": 1.39,
1110
+ "learning_rate": 1.1808805343321102e-05,
1111
+ "loss": 0.3693,
1112
+ "step": 184
1113
+ },
1114
+ {
1115
+ "epoch": 1.4,
1116
+ "learning_rate": 1.1727434656996306e-05,
1117
+ "loss": 0.3726,
1118
+ "step": 185
1119
+ },
1120
+ {
1121
+ "epoch": 1.4,
1122
+ "learning_rate": 1.164594590280734e-05,
1123
+ "loss": 0.3628,
1124
+ "step": 186
1125
+ },
1126
+ {
1127
+ "epoch": 1.41,
1128
+ "learning_rate": 1.156434465040231e-05,
1129
+ "loss": 0.3836,
1130
+ "step": 187
1131
+ },
1132
+ {
1133
+ "epoch": 1.42,
1134
+ "learning_rate": 1.148263647711842e-05,
1135
+ "loss": 0.3655,
1136
+ "step": 188
1137
+ },
1138
+ {
1139
+ "epoch": 1.43,
1140
+ "learning_rate": 1.140082696760078e-05,
1141
+ "loss": 0.3545,
1142
+ "step": 189
1143
+ },
1144
+ {
1145
+ "epoch": 1.43,
1146
+ "learning_rate": 1.1318921713420691e-05,
1147
+ "loss": 0.3729,
1148
+ "step": 190
1149
+ },
1150
+ {
1151
+ "epoch": 1.44,
1152
+ "learning_rate": 1.123692631269348e-05,
1153
+ "loss": 0.3612,
1154
+ "step": 191
1155
+ },
1156
+ {
1157
+ "epoch": 1.45,
1158
+ "learning_rate": 1.1154846369695864e-05,
1159
+ "loss": 0.3791,
1160
+ "step": 192
1161
+ },
1162
+ {
1163
+ "epoch": 1.46,
1164
+ "learning_rate": 1.107268749448292e-05,
1165
+ "loss": 0.3442,
1166
+ "step": 193
1167
+ },
1168
+ {
1169
+ "epoch": 1.46,
1170
+ "learning_rate": 1.099045530250463e-05,
1171
+ "loss": 0.3685,
1172
+ "step": 194
1173
+ },
1174
+ {
1175
+ "epoch": 1.47,
1176
+ "learning_rate": 1.0908155414222083e-05,
1177
+ "loss": 0.3605,
1178
+ "step": 195
1179
+ },
1180
+ {
1181
+ "epoch": 1.48,
1182
+ "learning_rate": 1.0825793454723325e-05,
1183
+ "loss": 0.3586,
1184
+ "step": 196
1185
+ },
1186
+ {
1187
+ "epoch": 1.49,
1188
+ "learning_rate": 1.0743375053338879e-05,
1189
+ "loss": 0.3606,
1190
+ "step": 197
1191
+ },
1192
+ {
1193
+ "epoch": 1.49,
1194
+ "learning_rate": 1.0660905843256995e-05,
1195
+ "loss": 0.3685,
1196
+ "step": 198
1197
+ },
1198
+ {
1199
+ "epoch": 1.51,
1200
+ "learning_rate": 1.0578391461138642e-05,
1201
+ "loss": 0.3423,
1202
+ "step": 199
1203
+ },
1204
+ {
1205
+ "epoch": 1.51,
1206
+ "learning_rate": 1.0495837546732224e-05,
1207
+ "loss": 0.3627,
1208
+ "step": 200
1209
+ },
1210
+ {
1211
+ "epoch": 1.52,
1212
+ "learning_rate": 1.0413249742488132e-05,
1213
+ "loss": 0.3553,
1214
+ "step": 201
1215
+ },
1216
+ {
1217
+ "epoch": 1.53,
1218
+ "learning_rate": 1.0330633693173083e-05,
1219
+ "loss": 0.3589,
1220
+ "step": 202
1221
+ },
1222
+ {
1223
+ "epoch": 1.54,
1224
+ "learning_rate": 1.0247995045484303e-05,
1225
+ "loss": 0.3459,
1226
+ "step": 203
1227
+ },
1228
+ {
1229
+ "epoch": 1.54,
1230
+ "learning_rate": 1.0165339447663586e-05,
1231
+ "loss": 0.3685,
1232
+ "step": 204
1233
+ },
1234
+ {
1235
+ "epoch": 1.55,
1236
+ "learning_rate": 1.008267254911125e-05,
1237
+ "loss": 0.362,
1238
+ "step": 205
1239
+ },
1240
+ {
1241
+ "epoch": 1.56,
1242
+ "learning_rate": 1e-05,
1243
+ "loss": 0.3458,
1244
+ "step": 206
1245
+ },
1246
+ {
1247
+ "epoch": 1.57,
1248
+ "learning_rate": 9.917327450888751e-06,
1249
+ "loss": 0.3453,
1250
+ "step": 207
1251
+ },
1252
+ {
1253
+ "epoch": 1.57,
1254
+ "learning_rate": 9.834660552336415e-06,
1255
+ "loss": 0.3375,
1256
+ "step": 208
1257
+ },
1258
+ {
1259
+ "epoch": 1.58,
1260
+ "learning_rate": 9.7520049545157e-06,
1261
+ "loss": 0.3785,
1262
+ "step": 209
1263
+ },
1264
+ {
1265
+ "epoch": 1.59,
1266
+ "learning_rate": 9.669366306826919e-06,
1267
+ "loss": 0.3649,
1268
+ "step": 210
1269
+ },
1270
+ {
1271
+ "epoch": 1.6,
1272
+ "learning_rate": 9.586750257511868e-06,
1273
+ "loss": 0.3462,
1274
+ "step": 211
1275
+ },
1276
+ {
1277
+ "epoch": 1.6,
1278
+ "learning_rate": 9.504162453267776e-06,
1279
+ "loss": 0.3458,
1280
+ "step": 212
1281
+ },
1282
+ {
1283
+ "epoch": 1.61,
1284
+ "learning_rate": 9.421608538861361e-06,
1285
+ "loss": 0.3471,
1286
+ "step": 213
1287
+ },
1288
+ {
1289
+ "epoch": 1.62,
1290
+ "learning_rate": 9.339094156743007e-06,
1291
+ "loss": 0.3649,
1292
+ "step": 214
1293
+ },
1294
+ {
1295
+ "epoch": 1.63,
1296
+ "learning_rate": 9.256624946661126e-06,
1297
+ "loss": 0.3558,
1298
+ "step": 215
1299
+ },
1300
+ {
1301
+ "epoch": 1.63,
1302
+ "learning_rate": 9.174206545276678e-06,
1303
+ "loss": 0.3356,
1304
+ "step": 216
1305
+ },
1306
+ {
1307
+ "epoch": 1.64,
1308
+ "learning_rate": 9.091844585777919e-06,
1309
+ "loss": 0.3252,
1310
+ "step": 217
1311
+ },
1312
+ {
1313
+ "epoch": 1.65,
1314
+ "learning_rate": 9.009544697495373e-06,
1315
+ "loss": 0.3354,
1316
+ "step": 218
1317
+ },
1318
+ {
1319
+ "epoch": 1.66,
1320
+ "learning_rate": 8.927312505517086e-06,
1321
+ "loss": 0.323,
1322
+ "step": 219
1323
+ },
1324
+ {
1325
+ "epoch": 1.66,
1326
+ "learning_rate": 8.84515363030414e-06,
1327
+ "loss": 0.3416,
1328
+ "step": 220
1329
+ },
1330
+ {
1331
+ "epoch": 1.67,
1332
+ "learning_rate": 8.763073687306523e-06,
1333
+ "loss": 0.3422,
1334
+ "step": 221
1335
+ },
1336
+ {
1337
+ "epoch": 1.68,
1338
+ "learning_rate": 8.68107828657931e-06,
1339
+ "loss": 0.3504,
1340
+ "step": 222
1341
+ },
1342
+ {
1343
+ "epoch": 1.69,
1344
+ "learning_rate": 8.599173032399222e-06,
1345
+ "loss": 0.3416,
1346
+ "step": 223
1347
+ },
1348
+ {
1349
+ "epoch": 1.69,
1350
+ "learning_rate": 8.51736352288158e-06,
1351
+ "loss": 0.3433,
1352
+ "step": 224
1353
+ },
1354
+ {
1355
+ "epoch": 1.7,
1356
+ "learning_rate": 8.43565534959769e-06,
1357
+ "loss": 0.3428,
1358
+ "step": 225
1359
+ },
1360
+ {
1361
+ "epoch": 1.71,
1362
+ "learning_rate": 8.35405409719266e-06,
1363
+ "loss": 0.3247,
1364
+ "step": 226
1365
+ },
1366
+ {
1367
+ "epoch": 1.72,
1368
+ "learning_rate": 8.2725653430037e-06,
1369
+ "loss": 0.3468,
1370
+ "step": 227
1371
+ },
1372
+ {
1373
+ "epoch": 1.72,
1374
+ "learning_rate": 8.191194656678905e-06,
1375
+ "loss": 0.3345,
1376
+ "step": 228
1377
+ },
1378
+ {
1379
+ "epoch": 1.73,
1380
+ "learning_rate": 8.109947599796599e-06,
1381
+ "loss": 0.3538,
1382
+ "step": 229
1383
+ },
1384
+ {
1385
+ "epoch": 1.74,
1386
+ "learning_rate": 8.0288297254852e-06,
1387
+ "loss": 0.3226,
1388
+ "step": 230
1389
+ },
1390
+ {
1391
+ "epoch": 1.75,
1392
+ "learning_rate": 7.947846578043658e-06,
1393
+ "loss": 0.3383,
1394
+ "step": 231
1395
+ },
1396
+ {
1397
+ "epoch": 1.75,
1398
+ "learning_rate": 7.867003692562533e-06,
1399
+ "loss": 0.3451,
1400
+ "step": 232
1401
+ },
1402
+ {
1403
+ "epoch": 1.76,
1404
+ "learning_rate": 7.786306594545658e-06,
1405
+ "loss": 0.3489,
1406
+ "step": 233
1407
+ },
1408
+ {
1409
+ "epoch": 1.77,
1410
+ "learning_rate": 7.705760799532485e-06,
1411
+ "loss": 0.3336,
1412
+ "step": 234
1413
+ },
1414
+ {
1415
+ "epoch": 1.78,
1416
+ "learning_rate": 7.625371812721115e-06,
1417
+ "loss": 0.343,
1418
+ "step": 235
1419
+ },
1420
+ {
1421
+ "epoch": 1.78,
1422
+ "learning_rate": 7.545145128592009e-06,
1423
+ "loss": 0.3559,
1424
+ "step": 236
1425
+ },
1426
+ {
1427
+ "epoch": 1.79,
1428
+ "learning_rate": 7.46508623053246e-06,
1429
+ "loss": 0.3399,
1430
+ "step": 237
1431
+ },
1432
+ {
1433
+ "epoch": 1.8,
1434
+ "learning_rate": 7.385200590461803e-06,
1435
+ "loss": 0.3366,
1436
+ "step": 238
1437
+ },
1438
+ {
1439
+ "epoch": 1.81,
1440
+ "learning_rate": 7.305493668457421e-06,
1441
+ "loss": 0.3328,
1442
+ "step": 239
1443
+ },
1444
+ {
1445
+ "epoch": 1.82,
1446
+ "learning_rate": 7.225970912381557e-06,
1447
+ "loss": 0.3518,
1448
+ "step": 240
1449
+ },
1450
+ {
1451
+ "epoch": 1.82,
1452
+ "learning_rate": 7.14663775750895e-06,
1453
+ "loss": 0.3355,
1454
+ "step": 241
1455
+ },
1456
+ {
1457
+ "epoch": 1.83,
1458
+ "learning_rate": 7.067499626155354e-06,
1459
+ "loss": 0.309,
1460
+ "step": 242
1461
+ },
1462
+ {
1463
+ "epoch": 1.84,
1464
+ "learning_rate": 6.988561927306927e-06,
1465
+ "loss": 0.3517,
1466
+ "step": 243
1467
+ },
1468
+ {
1469
+ "epoch": 1.85,
1470
+ "learning_rate": 6.909830056250527e-06,
1471
+ "loss": 0.3139,
1472
+ "step": 244
1473
+ },
1474
+ {
1475
+ "epoch": 1.85,
1476
+ "learning_rate": 6.831309394204957e-06,
1477
+ "loss": 0.3344,
1478
+ "step": 245
1479
+ },
1480
+ {
1481
+ "epoch": 1.86,
1482
+ "learning_rate": 6.7530053079531664e-06,
1483
+ "loss": 0.3276,
1484
+ "step": 246
1485
+ },
1486
+ {
1487
+ "epoch": 1.87,
1488
+ "learning_rate": 6.674923149475433e-06,
1489
+ "loss": 0.3427,
1490
+ "step": 247
1491
+ },
1492
+ {
1493
+ "epoch": 1.88,
1494
+ "learning_rate": 6.59706825558357e-06,
1495
+ "loss": 0.3124,
1496
+ "step": 248
1497
+ },
1498
+ {
1499
+ "epoch": 1.88,
1500
+ "learning_rate": 6.519445947556156e-06,
1501
+ "loss": 0.3219,
1502
+ "step": 249
1503
+ },
1504
+ {
1505
+ "epoch": 1.89,
1506
+ "learning_rate": 6.442061530774835e-06,
1507
+ "loss": 0.3247,
1508
+ "step": 250
1509
+ },
1510
+ {
1511
+ "epoch": 1.9,
1512
+ "learning_rate": 6.364920294361701e-06,
1513
+ "loss": 0.323,
1514
+ "step": 251
1515
+ },
1516
+ {
1517
+ "epoch": 1.91,
1518
+ "learning_rate": 6.2880275108177915e-06,
1519
+ "loss": 0.3198,
1520
+ "step": 252
1521
+ },
1522
+ {
1523
+ "epoch": 1.91,
1524
+ "learning_rate": 6.211388435662722e-06,
1525
+ "loss": 0.3259,
1526
+ "step": 253
1527
+ },
1528
+ {
1529
+ "epoch": 1.92,
1530
+ "learning_rate": 6.13500830707548e-06,
1531
+ "loss": 0.3237,
1532
+ "step": 254
1533
+ },
1534
+ {
1535
+ "epoch": 1.93,
1536
+ "learning_rate": 6.058892345536387e-06,
1537
+ "loss": 0.3235,
1538
+ "step": 255
1539
+ },
1540
+ {
1541
+ "epoch": 1.94,
1542
+ "learning_rate": 5.983045753470308e-06,
1543
+ "loss": 0.3133,
1544
+ "step": 256
1545
+ },
1546
+ {
1547
+ "epoch": 1.94,
1548
+ "learning_rate": 5.907473714891061e-06,
1549
+ "loss": 0.3261,
1550
+ "step": 257
1551
+ },
1552
+ {
1553
+ "epoch": 1.95,
1554
+ "learning_rate": 5.832181395047099e-06,
1555
+ "loss": 0.3524,
1556
+ "step": 258
1557
+ },
1558
+ {
1559
+ "epoch": 1.96,
1560
+ "learning_rate": 5.7571739400684644e-06,
1561
+ "loss": 0.3447,
1562
+ "step": 259
1563
+ },
1564
+ {
1565
+ "epoch": 1.97,
1566
+ "learning_rate": 5.6824564766150724e-06,
1567
+ "loss": 0.3377,
1568
+ "step": 260
1569
+ },
1570
+ {
1571
+ "epoch": 1.97,
1572
+ "learning_rate": 5.608034111526298e-06,
1573
+ "loss": 0.3128,
1574
+ "step": 261
1575
+ },
1576
+ {
1577
+ "epoch": 1.98,
1578
+ "learning_rate": 5.533911931471936e-06,
1579
+ "loss": 0.3247,
1580
+ "step": 262
1581
+ },
1582
+ {
1583
+ "epoch": 1.99,
1584
+ "learning_rate": 5.460095002604533e-06,
1585
+ "loss": 0.3273,
1586
+ "step": 263
1587
+ },
1588
+ {
1589
+ "epoch": 2.0,
1590
+ "learning_rate": 5.386588370213124e-06,
1591
+ "loss": 0.334,
1592
+ "step": 264
1593
+ },
1594
+ {
1595
+ "epoch": 2.0,
1596
+ "learning_rate": 5.3133970583783865e-06,
1597
+ "loss": 0.3001,
1598
+ "step": 265
1599
+ },
1600
+ {
1601
+ "epoch": 2.01,
1602
+ "learning_rate": 5.240526069629265e-06,
1603
+ "loss": 0.3199,
1604
+ "step": 266
1605
+ },
1606
+ {
1607
+ "epoch": 2.02,
1608
+ "learning_rate": 5.167980384601041e-06,
1609
+ "loss": 0.3307,
1610
+ "step": 267
1611
+ },
1612
+ {
1613
+ "epoch": 2.03,
1614
+ "learning_rate": 5.095764961694923e-06,
1615
+ "loss": 0.2982,
1616
+ "step": 268
1617
+ },
1618
+ {
1619
+ "epoch": 2.03,
1620
+ "learning_rate": 5.023884736739132e-06,
1621
+ "loss": 0.3142,
1622
+ "step": 269
1623
+ },
1624
+ {
1625
+ "epoch": 2.04,
1626
+ "learning_rate": 4.952344622651566e-06,
1627
+ "loss": 0.3093,
1628
+ "step": 270
1629
+ },
1630
+ {
1631
+ "epoch": 2.05,
1632
+ "learning_rate": 4.881149509103993e-06,
1633
+ "loss": 0.2976,
1634
+ "step": 271
1635
+ },
1636
+ {
1637
+ "epoch": 2.06,
1638
+ "learning_rate": 4.8103042621878515e-06,
1639
+ "loss": 0.3136,
1640
+ "step": 272
1641
+ },
1642
+ {
1643
+ "epoch": 2.06,
1644
+ "learning_rate": 4.739813724081661e-06,
1645
+ "loss": 0.2955,
1646
+ "step": 273
1647
+ },
1648
+ {
1649
+ "epoch": 2.07,
1650
+ "learning_rate": 4.669682712720065e-06,
1651
+ "loss": 0.2928,
1652
+ "step": 274
1653
+ },
1654
+ {
1655
+ "epoch": 2.08,
1656
+ "learning_rate": 4.599916021464531e-06,
1657
+ "loss": 0.2877,
1658
+ "step": 275
1659
+ },
1660
+ {
1661
+ "epoch": 2.09,
1662
+ "learning_rate": 4.530518418775734e-06,
1663
+ "loss": 0.2912,
1664
+ "step": 276
1665
+ },
1666
+ {
1667
+ "epoch": 2.09,
1668
+ "learning_rate": 4.4614946478876305e-06,
1669
+ "loss": 0.2907,
1670
+ "step": 277
1671
+ },
1672
+ {
1673
+ "epoch": 2.1,
1674
+ "learning_rate": 4.392849426483275e-06,
1675
+ "loss": 0.3007,
1676
+ "step": 278
1677
+ },
1678
+ {
1679
+ "epoch": 2.11,
1680
+ "learning_rate": 4.324587446372365e-06,
1681
+ "loss": 0.2926,
1682
+ "step": 279
1683
+ },
1684
+ {
1685
+ "epoch": 2.12,
1686
+ "learning_rate": 4.256713373170565e-06,
1687
+ "loss": 0.283,
1688
+ "step": 280
1689
+ },
1690
+ {
1691
+ "epoch": 2.12,
1692
+ "learning_rate": 4.189231845980618e-06,
1693
+ "loss": 0.2867,
1694
+ "step": 281
1695
+ },
1696
+ {
1697
+ "epoch": 2.13,
1698
+ "learning_rate": 4.12214747707527e-06,
1699
+ "loss": 0.2894,
1700
+ "step": 282
1701
+ },
1702
+ {
1703
+ "epoch": 2.14,
1704
+ "learning_rate": 4.055464851582022e-06,
1705
+ "loss": 0.3048,
1706
+ "step": 283
1707
+ },
1708
+ {
1709
+ "epoch": 2.15,
1710
+ "learning_rate": 3.989188527169749e-06,
1711
+ "loss": 0.2936,
1712
+ "step": 284
1713
+ },
1714
+ {
1715
+ "epoch": 2.15,
1716
+ "learning_rate": 3.923323033737188e-06,
1717
+ "loss": 0.2986,
1718
+ "step": 285
1719
+ },
1720
+ {
1721
+ "epoch": 2.16,
1722
+ "learning_rate": 3.857872873103322e-06,
1723
+ "loss": 0.2929,
1724
+ "step": 286
1725
+ },
1726
+ {
1727
+ "epoch": 2.17,
1728
+ "learning_rate": 3.792842518699689e-06,
1729
+ "loss": 0.292,
1730
+ "step": 287
1731
+ },
1732
+ {
1733
+ "epoch": 2.18,
1734
+ "learning_rate": 3.72823641526463e-06,
1735
+ "loss": 0.3029,
1736
+ "step": 288
1737
+ },
1738
+ {
1739
+ "epoch": 2.18,
1740
+ "learning_rate": 3.6640589785394955e-06,
1741
+ "loss": 0.2779,
1742
+ "step": 289
1743
+ },
1744
+ {
1745
+ "epoch": 2.19,
1746
+ "learning_rate": 3.6003145949668338e-06,
1747
+ "loss": 0.2852,
1748
+ "step": 290
1749
+ },
1750
+ {
1751
+ "epoch": 2.2,
1752
+ "learning_rate": 3.5370076213905904e-06,
1753
+ "loss": 0.2813,
1754
+ "step": 291
1755
+ },
1756
+ {
1757
+ "epoch": 2.21,
1758
+ "learning_rate": 3.4741423847583134e-06,
1759
+ "loss": 0.3127,
1760
+ "step": 292
1761
+ },
1762
+ {
1763
+ "epoch": 2.22,
1764
+ "learning_rate": 3.4117231818254205e-06,
1765
+ "loss": 0.3079,
1766
+ "step": 293
1767
+ },
1768
+ {
1769
+ "epoch": 2.22,
1770
+ "learning_rate": 3.349754278861517e-06,
1771
+ "loss": 0.2902,
1772
+ "step": 294
1773
+ },
1774
+ {
1775
+ "epoch": 2.23,
1776
+ "learning_rate": 3.288239911358807e-06,
1777
+ "loss": 0.3058,
1778
+ "step": 295
1779
+ },
1780
+ {
1781
+ "epoch": 2.24,
1782
+ "learning_rate": 3.2271842837425917e-06,
1783
+ "loss": 0.3047,
1784
+ "step": 296
1785
+ },
1786
+ {
1787
+ "epoch": 2.25,
1788
+ "learning_rate": 3.1665915690839165e-06,
1789
+ "loss": 0.2896,
1790
+ "step": 297
1791
+ },
1792
+ {
1793
+ "epoch": 2.25,
1794
+ "learning_rate": 3.1064659088143424e-06,
1795
+ "loss": 0.2874,
1796
+ "step": 298
1797
+ },
1798
+ {
1799
+ "epoch": 2.26,
1800
+ "learning_rate": 3.0468114124428806e-06,
1801
+ "loss": 0.3135,
1802
+ "step": 299
1803
+ },
1804
+ {
1805
+ "epoch": 2.27,
1806
+ "learning_rate": 2.9876321572751143e-06,
1807
+ "loss": 0.2885,
1808
+ "step": 300
1809
+ },
1810
+ {
1811
+ "epoch": 2.28,
1812
+ "learning_rate": 2.9289321881345257e-06,
1813
+ "loss": 0.3046,
1814
+ "step": 301
1815
+ },
1816
+ {
1817
+ "epoch": 2.28,
1818
+ "learning_rate": 2.8707155170860303e-06,
1819
+ "loss": 0.3039,
1820
+ "step": 302
1821
+ },
1822
+ {
1823
+ "epoch": 2.29,
1824
+ "learning_rate": 2.812986123161762e-06,
1825
+ "loss": 0.2818,
1826
+ "step": 303
1827
+ },
1828
+ {
1829
+ "epoch": 2.3,
1830
+ "learning_rate": 2.7557479520891104e-06,
1831
+ "loss": 0.2984,
1832
+ "step": 304
1833
+ },
1834
+ {
1835
+ "epoch": 2.31,
1836
+ "learning_rate": 2.6990049160210386e-06,
1837
+ "loss": 0.3234,
1838
+ "step": 305
1839
+ },
1840
+ {
1841
+ "epoch": 2.31,
1842
+ "learning_rate": 2.642760893268684e-06,
1843
+ "loss": 0.2838,
1844
+ "step": 306
1845
+ },
1846
+ {
1847
+ "epoch": 2.32,
1848
+ "learning_rate": 2.587019728036292e-06,
1849
+ "loss": 0.2963,
1850
+ "step": 307
1851
+ },
1852
+ {
1853
+ "epoch": 2.33,
1854
+ "learning_rate": 2.5317852301584642e-06,
1855
+ "loss": 0.2988,
1856
+ "step": 308
1857
+ },
1858
+ {
1859
+ "epoch": 2.34,
1860
+ "learning_rate": 2.4770611748397556e-06,
1861
+ "loss": 0.2874,
1862
+ "step": 309
1863
+ },
1864
+ {
1865
+ "epoch": 2.34,
1866
+ "learning_rate": 2.422851302396655e-06,
1867
+ "loss": 0.288,
1868
+ "step": 310
1869
+ },
1870
+ {
1871
+ "epoch": 2.35,
1872
+ "learning_rate": 2.369159318001937e-06,
1873
+ "loss": 0.2732,
1874
+ "step": 311
1875
+ },
1876
+ {
1877
+ "epoch": 2.36,
1878
+ "learning_rate": 2.315988891431412e-06,
1879
+ "loss": 0.2854,
1880
+ "step": 312
1881
+ },
1882
+ {
1883
+ "epoch": 2.37,
1884
+ "learning_rate": 2.263343656813107e-06,
1885
+ "loss": 0.2872,
1886
+ "step": 313
1887
+ },
1888
+ {
1889
+ "epoch": 2.37,
1890
+ "learning_rate": 2.211227212378877e-06,
1891
+ "loss": 0.2925,
1892
+ "step": 314
1893
+ },
1894
+ {
1895
+ "epoch": 2.38,
1896
+ "learning_rate": 2.1596431202184707e-06,
1897
+ "loss": 0.2951,
1898
+ "step": 315
1899
+ },
1900
+ {
1901
+ "epoch": 2.39,
1902
+ "learning_rate": 2.1085949060360654e-06,
1903
+ "loss": 0.2824,
1904
+ "step": 316
1905
+ },
1906
+ {
1907
+ "epoch": 2.4,
1908
+ "learning_rate": 2.0580860589092897e-06,
1909
+ "loss": 0.3029,
1910
+ "step": 317
1911
+ },
1912
+ {
1913
+ "epoch": 2.4,
1914
+ "learning_rate": 2.008120031050753e-06,
1915
+ "loss": 0.3057,
1916
+ "step": 318
1917
+ },
1918
+ {
1919
+ "epoch": 2.41,
1920
+ "learning_rate": 1.9587002375720864e-06,
1921
+ "loss": 0.2956,
1922
+ "step": 319
1923
+ },
1924
+ {
1925
+ "epoch": 2.42,
1926
+ "learning_rate": 1.9098300562505266e-06,
1927
+ "loss": 0.2965,
1928
+ "step": 320
1929
+ },
1930
+ {
1931
+ "epoch": 2.43,
1932
+ "learning_rate": 1.861512827298051e-06,
1933
+ "loss": 0.2938,
1934
+ "step": 321
1935
+ },
1936
+ {
1937
+ "epoch": 2.43,
1938
+ "learning_rate": 1.8137518531330768e-06,
1939
+ "loss": 0.2835,
1940
+ "step": 322
1941
+ },
1942
+ {
1943
+ "epoch": 2.44,
1944
+ "learning_rate": 1.7665503981547428e-06,
1945
+ "loss": 0.2936,
1946
+ "step": 323
1947
+ },
1948
+ {
1949
+ "epoch": 2.45,
1950
+ "learning_rate": 1.7199116885197996e-06,
1951
+ "loss": 0.289,
1952
+ "step": 324
1953
+ },
1954
+ {
1955
+ "epoch": 2.46,
1956
+ "learning_rate": 1.6738389119220966e-06,
1957
+ "loss": 0.2895,
1958
+ "step": 325
1959
+ },
1960
+ {
1961
+ "epoch": 2.46,
1962
+ "learning_rate": 1.6283352173747148e-06,
1963
+ "loss": 0.279,
1964
+ "step": 326
1965
+ },
1966
+ {
1967
+ "epoch": 2.47,
1968
+ "learning_rate": 1.5834037149947291e-06,
1969
+ "loss": 0.271,
1970
+ "step": 327
1971
+ },
1972
+ {
1973
+ "epoch": 2.48,
1974
+ "learning_rate": 1.5390474757906449e-06,
1975
+ "loss": 0.2819,
1976
+ "step": 328
1977
+ },
1978
+ {
1979
+ "epoch": 2.49,
1980
+ "learning_rate": 1.4952695314524912e-06,
1981
+ "loss": 0.2952,
1982
+ "step": 329
1983
+ },
1984
+ {
1985
+ "epoch": 2.49,
1986
+ "learning_rate": 1.4520728741446087e-06,
1987
+ "loss": 0.2705,
1988
+ "step": 330
1989
+ },
1990
+ {
1991
+ "epoch": 2.5,
1992
+ "learning_rate": 1.409460456301147e-06,
1993
+ "loss": 0.2822,
1994
+ "step": 331
1995
+ },
1996
+ {
1997
+ "epoch": 2.51,
1998
+ "learning_rate": 1.367435190424261e-06,
1999
+ "loss": 0.2668,
2000
+ "step": 332
2001
+ },
2002
+ {
2003
+ "epoch": 2.52,
2004
+ "learning_rate": 1.3259999488850473e-06,
2005
+ "loss": 0.2896,
2006
+ "step": 333
2007
+ },
2008
+ {
2009
+ "epoch": 2.52,
2010
+ "learning_rate": 1.2851575637272262e-06,
2011
+ "loss": 0.2762,
2012
+ "step": 334
2013
+ },
2014
+ {
2015
+ "epoch": 2.53,
2016
+ "learning_rate": 1.2449108264735721e-06,
2017
+ "loss": 0.2644,
2018
+ "step": 335
2019
+ },
2020
+ {
2021
+ "epoch": 2.54,
2022
+ "learning_rate": 1.2052624879351105e-06,
2023
+ "loss": 0.2785,
2024
+ "step": 336
2025
+ },
2026
+ {
2027
+ "epoch": 2.55,
2028
+ "learning_rate": 1.1662152580231145e-06,
2029
+ "loss": 0.2992,
2030
+ "step": 337
2031
+ },
2032
+ {
2033
+ "epoch": 2.55,
2034
+ "learning_rate": 1.127771805563882e-06,
2035
+ "loss": 0.2974,
2036
+ "step": 338
2037
+ },
2038
+ {
2039
+ "epoch": 2.56,
2040
+ "learning_rate": 1.0899347581163222e-06,
2041
+ "loss": 0.2813,
2042
+ "step": 339
2043
+ },
2044
+ {
2045
+ "epoch": 2.57,
2046
+ "learning_rate": 1.0527067017923654e-06,
2047
+ "loss": 0.2929,
2048
+ "step": 340
2049
+ },
2050
+ {
2051
+ "epoch": 2.58,
2052
+ "learning_rate": 1.0160901810802114e-06,
2053
+ "loss": 0.27,
2054
+ "step": 341
2055
+ },
2056
+ {
2057
+ "epoch": 2.58,
2058
+ "learning_rate": 9.800876986704111e-07,
2059
+ "loss": 0.2945,
2060
+ "step": 342
2061
+ },
2062
+ {
2063
+ "epoch": 2.59,
2064
+ "learning_rate": 9.447017152848126e-07,
2065
+ "loss": 0.2761,
2066
+ "step": 343
2067
+ },
2068
+ {
2069
+ "epoch": 2.6,
2070
+ "learning_rate": 9.09934649508375e-07,
2071
+ "loss": 0.2895,
2072
+ "step": 344
2073
+ },
2074
+ {
2075
+ "epoch": 2.61,
2076
+ "learning_rate": 8.757888776238621e-07,
2077
+ "loss": 0.2921,
2078
+ "step": 345
2079
+ },
2080
+ {
2081
+ "epoch": 2.62,
2082
+ "learning_rate": 8.42266733449425e-07,
2083
+ "loss": 0.2847,
2084
+ "step": 346
2085
+ },
2086
+ {
2087
+ "epoch": 2.62,
2088
+ "learning_rate": 8.093705081790892e-07,
2089
+ "loss": 0.2773,
2090
+ "step": 347
2091
+ },
2092
+ {
2093
+ "epoch": 2.63,
2094
+ "learning_rate": 7.771024502261526e-07,
2095
+ "loss": 0.2861,
2096
+ "step": 348
2097
+ },
2098
+ {
2099
+ "epoch": 2.64,
2100
+ "learning_rate": 7.454647650695157e-07,
2101
+ "loss": 0.2662,
2102
+ "step": 349
2103
+ },
2104
+ {
2105
+ "epoch": 2.65,
2106
+ "learning_rate": 7.144596151029304e-07,
2107
+ "loss": 0.285,
2108
+ "step": 350
2109
+ },
2110
+ {
2111
+ "epoch": 2.65,
2112
+ "learning_rate": 6.840891194872112e-07,
2113
+ "loss": 0.2989,
2114
+ "step": 351
2115
+ },
2116
+ {
2117
+ "epoch": 2.66,
2118
+ "learning_rate": 6.543553540053926e-07,
2119
+ "loss": 0.2809,
2120
+ "step": 352
2121
+ },
2122
+ {
2123
+ "epoch": 2.67,
2124
+ "learning_rate": 6.252603509208466e-07,
2125
+ "loss": 0.2923,
2126
+ "step": 353
2127
+ },
2128
+ {
2129
+ "epoch": 2.68,
2130
+ "learning_rate": 5.968060988383884e-07,
2131
+ "loss": 0.2738,
2132
+ "step": 354
2133
+ },
2134
+ {
2135
+ "epoch": 2.68,
2136
+ "learning_rate": 5.689945425683474e-07,
2137
+ "loss": 0.274,
2138
+ "step": 355
2139
+ },
2140
+ {
2141
+ "epoch": 2.69,
2142
+ "learning_rate": 5.418275829936537e-07,
2143
+ "loss": 0.2899,
2144
+ "step": 356
2145
+ },
2146
+ {
2147
+ "epoch": 2.7,
2148
+ "learning_rate": 5.15307076939906e-07,
2149
+ "loss": 0.2874,
2150
+ "step": 357
2151
+ },
2152
+ {
2153
+ "epoch": 2.71,
2154
+ "learning_rate": 4.894348370484648e-07,
2155
+ "loss": 0.2747,
2156
+ "step": 358
2157
+ },
2158
+ {
2159
+ "epoch": 2.71,
2160
+ "learning_rate": 4.642126316525586e-07,
2161
+ "loss": 0.293,
2162
+ "step": 359
2163
+ },
2164
+ {
2165
+ "epoch": 2.72,
2166
+ "learning_rate": 4.396421846564236e-07,
2167
+ "loss": 0.2835,
2168
+ "step": 360
2169
+ },
2170
+ {
2171
+ "epoch": 2.73,
2172
+ "learning_rate": 4.1572517541747294e-07,
2173
+ "loss": 0.2714,
2174
+ "step": 361
2175
+ },
2176
+ {
2177
+ "epoch": 2.74,
2178
+ "learning_rate": 3.924632386315186e-07,
2179
+ "loss": 0.2856,
2180
+ "step": 362
2181
+ },
2182
+ {
2183
+ "epoch": 2.74,
2184
+ "learning_rate": 3.698579642210398e-07,
2185
+ "loss": 0.285,
2186
+ "step": 363
2187
+ },
2188
+ {
2189
+ "epoch": 2.75,
2190
+ "learning_rate": 3.4791089722651437e-07,
2191
+ "loss": 0.2772,
2192
+ "step": 364
2193
+ },
2194
+ {
2195
+ "epoch": 2.76,
2196
+ "learning_rate": 3.2662353770081755e-07,
2197
+ "loss": 0.2763,
2198
+ "step": 365
2199
+ },
2200
+ {
2201
+ "epoch": 2.77,
2202
+ "learning_rate": 3.059973406066963e-07,
2203
+ "loss": 0.2934,
2204
+ "step": 366
2205
+ },
2206
+ {
2207
+ "epoch": 2.77,
2208
+ "learning_rate": 2.860337157173243e-07,
2209
+ "loss": 0.3004,
2210
+ "step": 367
2211
+ },
2212
+ {
2213
+ "epoch": 2.78,
2214
+ "learning_rate": 2.667340275199426e-07,
2215
+ "loss": 0.2779,
2216
+ "step": 368
2217
+ },
2218
+ {
2219
+ "epoch": 2.79,
2220
+ "learning_rate": 2.4809959512260285e-07,
2221
+ "loss": 0.2934,
2222
+ "step": 369
2223
+ },
2224
+ {
2225
+ "epoch": 2.8,
2226
+ "learning_rate": 2.3013169216400732e-07,
2227
+ "loss": 0.2791,
2228
+ "step": 370
2229
+ },
2230
+ {
2231
+ "epoch": 2.8,
2232
+ "learning_rate": 2.1283154672645522e-07,
2233
+ "loss": 0.2807,
2234
+ "step": 371
2235
+ },
2236
+ {
2237
+ "epoch": 2.81,
2238
+ "learning_rate": 1.9620034125190645e-07,
2239
+ "loss": 0.2775,
2240
+ "step": 372
2241
+ },
2242
+ {
2243
+ "epoch": 2.82,
2244
+ "learning_rate": 1.8023921246116405e-07,
2245
+ "loss": 0.2835,
2246
+ "step": 373
2247
+ },
2248
+ {
2249
+ "epoch": 2.83,
2250
+ "learning_rate": 1.6494925127617632e-07,
2251
+ "loss": 0.2821,
2252
+ "step": 374
2253
+ },
2254
+ {
2255
+ "epoch": 2.83,
2256
+ "learning_rate": 1.5033150274548325e-07,
2257
+ "loss": 0.2876,
2258
+ "step": 375
2259
+ },
2260
+ {
2261
+ "epoch": 2.84,
2262
+ "learning_rate": 1.3638696597277678e-07,
2263
+ "loss": 0.2921,
2264
+ "step": 376
2265
+ },
2266
+ {
2267
+ "epoch": 2.85,
2268
+ "learning_rate": 1.231165940486234e-07,
2269
+ "loss": 0.283,
2270
+ "step": 377
2271
+ },
2272
+ {
2273
+ "epoch": 2.86,
2274
+ "learning_rate": 1.1052129398531508e-07,
2275
+ "loss": 0.2879,
2276
+ "step": 378
2277
+ },
2278
+ {
2279
+ "epoch": 2.86,
2280
+ "learning_rate": 9.86019266548821e-08,
2281
+ "loss": 0.2763,
2282
+ "step": 379
2283
+ },
2284
+ {
2285
+ "epoch": 2.87,
2286
+ "learning_rate": 8.735930673024806e-08,
2287
+ "loss": 0.2684,
2288
+ "step": 380
2289
+ },
2290
+ {
2291
+ "epoch": 2.88,
2292
+ "learning_rate": 7.679420262954984e-08,
2293
+ "loss": 0.2827,
2294
+ "step": 381
2295
+ },
2296
+ {
2297
+ "epoch": 2.89,
2298
+ "learning_rate": 6.690733646361858e-08,
2299
+ "loss": 0.2894,
2300
+ "step": 382
2301
+ },
2302
+ {
2303
+ "epoch": 2.89,
2304
+ "learning_rate": 5.769938398662356e-08,
2305
+ "loss": 0.2813,
2306
+ "step": 383
2307
+ },
2308
+ {
2309
+ "epoch": 2.9,
2310
+ "learning_rate": 4.9170974549885844e-08,
2311
+ "loss": 0.2731,
2312
+ "step": 384
2313
+ },
2314
+ {
2315
+ "epoch": 2.91,
2316
+ "learning_rate": 4.132269105886155e-08,
2317
+ "loss": 0.286,
2318
+ "step": 385
2319
+ },
2320
+ {
2321
+ "epoch": 2.92,
2322
+ "learning_rate": 3.4155069933301535e-08,
2323
+ "loss": 0.2863,
2324
+ "step": 386
2325
+ },
2326
+ {
2327
+ "epoch": 2.92,
2328
+ "learning_rate": 2.766860107058844e-08,
2329
+ "loss": 0.2895,
2330
+ "step": 387
2331
+ },
2332
+ {
2333
+ "epoch": 2.93,
2334
+ "learning_rate": 2.1863727812254653e-08,
2335
+ "loss": 0.2779,
2336
+ "step": 388
2337
+ },
2338
+ {
2339
+ "epoch": 2.94,
2340
+ "learning_rate": 1.674084691367428e-08,
2341
+ "loss": 0.2736,
2342
+ "step": 389
2343
+ },
2344
+ {
2345
+ "epoch": 2.95,
2346
+ "learning_rate": 1.230030851695263e-08,
2347
+ "loss": 0.2897,
2348
+ "step": 390
2349
+ },
2350
+ {
2351
+ "epoch": 2.95,
2352
+ "learning_rate": 8.542416126989805e-09,
2353
+ "loss": 0.291,
2354
+ "step": 391
2355
+ },
2356
+ {
2357
+ "epoch": 2.96,
2358
+ "learning_rate": 5.467426590739511e-09,
2359
+ "loss": 0.2785,
2360
+ "step": 392
2361
+ },
2362
+ {
2363
+ "epoch": 2.97,
2364
+ "learning_rate": 3.0755500796531e-09,
2365
+ "loss": 0.2855,
2366
+ "step": 393
2367
+ },
2368
+ {
2369
+ "epoch": 2.98,
2370
+ "learning_rate": 1.3669500753099586e-09,
2371
+ "loss": 0.2707,
2372
+ "step": 394
2373
+ },
2374
+ {
2375
+ "epoch": 2.98,
2376
+ "learning_rate": 3.4174335825420955e-10,
2377
+ "loss": 0.2936,
2378
+ "step": 395
2379
+ },
2380
+ {
2381
+ "epoch": 2.99,
2382
+ "learning_rate": 0.0,
2383
+ "loss": 0.2833,
2384
+ "step": 396
2385
+ },
2386
+ {
2387
+ "epoch": 2.99,
2388
+ "step": 396,
2389
+ "total_flos": 4.052532987894432e+18,
2390
+ "train_loss": 0.15283302684323957,
2391
+ "train_runtime": 60602.7591,
2392
+ "train_samples_per_second": 1.678,
2393
+ "train_steps_per_second": 0.007
2394
+ }
2395
+ ],
2396
+ "max_steps": 396,
2397
+ "num_train_epochs": 3,
2398
+ "total_flos": 4.052532987894432e+18,
2399
+ "trial_name": null,
2400
+ "trial_params": null
2401
+ }