Training in progress, epoch 38, checkpoint
Browse files- checkpoint-12977/README.md +9 -0
- checkpoint-12977/adapter_config.json +25 -0
- checkpoint-12977/adapter_model.safetensors +3 -0
- checkpoint-12977/added_tokens.json +5 -0
- checkpoint-12977/merges.txt +0 -0
- checkpoint-12977/optimizer.pt +3 -0
- checkpoint-12977/rng_state_0.pth +3 -0
- checkpoint-12977/rng_state_1.pth +3 -0
- checkpoint-12977/rng_state_2.pth +3 -0
- checkpoint-12977/rng_state_3.pth +3 -0
- checkpoint-12977/scheduler.pt +3 -0
- checkpoint-12977/special_tokens_map.json +14 -0
- checkpoint-12977/tokenizer.json +0 -0
- checkpoint-12977/tokenizer_config.json +43 -0
- checkpoint-12977/trainer_state.json +1544 -0
- checkpoint-12977/training_args.bin +3 -0
- checkpoint-12977/vocab.json +0 -0
checkpoint-12977/README.md
ADDED
|
@@ -0,0 +1,9 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
---
|
| 2 |
+
library_name: peft
|
| 3 |
+
---
|
| 4 |
+
## Training procedure
|
| 5 |
+
|
| 6 |
+
### Framework versions
|
| 7 |
+
|
| 8 |
+
|
| 9 |
+
- PEFT 0.5.0
|
checkpoint-12977/adapter_config.json
ADDED
|
@@ -0,0 +1,25 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
{
|
| 2 |
+
"auto_mapping": null,
|
| 3 |
+
"base_model_name_or_path": "Qwen/Qwen1.5-4B",
|
| 4 |
+
"bias": "none",
|
| 5 |
+
"fan_in_fan_out": false,
|
| 6 |
+
"inference_mode": true,
|
| 7 |
+
"init_lora_weights": true,
|
| 8 |
+
"layers_pattern": null,
|
| 9 |
+
"layers_to_transform": null,
|
| 10 |
+
"lora_alpha": 16,
|
| 11 |
+
"lora_dropout": 0.05,
|
| 12 |
+
"modules_to_save": null,
|
| 13 |
+
"peft_type": "LORA",
|
| 14 |
+
"r": 16,
|
| 15 |
+
"revision": null,
|
| 16 |
+
"target_modules": [
|
| 17 |
+
"q_proj",
|
| 18 |
+
"k_proj",
|
| 19 |
+
"v_proj",
|
| 20 |
+
"down_proj",
|
| 21 |
+
"gate_proj",
|
| 22 |
+
"up_proj"
|
| 23 |
+
],
|
| 24 |
+
"task_type": "CAUSAL_LM"
|
| 25 |
+
}
|
checkpoint-12977/adapter_model.safetensors
ADDED
|
@@ -0,0 +1,3 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
version https://git-lfs.github.com/spec/v1
|
| 2 |
+
oid sha256:686da3c6b7d5c6df2f6d905818ab74e8bcf4ad2bf41e258962b1a4cae5329a0f
|
| 3 |
+
size 112130416
|
checkpoint-12977/added_tokens.json
ADDED
|
@@ -0,0 +1,5 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
{
|
| 2 |
+
"<|endoftext|>": 151643,
|
| 3 |
+
"<|im_end|>": 151645,
|
| 4 |
+
"<|im_start|>": 151644
|
| 5 |
+
}
|
checkpoint-12977/merges.txt
ADDED
|
The diff for this file is too large to render.
See raw diff
|
|
|
checkpoint-12977/optimizer.pt
ADDED
|
@@ -0,0 +1,3 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
version https://git-lfs.github.com/spec/v1
|
| 2 |
+
oid sha256:055bdb14e758889e4413fc025ecaefab5f65c088d2efa1e6ae4ad23d61252d24
|
| 3 |
+
size 224537202
|
checkpoint-12977/rng_state_0.pth
ADDED
|
@@ -0,0 +1,3 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
version https://git-lfs.github.com/spec/v1
|
| 2 |
+
oid sha256:2f35becf0c2f7c219c3880066de6ab824b512f882a694ab6fa6798b61005ba92
|
| 3 |
+
size 15024
|
checkpoint-12977/rng_state_1.pth
ADDED
|
@@ -0,0 +1,3 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
version https://git-lfs.github.com/spec/v1
|
| 2 |
+
oid sha256:1442272e545b17f4e58565300ed5ece8d98364dfb3183e20b414b5eef5e7b8b3
|
| 3 |
+
size 15024
|
checkpoint-12977/rng_state_2.pth
ADDED
|
@@ -0,0 +1,3 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
version https://git-lfs.github.com/spec/v1
|
| 2 |
+
oid sha256:a36039734f19029b83be46fdd2a4e228271f5fe8bb6ca195f23098bdc0ae8691
|
| 3 |
+
size 15024
|
checkpoint-12977/rng_state_3.pth
ADDED
|
@@ -0,0 +1,3 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
version https://git-lfs.github.com/spec/v1
|
| 2 |
+
oid sha256:1a5529fdc7b5e0a9e5a1a10de2a6d6513beebac56ef8762dfcfc8c7f56bc718e
|
| 3 |
+
size 15024
|
checkpoint-12977/scheduler.pt
ADDED
|
@@ -0,0 +1,3 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
version https://git-lfs.github.com/spec/v1
|
| 2 |
+
oid sha256:30314ffc7b7b736d8849ba2b38116d7649e938828c80198fe5b3251d85af92ac
|
| 3 |
+
size 1064
|
checkpoint-12977/special_tokens_map.json
ADDED
|
@@ -0,0 +1,14 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
{
|
| 2 |
+
"additional_special_tokens": [
|
| 3 |
+
"<|im_start|>",
|
| 4 |
+
"<|im_end|>"
|
| 5 |
+
],
|
| 6 |
+
"eos_token": {
|
| 7 |
+
"content": "<|endoftext|>",
|
| 8 |
+
"lstrip": false,
|
| 9 |
+
"normalized": false,
|
| 10 |
+
"rstrip": false,
|
| 11 |
+
"single_word": false
|
| 12 |
+
},
|
| 13 |
+
"pad_token": "<|endoftext|>"
|
| 14 |
+
}
|
checkpoint-12977/tokenizer.json
ADDED
|
The diff for this file is too large to render.
See raw diff
|
|
|
checkpoint-12977/tokenizer_config.json
ADDED
|
@@ -0,0 +1,43 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
{
|
| 2 |
+
"add_prefix_space": false,
|
| 3 |
+
"added_tokens_decoder": {
|
| 4 |
+
"151643": {
|
| 5 |
+
"content": "<|endoftext|>",
|
| 6 |
+
"lstrip": false,
|
| 7 |
+
"normalized": false,
|
| 8 |
+
"rstrip": false,
|
| 9 |
+
"single_word": false,
|
| 10 |
+
"special": true
|
| 11 |
+
},
|
| 12 |
+
"151644": {
|
| 13 |
+
"content": "<|im_start|>",
|
| 14 |
+
"lstrip": false,
|
| 15 |
+
"normalized": false,
|
| 16 |
+
"rstrip": false,
|
| 17 |
+
"single_word": false,
|
| 18 |
+
"special": true
|
| 19 |
+
},
|
| 20 |
+
"151645": {
|
| 21 |
+
"content": "<|im_end|>",
|
| 22 |
+
"lstrip": false,
|
| 23 |
+
"normalized": false,
|
| 24 |
+
"rstrip": false,
|
| 25 |
+
"single_word": false,
|
| 26 |
+
"special": true
|
| 27 |
+
}
|
| 28 |
+
},
|
| 29 |
+
"additional_special_tokens": [
|
| 30 |
+
"<|im_start|>",
|
| 31 |
+
"<|im_end|>"
|
| 32 |
+
],
|
| 33 |
+
"bos_token": null,
|
| 34 |
+
"chat_template": "{% for message in messages %}{% if loop.first and messages[0]['role'] != 'system' %}{{ '<|im_start|>system\nYou are a helpful assistant<|im_end|>\n' }}{% endif %}{{'<|im_start|>' + message['role'] + '\n' + message['content'] + '<|im_end|>' + '\n'}}{% endfor %}{% if add_generation_prompt %}{{ '<|im_start|>assistant\n' }}{% endif %}",
|
| 35 |
+
"clean_up_tokenization_spaces": false,
|
| 36 |
+
"eos_token": "<|endoftext|>",
|
| 37 |
+
"errors": "replace",
|
| 38 |
+
"model_max_length": 32768,
|
| 39 |
+
"pad_token": "<|endoftext|>",
|
| 40 |
+
"split_special_tokens": false,
|
| 41 |
+
"tokenizer_class": "Qwen2Tokenizer",
|
| 42 |
+
"unk_token": null
|
| 43 |
+
}
|
checkpoint-12977/trainer_state.json
ADDED
|
@@ -0,0 +1,1544 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
{
|
| 2 |
+
"best_metric": null,
|
| 3 |
+
"best_model_checkpoint": null,
|
| 4 |
+
"epoch": 38.0,
|
| 5 |
+
"eval_steps": 500,
|
| 6 |
+
"global_step": 12977,
|
| 7 |
+
"is_hyper_param_search": false,
|
| 8 |
+
"is_local_process_zero": true,
|
| 9 |
+
"is_world_process_zero": true,
|
| 10 |
+
"log_history": [
|
| 11 |
+
{
|
| 12 |
+
"epoch": 0.29282576866764276,
|
| 13 |
+
"grad_norm": 0.2411503791809082,
|
| 14 |
+
"learning_rate": 0.0001,
|
| 15 |
+
"loss": 2.0009,
|
| 16 |
+
"step": 100
|
| 17 |
+
},
|
| 18 |
+
{
|
| 19 |
+
"epoch": 0.5856515373352855,
|
| 20 |
+
"grad_norm": 0.22286373376846313,
|
| 21 |
+
"learning_rate": 0.0001,
|
| 22 |
+
"loss": 1.9538,
|
| 23 |
+
"step": 200
|
| 24 |
+
},
|
| 25 |
+
{
|
| 26 |
+
"epoch": 0.8784773060029283,
|
| 27 |
+
"grad_norm": 0.24079065024852753,
|
| 28 |
+
"learning_rate": 0.0001,
|
| 29 |
+
"loss": 1.9571,
|
| 30 |
+
"step": 300
|
| 31 |
+
},
|
| 32 |
+
{
|
| 33 |
+
"epoch": 0.9985358711566618,
|
| 34 |
+
"eval_accuracy": 0.45379487179487177,
|
| 35 |
+
"eval_loss": 3.9512319564819336,
|
| 36 |
+
"eval_runtime": 5.0647,
|
| 37 |
+
"eval_samples_per_second": 98.722,
|
| 38 |
+
"eval_steps_per_second": 12.439,
|
| 39 |
+
"step": 341
|
| 40 |
+
},
|
| 41 |
+
{
|
| 42 |
+
"epoch": 0.9985358711566618,
|
| 43 |
+
"eval_bleu": 0.28767194245703276,
|
| 44 |
+
"eval_exact_match": 0.03,
|
| 45 |
+
"eval_prefix_exact_match": 0.042,
|
| 46 |
+
"step": 341
|
| 47 |
+
},
|
| 48 |
+
{
|
| 49 |
+
"epoch": 1.171303074670571,
|
| 50 |
+
"grad_norm": 0.33546799421310425,
|
| 51 |
+
"learning_rate": 0.0001,
|
| 52 |
+
"loss": 1.9255,
|
| 53 |
+
"step": 400
|
| 54 |
+
},
|
| 55 |
+
{
|
| 56 |
+
"epoch": 1.4641288433382138,
|
| 57 |
+
"grad_norm": 0.3515666723251343,
|
| 58 |
+
"learning_rate": 0.0001,
|
| 59 |
+
"loss": 1.8684,
|
| 60 |
+
"step": 500
|
| 61 |
+
},
|
| 62 |
+
{
|
| 63 |
+
"epoch": 1.7569546120058566,
|
| 64 |
+
"grad_norm": 0.35733896493911743,
|
| 65 |
+
"learning_rate": 0.0001,
|
| 66 |
+
"loss": 1.8819,
|
| 67 |
+
"step": 600
|
| 68 |
+
},
|
| 69 |
+
{
|
| 70 |
+
"epoch": 2.0,
|
| 71 |
+
"eval_accuracy": 0.4482564102564103,
|
| 72 |
+
"eval_loss": 4.112804412841797,
|
| 73 |
+
"eval_runtime": 5.0848,
|
| 74 |
+
"eval_samples_per_second": 98.333,
|
| 75 |
+
"eval_steps_per_second": 12.39,
|
| 76 |
+
"step": 683
|
| 77 |
+
},
|
| 78 |
+
{
|
| 79 |
+
"epoch": 2.0,
|
| 80 |
+
"eval_bleu": 0.24521038067633488,
|
| 81 |
+
"eval_exact_match": 0.02,
|
| 82 |
+
"eval_prefix_exact_match": 0.044,
|
| 83 |
+
"step": 683
|
| 84 |
+
},
|
| 85 |
+
{
|
| 86 |
+
"epoch": 2.049780380673499,
|
| 87 |
+
"grad_norm": 0.3784458339214325,
|
| 88 |
+
"learning_rate": 0.0001,
|
| 89 |
+
"loss": 1.8613,
|
| 90 |
+
"step": 700
|
| 91 |
+
},
|
| 92 |
+
{
|
| 93 |
+
"epoch": 2.342606149341142,
|
| 94 |
+
"grad_norm": 0.46702849864959717,
|
| 95 |
+
"learning_rate": 0.0001,
|
| 96 |
+
"loss": 1.743,
|
| 97 |
+
"step": 800
|
| 98 |
+
},
|
| 99 |
+
{
|
| 100 |
+
"epoch": 2.6354319180087846,
|
| 101 |
+
"grad_norm": 0.5156407356262207,
|
| 102 |
+
"learning_rate": 0.0001,
|
| 103 |
+
"loss": 1.7653,
|
| 104 |
+
"step": 900
|
| 105 |
+
},
|
| 106 |
+
{
|
| 107 |
+
"epoch": 2.9282576866764276,
|
| 108 |
+
"grad_norm": 0.4755038619041443,
|
| 109 |
+
"learning_rate": 0.0001,
|
| 110 |
+
"loss": 1.7702,
|
| 111 |
+
"step": 1000
|
| 112 |
+
},
|
| 113 |
+
{
|
| 114 |
+
"epoch": 2.998535871156662,
|
| 115 |
+
"eval_accuracy": 0.446051282051282,
|
| 116 |
+
"eval_loss": 4.327652454376221,
|
| 117 |
+
"eval_runtime": 5.1641,
|
| 118 |
+
"eval_samples_per_second": 96.822,
|
| 119 |
+
"eval_steps_per_second": 12.2,
|
| 120 |
+
"step": 1024
|
| 121 |
+
},
|
| 122 |
+
{
|
| 123 |
+
"epoch": 2.998535871156662,
|
| 124 |
+
"eval_bleu": 0.2598440457703149,
|
| 125 |
+
"eval_exact_match": 0.018,
|
| 126 |
+
"eval_prefix_exact_match": 0.034,
|
| 127 |
+
"step": 1024
|
| 128 |
+
},
|
| 129 |
+
{
|
| 130 |
+
"epoch": 3.22108345534407,
|
| 131 |
+
"grad_norm": 0.5288150906562805,
|
| 132 |
+
"learning_rate": 0.0001,
|
| 133 |
+
"loss": 1.6325,
|
| 134 |
+
"step": 1100
|
| 135 |
+
},
|
| 136 |
+
{
|
| 137 |
+
"epoch": 3.513909224011713,
|
| 138 |
+
"grad_norm": 0.5784010887145996,
|
| 139 |
+
"learning_rate": 0.0001,
|
| 140 |
+
"loss": 1.5896,
|
| 141 |
+
"step": 1200
|
| 142 |
+
},
|
| 143 |
+
{
|
| 144 |
+
"epoch": 3.8067349926793557,
|
| 145 |
+
"grad_norm": 0.6458429098129272,
|
| 146 |
+
"learning_rate": 0.0001,
|
| 147 |
+
"loss": 1.6163,
|
| 148 |
+
"step": 1300
|
| 149 |
+
},
|
| 150 |
+
{
|
| 151 |
+
"epoch": 4.0,
|
| 152 |
+
"eval_accuracy": 0.44235897435897437,
|
| 153 |
+
"eval_loss": 4.584905624389648,
|
| 154 |
+
"eval_runtime": 4.9243,
|
| 155 |
+
"eval_samples_per_second": 101.538,
|
| 156 |
+
"eval_steps_per_second": 12.794,
|
| 157 |
+
"step": 1366
|
| 158 |
+
},
|
| 159 |
+
{
|
| 160 |
+
"epoch": 4.0,
|
| 161 |
+
"eval_bleu": 0.27967222841565076,
|
| 162 |
+
"eval_exact_match": 0.01,
|
| 163 |
+
"eval_prefix_exact_match": 0.038,
|
| 164 |
+
"step": 1366
|
| 165 |
+
},
|
| 166 |
+
{
|
| 167 |
+
"epoch": 4.099560761346998,
|
| 168 |
+
"grad_norm": 0.732841432094574,
|
| 169 |
+
"learning_rate": 0.0001,
|
| 170 |
+
"loss": 1.5441,
|
| 171 |
+
"step": 1400
|
| 172 |
+
},
|
| 173 |
+
{
|
| 174 |
+
"epoch": 4.392386530014641,
|
| 175 |
+
"grad_norm": 0.7372816205024719,
|
| 176 |
+
"learning_rate": 0.0001,
|
| 177 |
+
"loss": 1.4142,
|
| 178 |
+
"step": 1500
|
| 179 |
+
},
|
| 180 |
+
{
|
| 181 |
+
"epoch": 4.685212298682284,
|
| 182 |
+
"grad_norm": 0.8235049247741699,
|
| 183 |
+
"learning_rate": 0.0001,
|
| 184 |
+
"loss": 1.4198,
|
| 185 |
+
"step": 1600
|
| 186 |
+
},
|
| 187 |
+
{
|
| 188 |
+
"epoch": 4.978038067349927,
|
| 189 |
+
"grad_norm": 0.7765743732452393,
|
| 190 |
+
"learning_rate": 0.0001,
|
| 191 |
+
"loss": 1.4427,
|
| 192 |
+
"step": 1700
|
| 193 |
+
},
|
| 194 |
+
{
|
| 195 |
+
"epoch": 4.998535871156662,
|
| 196 |
+
"eval_accuracy": 0.43861538461538463,
|
| 197 |
+
"eval_loss": 4.850330829620361,
|
| 198 |
+
"eval_runtime": 4.9395,
|
| 199 |
+
"eval_samples_per_second": 101.224,
|
| 200 |
+
"eval_steps_per_second": 12.754,
|
| 201 |
+
"step": 1707
|
| 202 |
+
},
|
| 203 |
+
{
|
| 204 |
+
"epoch": 4.998535871156662,
|
| 205 |
+
"eval_bleu": 0.2170008640509292,
|
| 206 |
+
"eval_exact_match": 0.006,
|
| 207 |
+
"eval_prefix_exact_match": 0.032,
|
| 208 |
+
"step": 1707
|
| 209 |
+
},
|
| 210 |
+
{
|
| 211 |
+
"epoch": 5.270863836017569,
|
| 212 |
+
"grad_norm": 0.9119400978088379,
|
| 213 |
+
"learning_rate": 0.0001,
|
| 214 |
+
"loss": 1.2469,
|
| 215 |
+
"step": 1800
|
| 216 |
+
},
|
| 217 |
+
{
|
| 218 |
+
"epoch": 5.563689604685212,
|
| 219 |
+
"grad_norm": 0.8966185450553894,
|
| 220 |
+
"learning_rate": 0.0001,
|
| 221 |
+
"loss": 1.2294,
|
| 222 |
+
"step": 1900
|
| 223 |
+
},
|
| 224 |
+
{
|
| 225 |
+
"epoch": 5.856515373352855,
|
| 226 |
+
"grad_norm": 0.9629669189453125,
|
| 227 |
+
"learning_rate": 0.0001,
|
| 228 |
+
"loss": 1.2498,
|
| 229 |
+
"step": 2000
|
| 230 |
+
},
|
| 231 |
+
{
|
| 232 |
+
"epoch": 6.0,
|
| 233 |
+
"eval_accuracy": 0.4348717948717949,
|
| 234 |
+
"eval_loss": 5.09260892868042,
|
| 235 |
+
"eval_runtime": 4.9572,
|
| 236 |
+
"eval_samples_per_second": 100.863,
|
| 237 |
+
"eval_steps_per_second": 12.709,
|
| 238 |
+
"step": 2049
|
| 239 |
+
},
|
| 240 |
+
{
|
| 241 |
+
"epoch": 6.0,
|
| 242 |
+
"eval_bleu": 0.20196479624851718,
|
| 243 |
+
"eval_exact_match": 0.002,
|
| 244 |
+
"eval_prefix_exact_match": 0.03,
|
| 245 |
+
"step": 2049
|
| 246 |
+
},
|
| 247 |
+
{
|
| 248 |
+
"epoch": 6.149341142020498,
|
| 249 |
+
"grad_norm": 1.0808535814285278,
|
| 250 |
+
"learning_rate": 0.0001,
|
| 251 |
+
"loss": 1.1238,
|
| 252 |
+
"step": 2100
|
| 253 |
+
},
|
| 254 |
+
{
|
| 255 |
+
"epoch": 6.44216691068814,
|
| 256 |
+
"grad_norm": 1.1639196872711182,
|
| 257 |
+
"learning_rate": 0.0001,
|
| 258 |
+
"loss": 1.0301,
|
| 259 |
+
"step": 2200
|
| 260 |
+
},
|
| 261 |
+
{
|
| 262 |
+
"epoch": 6.734992679355783,
|
| 263 |
+
"grad_norm": 1.0662291049957275,
|
| 264 |
+
"learning_rate": 0.0001,
|
| 265 |
+
"loss": 1.0655,
|
| 266 |
+
"step": 2300
|
| 267 |
+
},
|
| 268 |
+
{
|
| 269 |
+
"epoch": 6.998535871156662,
|
| 270 |
+
"eval_accuracy": 0.4326153846153846,
|
| 271 |
+
"eval_loss": 5.270845890045166,
|
| 272 |
+
"eval_runtime": 5.089,
|
| 273 |
+
"eval_samples_per_second": 98.251,
|
| 274 |
+
"eval_steps_per_second": 12.38,
|
| 275 |
+
"step": 2390
|
| 276 |
+
},
|
| 277 |
+
{
|
| 278 |
+
"epoch": 6.998535871156662,
|
| 279 |
+
"eval_bleu": 0.17780932376070374,
|
| 280 |
+
"eval_exact_match": 0.004,
|
| 281 |
+
"eval_prefix_exact_match": 0.024,
|
| 282 |
+
"step": 2390
|
| 283 |
+
},
|
| 284 |
+
{
|
| 285 |
+
"epoch": 7.027818448023426,
|
| 286 |
+
"grad_norm": 1.6462695598602295,
|
| 287 |
+
"learning_rate": 0.0001,
|
| 288 |
+
"loss": 1.0358,
|
| 289 |
+
"step": 2400
|
| 290 |
+
},
|
| 291 |
+
{
|
| 292 |
+
"epoch": 7.320644216691068,
|
| 293 |
+
"grad_norm": 1.2542779445648193,
|
| 294 |
+
"learning_rate": 0.0001,
|
| 295 |
+
"loss": 0.8413,
|
| 296 |
+
"step": 2500
|
| 297 |
+
},
|
| 298 |
+
{
|
| 299 |
+
"epoch": 7.613469985358711,
|
| 300 |
+
"grad_norm": 1.334762454032898,
|
| 301 |
+
"learning_rate": 0.0001,
|
| 302 |
+
"loss": 0.8654,
|
| 303 |
+
"step": 2600
|
| 304 |
+
},
|
| 305 |
+
{
|
| 306 |
+
"epoch": 7.906295754026354,
|
| 307 |
+
"grad_norm": 1.2401787042617798,
|
| 308 |
+
"learning_rate": 0.0001,
|
| 309 |
+
"loss": 0.8733,
|
| 310 |
+
"step": 2700
|
| 311 |
+
},
|
| 312 |
+
{
|
| 313 |
+
"epoch": 8.0,
|
| 314 |
+
"eval_accuracy": 0.4316923076923077,
|
| 315 |
+
"eval_loss": 5.402440547943115,
|
| 316 |
+
"eval_runtime": 4.9383,
|
| 317 |
+
"eval_samples_per_second": 101.25,
|
| 318 |
+
"eval_steps_per_second": 12.758,
|
| 319 |
+
"step": 2732
|
| 320 |
+
},
|
| 321 |
+
{
|
| 322 |
+
"epoch": 8.0,
|
| 323 |
+
"eval_bleu": 0.21418356891762502,
|
| 324 |
+
"eval_exact_match": 0.002,
|
| 325 |
+
"eval_prefix_exact_match": 0.026,
|
| 326 |
+
"step": 2732
|
| 327 |
+
},
|
| 328 |
+
{
|
| 329 |
+
"epoch": 8.199121522693996,
|
| 330 |
+
"grad_norm": 1.255996584892273,
|
| 331 |
+
"learning_rate": 0.0001,
|
| 332 |
+
"loss": 0.7302,
|
| 333 |
+
"step": 2800
|
| 334 |
+
},
|
| 335 |
+
{
|
| 336 |
+
"epoch": 8.49194729136164,
|
| 337 |
+
"grad_norm": 1.2836525440216064,
|
| 338 |
+
"learning_rate": 0.0001,
|
| 339 |
+
"loss": 0.6875,
|
| 340 |
+
"step": 2900
|
| 341 |
+
},
|
| 342 |
+
{
|
| 343 |
+
"epoch": 8.784773060029282,
|
| 344 |
+
"grad_norm": 1.375715732574463,
|
| 345 |
+
"learning_rate": 0.0001,
|
| 346 |
+
"loss": 0.7219,
|
| 347 |
+
"step": 3000
|
| 348 |
+
},
|
| 349 |
+
{
|
| 350 |
+
"epoch": 8.998535871156662,
|
| 351 |
+
"eval_accuracy": 0.42938461538461536,
|
| 352 |
+
"eval_loss": 5.534832000732422,
|
| 353 |
+
"eval_runtime": 5.2363,
|
| 354 |
+
"eval_samples_per_second": 95.487,
|
| 355 |
+
"eval_steps_per_second": 12.031,
|
| 356 |
+
"step": 3073
|
| 357 |
+
},
|
| 358 |
+
{
|
| 359 |
+
"epoch": 8.998535871156662,
|
| 360 |
+
"eval_bleu": 0.20459866130314297,
|
| 361 |
+
"eval_exact_match": 0.0,
|
| 362 |
+
"eval_prefix_exact_match": 0.024,
|
| 363 |
+
"step": 3073
|
| 364 |
+
},
|
| 365 |
+
{
|
| 366 |
+
"epoch": 9.077598828696924,
|
| 367 |
+
"grad_norm": 1.481998324394226,
|
| 368 |
+
"learning_rate": 0.0001,
|
| 369 |
+
"loss": 0.6722,
|
| 370 |
+
"step": 3100
|
| 371 |
+
},
|
| 372 |
+
{
|
| 373 |
+
"epoch": 9.370424597364568,
|
| 374 |
+
"grad_norm": 1.4086748361587524,
|
| 375 |
+
"learning_rate": 0.0001,
|
| 376 |
+
"loss": 0.5489,
|
| 377 |
+
"step": 3200
|
| 378 |
+
},
|
| 379 |
+
{
|
| 380 |
+
"epoch": 9.66325036603221,
|
| 381 |
+
"grad_norm": 1.4381046295166016,
|
| 382 |
+
"learning_rate": 0.0001,
|
| 383 |
+
"loss": 0.565,
|
| 384 |
+
"step": 3300
|
| 385 |
+
},
|
| 386 |
+
{
|
| 387 |
+
"epoch": 9.956076134699854,
|
| 388 |
+
"grad_norm": 1.7364269495010376,
|
| 389 |
+
"learning_rate": 0.0001,
|
| 390 |
+
"loss": 0.5932,
|
| 391 |
+
"step": 3400
|
| 392 |
+
},
|
| 393 |
+
{
|
| 394 |
+
"epoch": 10.0,
|
| 395 |
+
"eval_accuracy": 0.42605128205128207,
|
| 396 |
+
"eval_loss": 5.768980503082275,
|
| 397 |
+
"eval_runtime": 5.0821,
|
| 398 |
+
"eval_samples_per_second": 98.384,
|
| 399 |
+
"eval_steps_per_second": 12.396,
|
| 400 |
+
"step": 3415
|
| 401 |
+
},
|
| 402 |
+
{
|
| 403 |
+
"epoch": 10.0,
|
| 404 |
+
"eval_bleu": 0.19515580420261575,
|
| 405 |
+
"eval_exact_match": 0.002,
|
| 406 |
+
"eval_prefix_exact_match": 0.026,
|
| 407 |
+
"step": 3415
|
| 408 |
+
},
|
| 409 |
+
{
|
| 410 |
+
"epoch": 10.248901903367496,
|
| 411 |
+
"grad_norm": 1.599753737449646,
|
| 412 |
+
"learning_rate": 0.0001,
|
| 413 |
+
"loss": 0.4581,
|
| 414 |
+
"step": 3500
|
| 415 |
+
},
|
| 416 |
+
{
|
| 417 |
+
"epoch": 10.541727672035138,
|
| 418 |
+
"grad_norm": 1.5595277547836304,
|
| 419 |
+
"learning_rate": 0.0001,
|
| 420 |
+
"loss": 0.4519,
|
| 421 |
+
"step": 3600
|
| 422 |
+
},
|
| 423 |
+
{
|
| 424 |
+
"epoch": 10.834553440702782,
|
| 425 |
+
"grad_norm": 1.6201436519622803,
|
| 426 |
+
"learning_rate": 0.0001,
|
| 427 |
+
"loss": 0.4719,
|
| 428 |
+
"step": 3700
|
| 429 |
+
},
|
| 430 |
+
{
|
| 431 |
+
"epoch": 10.998535871156662,
|
| 432 |
+
"eval_accuracy": 0.42538461538461536,
|
| 433 |
+
"eval_loss": 5.89434289932251,
|
| 434 |
+
"eval_runtime": 5.0717,
|
| 435 |
+
"eval_samples_per_second": 98.586,
|
| 436 |
+
"eval_steps_per_second": 12.422,
|
| 437 |
+
"step": 3756
|
| 438 |
+
},
|
| 439 |
+
{
|
| 440 |
+
"epoch": 10.998535871156662,
|
| 441 |
+
"eval_bleu": 0.2074466521823303,
|
| 442 |
+
"eval_exact_match": 0.0,
|
| 443 |
+
"eval_prefix_exact_match": 0.02,
|
| 444 |
+
"step": 3756
|
| 445 |
+
},
|
| 446 |
+
{
|
| 447 |
+
"epoch": 11.127379209370424,
|
| 448 |
+
"grad_norm": 1.6209241151809692,
|
| 449 |
+
"learning_rate": 0.0001,
|
| 450 |
+
"loss": 0.4263,
|
| 451 |
+
"step": 3800
|
| 452 |
+
},
|
| 453 |
+
{
|
| 454 |
+
"epoch": 11.420204978038067,
|
| 455 |
+
"grad_norm": 1.5148394107818604,
|
| 456 |
+
"learning_rate": 0.0001,
|
| 457 |
+
"loss": 0.3625,
|
| 458 |
+
"step": 3900
|
| 459 |
+
},
|
| 460 |
+
{
|
| 461 |
+
"epoch": 11.71303074670571,
|
| 462 |
+
"grad_norm": 1.6862399578094482,
|
| 463 |
+
"learning_rate": 0.0001,
|
| 464 |
+
"loss": 0.3838,
|
| 465 |
+
"step": 4000
|
| 466 |
+
},
|
| 467 |
+
{
|
| 468 |
+
"epoch": 12.0,
|
| 469 |
+
"eval_accuracy": 0.4247179487179487,
|
| 470 |
+
"eval_loss": 6.019149303436279,
|
| 471 |
+
"eval_runtime": 5.3028,
|
| 472 |
+
"eval_samples_per_second": 94.291,
|
| 473 |
+
"eval_steps_per_second": 11.881,
|
| 474 |
+
"step": 4098
|
| 475 |
+
},
|
| 476 |
+
{
|
| 477 |
+
"epoch": 12.0,
|
| 478 |
+
"eval_bleu": 0.16189430124724263,
|
| 479 |
+
"eval_exact_match": 0.002,
|
| 480 |
+
"eval_prefix_exact_match": 0.028,
|
| 481 |
+
"step": 4098
|
| 482 |
+
},
|
| 483 |
+
{
|
| 484 |
+
"epoch": 12.005856515373353,
|
| 485 |
+
"grad_norm": 1.3296407461166382,
|
| 486 |
+
"learning_rate": 0.0001,
|
| 487 |
+
"loss": 0.3952,
|
| 488 |
+
"step": 4100
|
| 489 |
+
},
|
| 490 |
+
{
|
| 491 |
+
"epoch": 12.298682284040996,
|
| 492 |
+
"grad_norm": 1.3439353704452515,
|
| 493 |
+
"learning_rate": 0.0001,
|
| 494 |
+
"loss": 0.2901,
|
| 495 |
+
"step": 4200
|
| 496 |
+
},
|
| 497 |
+
{
|
| 498 |
+
"epoch": 12.591508052708638,
|
| 499 |
+
"grad_norm": 1.5329118967056274,
|
| 500 |
+
"learning_rate": 0.0001,
|
| 501 |
+
"loss": 0.3114,
|
| 502 |
+
"step": 4300
|
| 503 |
+
},
|
| 504 |
+
{
|
| 505 |
+
"epoch": 12.88433382137628,
|
| 506 |
+
"grad_norm": 1.649775743484497,
|
| 507 |
+
"learning_rate": 0.0001,
|
| 508 |
+
"loss": 0.329,
|
| 509 |
+
"step": 4400
|
| 510 |
+
},
|
| 511 |
+
{
|
| 512 |
+
"epoch": 12.998535871156662,
|
| 513 |
+
"eval_accuracy": 0.42456410256410254,
|
| 514 |
+
"eval_loss": 6.104373455047607,
|
| 515 |
+
"eval_runtime": 5.9944,
|
| 516 |
+
"eval_samples_per_second": 83.411,
|
| 517 |
+
"eval_steps_per_second": 10.51,
|
| 518 |
+
"step": 4439
|
| 519 |
+
},
|
| 520 |
+
{
|
| 521 |
+
"epoch": 12.998535871156662,
|
| 522 |
+
"eval_bleu": 0.17170793542335236,
|
| 523 |
+
"eval_exact_match": 0.002,
|
| 524 |
+
"eval_prefix_exact_match": 0.028,
|
| 525 |
+
"step": 4439
|
| 526 |
+
},
|
| 527 |
+
{
|
| 528 |
+
"epoch": 13.177159590043924,
|
| 529 |
+
"grad_norm": 1.5262281894683838,
|
| 530 |
+
"learning_rate": 0.0001,
|
| 531 |
+
"loss": 0.2758,
|
| 532 |
+
"step": 4500
|
| 533 |
+
},
|
| 534 |
+
{
|
| 535 |
+
"epoch": 13.469985358711567,
|
| 536 |
+
"grad_norm": 1.5962131023406982,
|
| 537 |
+
"learning_rate": 0.0001,
|
| 538 |
+
"loss": 0.2594,
|
| 539 |
+
"step": 4600
|
| 540 |
+
},
|
| 541 |
+
{
|
| 542 |
+
"epoch": 13.762811127379209,
|
| 543 |
+
"grad_norm": 1.4258382320404053,
|
| 544 |
+
"learning_rate": 0.0001,
|
| 545 |
+
"loss": 0.2742,
|
| 546 |
+
"step": 4700
|
| 547 |
+
},
|
| 548 |
+
{
|
| 549 |
+
"epoch": 14.0,
|
| 550 |
+
"eval_accuracy": 0.4216410256410256,
|
| 551 |
+
"eval_loss": 6.146524906158447,
|
| 552 |
+
"eval_runtime": 4.9576,
|
| 553 |
+
"eval_samples_per_second": 100.855,
|
| 554 |
+
"eval_steps_per_second": 12.708,
|
| 555 |
+
"step": 4781
|
| 556 |
+
},
|
| 557 |
+
{
|
| 558 |
+
"epoch": 14.0,
|
| 559 |
+
"eval_bleu": 0.20004000747911055,
|
| 560 |
+
"eval_exact_match": 0.002,
|
| 561 |
+
"eval_prefix_exact_match": 0.026,
|
| 562 |
+
"step": 4781
|
| 563 |
+
},
|
| 564 |
+
{
|
| 565 |
+
"epoch": 14.055636896046853,
|
| 566 |
+
"grad_norm": 1.388717532157898,
|
| 567 |
+
"learning_rate": 0.0001,
|
| 568 |
+
"loss": 0.269,
|
| 569 |
+
"step": 4800
|
| 570 |
+
},
|
| 571 |
+
{
|
| 572 |
+
"epoch": 14.348462664714495,
|
| 573 |
+
"grad_norm": 1.2370668649673462,
|
| 574 |
+
"learning_rate": 0.0001,
|
| 575 |
+
"loss": 0.2226,
|
| 576 |
+
"step": 4900
|
| 577 |
+
},
|
| 578 |
+
{
|
| 579 |
+
"epoch": 14.641288433382137,
|
| 580 |
+
"grad_norm": 1.3778610229492188,
|
| 581 |
+
"learning_rate": 0.0001,
|
| 582 |
+
"loss": 0.2334,
|
| 583 |
+
"step": 5000
|
| 584 |
+
},
|
| 585 |
+
{
|
| 586 |
+
"epoch": 14.93411420204978,
|
| 587 |
+
"grad_norm": 1.6371277570724487,
|
| 588 |
+
"learning_rate": 0.0001,
|
| 589 |
+
"loss": 0.2432,
|
| 590 |
+
"step": 5100
|
| 591 |
+
},
|
| 592 |
+
{
|
| 593 |
+
"epoch": 14.998535871156662,
|
| 594 |
+
"eval_accuracy": 0.4226666666666667,
|
| 595 |
+
"eval_loss": 6.3254475593566895,
|
| 596 |
+
"eval_runtime": 5.2196,
|
| 597 |
+
"eval_samples_per_second": 95.794,
|
| 598 |
+
"eval_steps_per_second": 12.07,
|
| 599 |
+
"step": 5122
|
| 600 |
+
},
|
| 601 |
+
{
|
| 602 |
+
"epoch": 14.998535871156662,
|
| 603 |
+
"eval_bleu": 0.16083749853030369,
|
| 604 |
+
"eval_exact_match": 0.0,
|
| 605 |
+
"eval_prefix_exact_match": 0.024,
|
| 606 |
+
"step": 5122
|
| 607 |
+
},
|
| 608 |
+
{
|
| 609 |
+
"epoch": 15.226939970717423,
|
| 610 |
+
"grad_norm": 1.2761411666870117,
|
| 611 |
+
"learning_rate": 0.0001,
|
| 612 |
+
"loss": 0.2066,
|
| 613 |
+
"step": 5200
|
| 614 |
+
},
|
| 615 |
+
{
|
| 616 |
+
"epoch": 15.519765739385067,
|
| 617 |
+
"grad_norm": 1.6878156661987305,
|
| 618 |
+
"learning_rate": 0.0001,
|
| 619 |
+
"loss": 0.2044,
|
| 620 |
+
"step": 5300
|
| 621 |
+
},
|
| 622 |
+
{
|
| 623 |
+
"epoch": 15.812591508052709,
|
| 624 |
+
"grad_norm": 1.4964863061904907,
|
| 625 |
+
"learning_rate": 0.0001,
|
| 626 |
+
"loss": 0.2158,
|
| 627 |
+
"step": 5400
|
| 628 |
+
},
|
| 629 |
+
{
|
| 630 |
+
"epoch": 16.0,
|
| 631 |
+
"eval_accuracy": 0.4227692307692308,
|
| 632 |
+
"eval_loss": 6.441042423248291,
|
| 633 |
+
"eval_runtime": 4.9987,
|
| 634 |
+
"eval_samples_per_second": 100.025,
|
| 635 |
+
"eval_steps_per_second": 12.603,
|
| 636 |
+
"step": 5464
|
| 637 |
+
},
|
| 638 |
+
{
|
| 639 |
+
"epoch": 16.0,
|
| 640 |
+
"eval_bleu": 0.1349628182313855,
|
| 641 |
+
"eval_exact_match": 0.002,
|
| 642 |
+
"eval_prefix_exact_match": 0.026,
|
| 643 |
+
"step": 5464
|
| 644 |
+
},
|
| 645 |
+
{
|
| 646 |
+
"epoch": 16.105417276720353,
|
| 647 |
+
"grad_norm": 0.9965784549713135,
|
| 648 |
+
"learning_rate": 0.0001,
|
| 649 |
+
"loss": 0.2038,
|
| 650 |
+
"step": 5500
|
| 651 |
+
},
|
| 652 |
+
{
|
| 653 |
+
"epoch": 16.398243045387993,
|
| 654 |
+
"grad_norm": 1.2694511413574219,
|
| 655 |
+
"learning_rate": 0.0001,
|
| 656 |
+
"loss": 0.1857,
|
| 657 |
+
"step": 5600
|
| 658 |
+
},
|
| 659 |
+
{
|
| 660 |
+
"epoch": 16.691068814055637,
|
| 661 |
+
"grad_norm": 1.436286211013794,
|
| 662 |
+
"learning_rate": 0.0001,
|
| 663 |
+
"loss": 0.1944,
|
| 664 |
+
"step": 5700
|
| 665 |
+
},
|
| 666 |
+
{
|
| 667 |
+
"epoch": 16.98389458272328,
|
| 668 |
+
"grad_norm": 1.526081919670105,
|
| 669 |
+
"learning_rate": 0.0001,
|
| 670 |
+
"loss": 0.2013,
|
| 671 |
+
"step": 5800
|
| 672 |
+
},
|
| 673 |
+
{
|
| 674 |
+
"epoch": 16.998535871156662,
|
| 675 |
+
"eval_accuracy": 0.4214871794871795,
|
| 676 |
+
"eval_loss": 6.39243745803833,
|
| 677 |
+
"eval_runtime": 5.1033,
|
| 678 |
+
"eval_samples_per_second": 97.975,
|
| 679 |
+
"eval_steps_per_second": 12.345,
|
| 680 |
+
"step": 5805
|
| 681 |
+
},
|
| 682 |
+
{
|
| 683 |
+
"epoch": 16.998535871156662,
|
| 684 |
+
"eval_bleu": 0.18366001556911046,
|
| 685 |
+
"eval_exact_match": 0.0,
|
| 686 |
+
"eval_prefix_exact_match": 0.018,
|
| 687 |
+
"step": 5805
|
| 688 |
+
},
|
| 689 |
+
{
|
| 690 |
+
"epoch": 17.27672035139092,
|
| 691 |
+
"grad_norm": 1.1292760372161865,
|
| 692 |
+
"learning_rate": 0.0001,
|
| 693 |
+
"loss": 0.1694,
|
| 694 |
+
"step": 5900
|
| 695 |
+
},
|
| 696 |
+
{
|
| 697 |
+
"epoch": 17.569546120058565,
|
| 698 |
+
"grad_norm": 1.440766453742981,
|
| 699 |
+
"learning_rate": 0.0001,
|
| 700 |
+
"loss": 0.1782,
|
| 701 |
+
"step": 6000
|
| 702 |
+
},
|
| 703 |
+
{
|
| 704 |
+
"epoch": 17.86237188872621,
|
| 705 |
+
"grad_norm": 1.2640533447265625,
|
| 706 |
+
"learning_rate": 0.0001,
|
| 707 |
+
"loss": 0.1851,
|
| 708 |
+
"step": 6100
|
| 709 |
+
},
|
| 710 |
+
{
|
| 711 |
+
"epoch": 18.0,
|
| 712 |
+
"eval_accuracy": 0.42005128205128206,
|
| 713 |
+
"eval_loss": 6.5217366218566895,
|
| 714 |
+
"eval_runtime": 5.0655,
|
| 715 |
+
"eval_samples_per_second": 98.707,
|
| 716 |
+
"eval_steps_per_second": 12.437,
|
| 717 |
+
"step": 6147
|
| 718 |
+
},
|
| 719 |
+
{
|
| 720 |
+
"epoch": 18.0,
|
| 721 |
+
"eval_bleu": 0.14616910828479238,
|
| 722 |
+
"eval_exact_match": 0.0,
|
| 723 |
+
"eval_prefix_exact_match": 0.026,
|
| 724 |
+
"step": 6147
|
| 725 |
+
},
|
| 726 |
+
{
|
| 727 |
+
"epoch": 18.15519765739385,
|
| 728 |
+
"grad_norm": 1.430617094039917,
|
| 729 |
+
"learning_rate": 0.0001,
|
| 730 |
+
"loss": 0.1717,
|
| 731 |
+
"step": 6200
|
| 732 |
+
},
|
| 733 |
+
{
|
| 734 |
+
"epoch": 18.448023426061493,
|
| 735 |
+
"grad_norm": 1.1885381937026978,
|
| 736 |
+
"learning_rate": 0.0001,
|
| 737 |
+
"loss": 0.1656,
|
| 738 |
+
"step": 6300
|
| 739 |
+
},
|
| 740 |
+
{
|
| 741 |
+
"epoch": 18.740849194729137,
|
| 742 |
+
"grad_norm": 1.64545738697052,
|
| 743 |
+
"learning_rate": 0.0001,
|
| 744 |
+
"loss": 0.1721,
|
| 745 |
+
"step": 6400
|
| 746 |
+
},
|
| 747 |
+
{
|
| 748 |
+
"epoch": 18.998535871156662,
|
| 749 |
+
"eval_accuracy": 0.4208717948717949,
|
| 750 |
+
"eval_loss": 6.557307720184326,
|
| 751 |
+
"eval_runtime": 5.092,
|
| 752 |
+
"eval_samples_per_second": 98.193,
|
| 753 |
+
"eval_steps_per_second": 12.372,
|
| 754 |
+
"step": 6488
|
| 755 |
+
},
|
| 756 |
+
{
|
| 757 |
+
"epoch": 18.998535871156662,
|
| 758 |
+
"eval_bleu": 0.16916503510298256,
|
| 759 |
+
"eval_exact_match": 0.0,
|
| 760 |
+
"eval_prefix_exact_match": 0.02,
|
| 761 |
+
"step": 6488
|
| 762 |
+
},
|
| 763 |
+
{
|
| 764 |
+
"epoch": 19.03367496339678,
|
| 765 |
+
"grad_norm": 1.2168452739715576,
|
| 766 |
+
"learning_rate": 0.0001,
|
| 767 |
+
"loss": 0.1743,
|
| 768 |
+
"step": 6500
|
| 769 |
+
},
|
| 770 |
+
{
|
| 771 |
+
"epoch": 19.32650073206442,
|
| 772 |
+
"grad_norm": 1.2320917844772339,
|
| 773 |
+
"learning_rate": 0.0001,
|
| 774 |
+
"loss": 0.152,
|
| 775 |
+
"step": 6600
|
| 776 |
+
},
|
| 777 |
+
{
|
| 778 |
+
"epoch": 19.619326500732065,
|
| 779 |
+
"grad_norm": 1.1541388034820557,
|
| 780 |
+
"learning_rate": 0.0001,
|
| 781 |
+
"loss": 0.1619,
|
| 782 |
+
"step": 6700
|
| 783 |
+
},
|
| 784 |
+
{
|
| 785 |
+
"epoch": 19.91215226939971,
|
| 786 |
+
"grad_norm": 1.2821346521377563,
|
| 787 |
+
"learning_rate": 0.0001,
|
| 788 |
+
"loss": 0.1676,
|
| 789 |
+
"step": 6800
|
| 790 |
+
},
|
| 791 |
+
{
|
| 792 |
+
"epoch": 20.0,
|
| 793 |
+
"eval_accuracy": 0.42138461538461536,
|
| 794 |
+
"eval_loss": 6.566097736358643,
|
| 795 |
+
"eval_runtime": 5.0056,
|
| 796 |
+
"eval_samples_per_second": 99.888,
|
| 797 |
+
"eval_steps_per_second": 12.586,
|
| 798 |
+
"step": 6830
|
| 799 |
+
},
|
| 800 |
+
{
|
| 801 |
+
"epoch": 20.0,
|
| 802 |
+
"eval_bleu": 0.21213643830050385,
|
| 803 |
+
"eval_exact_match": 0.0,
|
| 804 |
+
"eval_prefix_exact_match": 0.026,
|
| 805 |
+
"step": 6830
|
| 806 |
+
},
|
| 807 |
+
{
|
| 808 |
+
"epoch": 20.20497803806735,
|
| 809 |
+
"grad_norm": 1.1950445175170898,
|
| 810 |
+
"learning_rate": 0.0001,
|
| 811 |
+
"loss": 0.1533,
|
| 812 |
+
"step": 6900
|
| 813 |
+
},
|
| 814 |
+
{
|
| 815 |
+
"epoch": 20.497803806734993,
|
| 816 |
+
"grad_norm": 1.253625750541687,
|
| 817 |
+
"learning_rate": 0.0001,
|
| 818 |
+
"loss": 0.1522,
|
| 819 |
+
"step": 7000
|
| 820 |
+
},
|
| 821 |
+
{
|
| 822 |
+
"epoch": 20.790629575402637,
|
| 823 |
+
"grad_norm": 1.193354606628418,
|
| 824 |
+
"learning_rate": 0.0001,
|
| 825 |
+
"loss": 0.1579,
|
| 826 |
+
"step": 7100
|
| 827 |
+
},
|
| 828 |
+
{
|
| 829 |
+
"epoch": 20.998535871156662,
|
| 830 |
+
"eval_accuracy": 0.42133333333333334,
|
| 831 |
+
"eval_loss": 6.566288948059082,
|
| 832 |
+
"eval_runtime": 5.2039,
|
| 833 |
+
"eval_samples_per_second": 96.082,
|
| 834 |
+
"eval_steps_per_second": 12.106,
|
| 835 |
+
"step": 7171
|
| 836 |
+
},
|
| 837 |
+
{
|
| 838 |
+
"epoch": 20.998535871156662,
|
| 839 |
+
"eval_bleu": 0.16340256740662695,
|
| 840 |
+
"eval_exact_match": 0.0,
|
| 841 |
+
"eval_prefix_exact_match": 0.022,
|
| 842 |
+
"step": 7171
|
| 843 |
+
},
|
| 844 |
+
{
|
| 845 |
+
"epoch": 21.083455344070277,
|
| 846 |
+
"grad_norm": 1.0565904378890991,
|
| 847 |
+
"learning_rate": 0.0001,
|
| 848 |
+
"loss": 0.1575,
|
| 849 |
+
"step": 7200
|
| 850 |
+
},
|
| 851 |
+
{
|
| 852 |
+
"epoch": 21.37628111273792,
|
| 853 |
+
"grad_norm": 1.1256818771362305,
|
| 854 |
+
"learning_rate": 0.0001,
|
| 855 |
+
"loss": 0.1459,
|
| 856 |
+
"step": 7300
|
| 857 |
+
},
|
| 858 |
+
{
|
| 859 |
+
"epoch": 21.669106881405565,
|
| 860 |
+
"grad_norm": 1.4385403394699097,
|
| 861 |
+
"learning_rate": 0.0001,
|
| 862 |
+
"loss": 0.1511,
|
| 863 |
+
"step": 7400
|
| 864 |
+
},
|
| 865 |
+
{
|
| 866 |
+
"epoch": 21.961932650073205,
|
| 867 |
+
"grad_norm": 1.1152088642120361,
|
| 868 |
+
"learning_rate": 0.0001,
|
| 869 |
+
"loss": 0.1575,
|
| 870 |
+
"step": 7500
|
| 871 |
+
},
|
| 872 |
+
{
|
| 873 |
+
"epoch": 22.0,
|
| 874 |
+
"eval_accuracy": 0.42015384615384616,
|
| 875 |
+
"eval_loss": 6.625944137573242,
|
| 876 |
+
"eval_runtime": 5.0735,
|
| 877 |
+
"eval_samples_per_second": 98.552,
|
| 878 |
+
"eval_steps_per_second": 12.418,
|
| 879 |
+
"step": 7513
|
| 880 |
+
},
|
| 881 |
+
{
|
| 882 |
+
"epoch": 22.0,
|
| 883 |
+
"eval_bleu": 0.19792591482411717,
|
| 884 |
+
"eval_exact_match": 0.0,
|
| 885 |
+
"eval_prefix_exact_match": 0.022,
|
| 886 |
+
"step": 7513
|
| 887 |
+
},
|
| 888 |
+
{
|
| 889 |
+
"epoch": 22.25475841874085,
|
| 890 |
+
"grad_norm": 1.1440153121948242,
|
| 891 |
+
"learning_rate": 0.0001,
|
| 892 |
+
"loss": 0.1414,
|
| 893 |
+
"step": 7600
|
| 894 |
+
},
|
| 895 |
+
{
|
| 896 |
+
"epoch": 22.547584187408493,
|
| 897 |
+
"grad_norm": 1.1086493730545044,
|
| 898 |
+
"learning_rate": 0.0001,
|
| 899 |
+
"loss": 0.1435,
|
| 900 |
+
"step": 7700
|
| 901 |
+
},
|
| 902 |
+
{
|
| 903 |
+
"epoch": 22.840409956076133,
|
| 904 |
+
"grad_norm": 1.2570273876190186,
|
| 905 |
+
"learning_rate": 0.0001,
|
| 906 |
+
"loss": 0.15,
|
| 907 |
+
"step": 7800
|
| 908 |
+
},
|
| 909 |
+
{
|
| 910 |
+
"epoch": 22.998535871156662,
|
| 911 |
+
"eval_accuracy": 0.42143589743589743,
|
| 912 |
+
"eval_loss": 6.595519065856934,
|
| 913 |
+
"eval_runtime": 5.2943,
|
| 914 |
+
"eval_samples_per_second": 94.442,
|
| 915 |
+
"eval_steps_per_second": 11.9,
|
| 916 |
+
"step": 7854
|
| 917 |
+
},
|
| 918 |
+
{
|
| 919 |
+
"epoch": 22.998535871156662,
|
| 920 |
+
"eval_bleu": 0.17402208428859392,
|
| 921 |
+
"eval_exact_match": 0.002,
|
| 922 |
+
"eval_prefix_exact_match": 0.028,
|
| 923 |
+
"step": 7854
|
| 924 |
+
},
|
| 925 |
+
{
|
| 926 |
+
"epoch": 23.133235724743777,
|
| 927 |
+
"grad_norm": 1.070988655090332,
|
| 928 |
+
"learning_rate": 0.0001,
|
| 929 |
+
"loss": 0.1455,
|
| 930 |
+
"step": 7900
|
| 931 |
+
},
|
| 932 |
+
{
|
| 933 |
+
"epoch": 23.42606149341142,
|
| 934 |
+
"grad_norm": 1.021813154220581,
|
| 935 |
+
"learning_rate": 0.0001,
|
| 936 |
+
"loss": 0.138,
|
| 937 |
+
"step": 8000
|
| 938 |
+
},
|
| 939 |
+
{
|
| 940 |
+
"epoch": 23.718887262079065,
|
| 941 |
+
"grad_norm": 1.2333232164382935,
|
| 942 |
+
"learning_rate": 0.0001,
|
| 943 |
+
"loss": 0.1427,
|
| 944 |
+
"step": 8100
|
| 945 |
+
},
|
| 946 |
+
{
|
| 947 |
+
"epoch": 24.0,
|
| 948 |
+
"eval_accuracy": 0.4216410256410256,
|
| 949 |
+
"eval_loss": 6.62971830368042,
|
| 950 |
+
"eval_runtime": 4.9115,
|
| 951 |
+
"eval_samples_per_second": 101.802,
|
| 952 |
+
"eval_steps_per_second": 12.827,
|
| 953 |
+
"step": 8196
|
| 954 |
+
},
|
| 955 |
+
{
|
| 956 |
+
"epoch": 24.0,
|
| 957 |
+
"eval_bleu": 0.1531918581052099,
|
| 958 |
+
"eval_exact_match": 0.0,
|
| 959 |
+
"eval_prefix_exact_match": 0.028,
|
| 960 |
+
"step": 8196
|
| 961 |
+
},
|
| 962 |
+
{
|
| 963 |
+
"epoch": 24.011713030746705,
|
| 964 |
+
"grad_norm": 0.9755038022994995,
|
| 965 |
+
"learning_rate": 0.0001,
|
| 966 |
+
"loss": 0.1488,
|
| 967 |
+
"step": 8200
|
| 968 |
+
},
|
| 969 |
+
{
|
| 970 |
+
"epoch": 24.30453879941435,
|
| 971 |
+
"grad_norm": 1.2587764263153076,
|
| 972 |
+
"learning_rate": 0.0001,
|
| 973 |
+
"loss": 0.1328,
|
| 974 |
+
"step": 8300
|
| 975 |
+
},
|
| 976 |
+
{
|
| 977 |
+
"epoch": 24.597364568081993,
|
| 978 |
+
"grad_norm": 0.9521226286888123,
|
| 979 |
+
"learning_rate": 0.0001,
|
| 980 |
+
"loss": 0.1384,
|
| 981 |
+
"step": 8400
|
| 982 |
+
},
|
| 983 |
+
{
|
| 984 |
+
"epoch": 24.890190336749633,
|
| 985 |
+
"grad_norm": 1.2158619165420532,
|
| 986 |
+
"learning_rate": 0.0001,
|
| 987 |
+
"loss": 0.145,
|
| 988 |
+
"step": 8500
|
| 989 |
+
},
|
| 990 |
+
{
|
| 991 |
+
"epoch": 24.998535871156662,
|
| 992 |
+
"eval_accuracy": 0.4226666666666667,
|
| 993 |
+
"eval_loss": 6.57574462890625,
|
| 994 |
+
"eval_runtime": 5.0443,
|
| 995 |
+
"eval_samples_per_second": 99.122,
|
| 996 |
+
"eval_steps_per_second": 12.489,
|
| 997 |
+
"step": 8537
|
| 998 |
+
},
|
| 999 |
+
{
|
| 1000 |
+
"epoch": 24.998535871156662,
|
| 1001 |
+
"eval_bleu": 0.17454308300545518,
|
| 1002 |
+
"eval_exact_match": 0.0,
|
| 1003 |
+
"eval_prefix_exact_match": 0.022,
|
| 1004 |
+
"step": 8537
|
| 1005 |
+
},
|
| 1006 |
+
{
|
| 1007 |
+
"epoch": 25.183016105417277,
|
| 1008 |
+
"grad_norm": 1.0446624755859375,
|
| 1009 |
+
"learning_rate": 0.0001,
|
| 1010 |
+
"loss": 0.1357,
|
| 1011 |
+
"step": 8600
|
| 1012 |
+
},
|
| 1013 |
+
{
|
| 1014 |
+
"epoch": 25.47584187408492,
|
| 1015 |
+
"grad_norm": 1.07607901096344,
|
| 1016 |
+
"learning_rate": 0.0001,
|
| 1017 |
+
"loss": 0.1341,
|
| 1018 |
+
"step": 8700
|
| 1019 |
+
},
|
| 1020 |
+
{
|
| 1021 |
+
"epoch": 25.76866764275256,
|
| 1022 |
+
"grad_norm": 1.0376451015472412,
|
| 1023 |
+
"learning_rate": 0.0001,
|
| 1024 |
+
"loss": 0.1393,
|
| 1025 |
+
"step": 8800
|
| 1026 |
+
},
|
| 1027 |
+
{
|
| 1028 |
+
"epoch": 26.0,
|
| 1029 |
+
"eval_accuracy": 0.42133333333333334,
|
| 1030 |
+
"eval_loss": 6.567473411560059,
|
| 1031 |
+
"eval_runtime": 5.3766,
|
| 1032 |
+
"eval_samples_per_second": 92.995,
|
| 1033 |
+
"eval_steps_per_second": 11.717,
|
| 1034 |
+
"step": 8879
|
| 1035 |
+
},
|
| 1036 |
+
{
|
| 1037 |
+
"epoch": 26.0,
|
| 1038 |
+
"eval_bleu": 0.183512215739427,
|
| 1039 |
+
"eval_exact_match": 0.0,
|
| 1040 |
+
"eval_prefix_exact_match": 0.026,
|
| 1041 |
+
"step": 8879
|
| 1042 |
+
},
|
| 1043 |
+
{
|
| 1044 |
+
"epoch": 26.061493411420205,
|
| 1045 |
+
"grad_norm": 0.8689504861831665,
|
| 1046 |
+
"learning_rate": 0.0001,
|
| 1047 |
+
"loss": 0.1415,
|
| 1048 |
+
"step": 8900
|
| 1049 |
+
},
|
| 1050 |
+
{
|
| 1051 |
+
"epoch": 26.35431918008785,
|
| 1052 |
+
"grad_norm": 0.8533344864845276,
|
| 1053 |
+
"learning_rate": 0.0001,
|
| 1054 |
+
"loss": 0.1292,
|
| 1055 |
+
"step": 9000
|
| 1056 |
+
},
|
| 1057 |
+
{
|
| 1058 |
+
"epoch": 26.64714494875549,
|
| 1059 |
+
"grad_norm": 1.0425468683242798,
|
| 1060 |
+
"learning_rate": 0.0001,
|
| 1061 |
+
"loss": 0.1348,
|
| 1062 |
+
"step": 9100
|
| 1063 |
+
},
|
| 1064 |
+
{
|
| 1065 |
+
"epoch": 26.939970717423133,
|
| 1066 |
+
"grad_norm": 1.2276948690414429,
|
| 1067 |
+
"learning_rate": 0.0001,
|
| 1068 |
+
"loss": 0.1405,
|
| 1069 |
+
"step": 9200
|
| 1070 |
+
},
|
| 1071 |
+
{
|
| 1072 |
+
"epoch": 26.998535871156662,
|
| 1073 |
+
"eval_accuracy": 0.42133333333333334,
|
| 1074 |
+
"eval_loss": 6.664970874786377,
|
| 1075 |
+
"eval_runtime": 5.6177,
|
| 1076 |
+
"eval_samples_per_second": 89.004,
|
| 1077 |
+
"eval_steps_per_second": 11.215,
|
| 1078 |
+
"step": 9220
|
| 1079 |
+
},
|
| 1080 |
+
{
|
| 1081 |
+
"epoch": 26.998535871156662,
|
| 1082 |
+
"eval_bleu": 0.19671240169357163,
|
| 1083 |
+
"eval_exact_match": 0.002,
|
| 1084 |
+
"eval_prefix_exact_match": 0.026,
|
| 1085 |
+
"step": 9220
|
| 1086 |
+
},
|
| 1087 |
+
{
|
| 1088 |
+
"epoch": 27.232796486090777,
|
| 1089 |
+
"grad_norm": 1.0451931953430176,
|
| 1090 |
+
"learning_rate": 0.0001,
|
| 1091 |
+
"loss": 0.1282,
|
| 1092 |
+
"step": 9300
|
| 1093 |
+
},
|
| 1094 |
+
{
|
| 1095 |
+
"epoch": 27.525622254758417,
|
| 1096 |
+
"grad_norm": 1.0023565292358398,
|
| 1097 |
+
"learning_rate": 0.0001,
|
| 1098 |
+
"loss": 0.1307,
|
| 1099 |
+
"step": 9400
|
| 1100 |
+
},
|
| 1101 |
+
{
|
| 1102 |
+
"epoch": 27.81844802342606,
|
| 1103 |
+
"grad_norm": 0.9287552833557129,
|
| 1104 |
+
"learning_rate": 0.0001,
|
| 1105 |
+
"loss": 0.1365,
|
| 1106 |
+
"step": 9500
|
| 1107 |
+
},
|
| 1108 |
+
{
|
| 1109 |
+
"epoch": 28.0,
|
| 1110 |
+
"eval_accuracy": 0.420974358974359,
|
| 1111 |
+
"eval_loss": 6.642728328704834,
|
| 1112 |
+
"eval_runtime": 4.9204,
|
| 1113 |
+
"eval_samples_per_second": 101.618,
|
| 1114 |
+
"eval_steps_per_second": 12.804,
|
| 1115 |
+
"step": 9562
|
| 1116 |
+
},
|
| 1117 |
+
{
|
| 1118 |
+
"epoch": 28.0,
|
| 1119 |
+
"eval_bleu": 0.17149559728372338,
|
| 1120 |
+
"eval_exact_match": 0.0,
|
| 1121 |
+
"eval_prefix_exact_match": 0.028,
|
| 1122 |
+
"step": 9562
|
| 1123 |
+
},
|
| 1124 |
+
{
|
| 1125 |
+
"epoch": 28.111273792093705,
|
| 1126 |
+
"grad_norm": 0.7279195785522461,
|
| 1127 |
+
"learning_rate": 0.0001,
|
| 1128 |
+
"loss": 0.133,
|
| 1129 |
+
"step": 9600
|
| 1130 |
+
},
|
| 1131 |
+
{
|
| 1132 |
+
"epoch": 28.404099560761345,
|
| 1133 |
+
"grad_norm": 0.9348853230476379,
|
| 1134 |
+
"learning_rate": 0.0001,
|
| 1135 |
+
"loss": 0.1268,
|
| 1136 |
+
"step": 9700
|
| 1137 |
+
},
|
| 1138 |
+
{
|
| 1139 |
+
"epoch": 28.69692532942899,
|
| 1140 |
+
"grad_norm": 0.9402393102645874,
|
| 1141 |
+
"learning_rate": 0.0001,
|
| 1142 |
+
"loss": 0.1323,
|
| 1143 |
+
"step": 9800
|
| 1144 |
+
},
|
| 1145 |
+
{
|
| 1146 |
+
"epoch": 28.989751098096633,
|
| 1147 |
+
"grad_norm": 0.9316591620445251,
|
| 1148 |
+
"learning_rate": 0.0001,
|
| 1149 |
+
"loss": 0.1372,
|
| 1150 |
+
"step": 9900
|
| 1151 |
+
},
|
| 1152 |
+
{
|
| 1153 |
+
"epoch": 28.998535871156662,
|
| 1154 |
+
"eval_accuracy": 0.4208717948717949,
|
| 1155 |
+
"eval_loss": 6.548105239868164,
|
| 1156 |
+
"eval_runtime": 4.8211,
|
| 1157 |
+
"eval_samples_per_second": 103.71,
|
| 1158 |
+
"eval_steps_per_second": 13.067,
|
| 1159 |
+
"step": 9903
|
| 1160 |
+
},
|
| 1161 |
+
{
|
| 1162 |
+
"epoch": 28.998535871156662,
|
| 1163 |
+
"eval_bleu": 0.16884007365872924,
|
| 1164 |
+
"eval_exact_match": 0.0,
|
| 1165 |
+
"eval_prefix_exact_match": 0.02,
|
| 1166 |
+
"step": 9903
|
| 1167 |
+
},
|
| 1168 |
+
{
|
| 1169 |
+
"epoch": 29.282576866764277,
|
| 1170 |
+
"grad_norm": 0.9301176071166992,
|
| 1171 |
+
"learning_rate": 0.0001,
|
| 1172 |
+
"loss": 0.1234,
|
| 1173 |
+
"step": 10000
|
| 1174 |
+
},
|
| 1175 |
+
{
|
| 1176 |
+
"epoch": 29.575402635431917,
|
| 1177 |
+
"grad_norm": 0.8069124221801758,
|
| 1178 |
+
"learning_rate": 0.0001,
|
| 1179 |
+
"loss": 0.1287,
|
| 1180 |
+
"step": 10100
|
| 1181 |
+
},
|
| 1182 |
+
{
|
| 1183 |
+
"epoch": 29.86822840409956,
|
| 1184 |
+
"grad_norm": 1.0093735456466675,
|
| 1185 |
+
"learning_rate": 0.0001,
|
| 1186 |
+
"loss": 0.134,
|
| 1187 |
+
"step": 10200
|
| 1188 |
+
},
|
| 1189 |
+
{
|
| 1190 |
+
"epoch": 30.0,
|
| 1191 |
+
"eval_accuracy": 0.41994871794871796,
|
| 1192 |
+
"eval_loss": 6.661746025085449,
|
| 1193 |
+
"eval_runtime": 4.8287,
|
| 1194 |
+
"eval_samples_per_second": 103.548,
|
| 1195 |
+
"eval_steps_per_second": 13.047,
|
| 1196 |
+
"step": 10245
|
| 1197 |
+
},
|
| 1198 |
+
{
|
| 1199 |
+
"epoch": 30.0,
|
| 1200 |
+
"eval_bleu": 0.19612712097933296,
|
| 1201 |
+
"eval_exact_match": 0.0,
|
| 1202 |
+
"eval_prefix_exact_match": 0.028,
|
| 1203 |
+
"step": 10245
|
| 1204 |
+
},
|
| 1205 |
+
{
|
| 1206 |
+
"epoch": 30.161054172767205,
|
| 1207 |
+
"grad_norm": 0.965752899646759,
|
| 1208 |
+
"learning_rate": 0.0001,
|
| 1209 |
+
"loss": 0.1268,
|
| 1210 |
+
"step": 10300
|
| 1211 |
+
},
|
| 1212 |
+
{
|
| 1213 |
+
"epoch": 30.453879941434845,
|
| 1214 |
+
"grad_norm": 0.9576678276062012,
|
| 1215 |
+
"learning_rate": 0.0001,
|
| 1216 |
+
"loss": 0.1241,
|
| 1217 |
+
"step": 10400
|
| 1218 |
+
},
|
| 1219 |
+
{
|
| 1220 |
+
"epoch": 30.74670571010249,
|
| 1221 |
+
"grad_norm": 0.8335288763046265,
|
| 1222 |
+
"learning_rate": 0.0001,
|
| 1223 |
+
"loss": 0.1287,
|
| 1224 |
+
"step": 10500
|
| 1225 |
+
},
|
| 1226 |
+
{
|
| 1227 |
+
"epoch": 30.998535871156662,
|
| 1228 |
+
"eval_accuracy": 0.4206666666666667,
|
| 1229 |
+
"eval_loss": 6.624051094055176,
|
| 1230 |
+
"eval_runtime": 5.0567,
|
| 1231 |
+
"eval_samples_per_second": 98.879,
|
| 1232 |
+
"eval_steps_per_second": 12.459,
|
| 1233 |
+
"step": 10586
|
| 1234 |
+
},
|
| 1235 |
+
{
|
| 1236 |
+
"epoch": 30.998535871156662,
|
| 1237 |
+
"eval_bleu": 0.17063063536886222,
|
| 1238 |
+
"eval_exact_match": 0.0,
|
| 1239 |
+
"eval_prefix_exact_match": 0.026,
|
| 1240 |
+
"step": 10586
|
| 1241 |
+
},
|
| 1242 |
+
{
|
| 1243 |
+
"epoch": 31.039531478770133,
|
| 1244 |
+
"grad_norm": 0.731296718120575,
|
| 1245 |
+
"learning_rate": 0.0001,
|
| 1246 |
+
"loss": 0.1327,
|
| 1247 |
+
"step": 10600
|
| 1248 |
+
},
|
| 1249 |
+
{
|
| 1250 |
+
"epoch": 31.332357247437773,
|
| 1251 |
+
"grad_norm": 0.9108264446258545,
|
| 1252 |
+
"learning_rate": 0.0001,
|
| 1253 |
+
"loss": 0.1204,
|
| 1254 |
+
"step": 10700
|
| 1255 |
+
},
|
| 1256 |
+
{
|
| 1257 |
+
"epoch": 31.625183016105417,
|
| 1258 |
+
"grad_norm": 0.9855822324752808,
|
| 1259 |
+
"learning_rate": 0.0001,
|
| 1260 |
+
"loss": 0.1259,
|
| 1261 |
+
"step": 10800
|
| 1262 |
+
},
|
| 1263 |
+
{
|
| 1264 |
+
"epoch": 31.91800878477306,
|
| 1265 |
+
"grad_norm": 1.1106101274490356,
|
| 1266 |
+
"learning_rate": 0.0001,
|
| 1267 |
+
"loss": 0.1305,
|
| 1268 |
+
"step": 10900
|
| 1269 |
+
},
|
| 1270 |
+
{
|
| 1271 |
+
"epoch": 32.0,
|
| 1272 |
+
"eval_accuracy": 0.41994871794871796,
|
| 1273 |
+
"eval_loss": 6.609373092651367,
|
| 1274 |
+
"eval_runtime": 5.0029,
|
| 1275 |
+
"eval_samples_per_second": 99.941,
|
| 1276 |
+
"eval_steps_per_second": 12.593,
|
| 1277 |
+
"step": 10928
|
| 1278 |
+
},
|
| 1279 |
+
{
|
| 1280 |
+
"epoch": 32.0,
|
| 1281 |
+
"eval_bleu": 0.1600123189131272,
|
| 1282 |
+
"eval_exact_match": 0.002,
|
| 1283 |
+
"eval_prefix_exact_match": 0.03,
|
| 1284 |
+
"step": 10928
|
| 1285 |
+
},
|
| 1286 |
+
{
|
| 1287 |
+
"epoch": 32.210834553440705,
|
| 1288 |
+
"grad_norm": 0.8511720895767212,
|
| 1289 |
+
"learning_rate": 0.0001,
|
| 1290 |
+
"loss": 0.1216,
|
| 1291 |
+
"step": 11000
|
| 1292 |
+
},
|
| 1293 |
+
{
|
| 1294 |
+
"epoch": 32.503660322108345,
|
| 1295 |
+
"grad_norm": 1.0120123624801636,
|
| 1296 |
+
"learning_rate": 0.0001,
|
| 1297 |
+
"loss": 0.1225,
|
| 1298 |
+
"step": 11100
|
| 1299 |
+
},
|
| 1300 |
+
{
|
| 1301 |
+
"epoch": 32.796486090775986,
|
| 1302 |
+
"grad_norm": 0.9636625051498413,
|
| 1303 |
+
"learning_rate": 0.0001,
|
| 1304 |
+
"loss": 0.1274,
|
| 1305 |
+
"step": 11200
|
| 1306 |
+
},
|
| 1307 |
+
{
|
| 1308 |
+
"epoch": 32.99853587115666,
|
| 1309 |
+
"eval_accuracy": 0.4165128205128205,
|
| 1310 |
+
"eval_loss": 6.682253360748291,
|
| 1311 |
+
"eval_runtime": 5.2415,
|
| 1312 |
+
"eval_samples_per_second": 95.392,
|
| 1313 |
+
"eval_steps_per_second": 12.019,
|
| 1314 |
+
"step": 11269
|
| 1315 |
+
},
|
| 1316 |
+
{
|
| 1317 |
+
"epoch": 32.99853587115666,
|
| 1318 |
+
"eval_bleu": 0.21084378912778606,
|
| 1319 |
+
"eval_exact_match": 0.002,
|
| 1320 |
+
"eval_prefix_exact_match": 0.022,
|
| 1321 |
+
"step": 11269
|
| 1322 |
+
},
|
| 1323 |
+
{
|
| 1324 |
+
"epoch": 33.08931185944363,
|
| 1325 |
+
"grad_norm": 0.991923451423645,
|
| 1326 |
+
"learning_rate": 0.0001,
|
| 1327 |
+
"loss": 0.1271,
|
| 1328 |
+
"step": 11300
|
| 1329 |
+
},
|
| 1330 |
+
{
|
| 1331 |
+
"epoch": 33.38213762811127,
|
| 1332 |
+
"grad_norm": 0.8520097136497498,
|
| 1333 |
+
"learning_rate": 0.0001,
|
| 1334 |
+
"loss": 0.1195,
|
| 1335 |
+
"step": 11400
|
| 1336 |
+
},
|
| 1337 |
+
{
|
| 1338 |
+
"epoch": 33.674963396778914,
|
| 1339 |
+
"grad_norm": 1.1480833292007446,
|
| 1340 |
+
"learning_rate": 0.0001,
|
| 1341 |
+
"loss": 0.1242,
|
| 1342 |
+
"step": 11500
|
| 1343 |
+
},
|
| 1344 |
+
{
|
| 1345 |
+
"epoch": 33.96778916544656,
|
| 1346 |
+
"grad_norm": 0.8655959367752075,
|
| 1347 |
+
"learning_rate": 0.0001,
|
| 1348 |
+
"loss": 0.1296,
|
| 1349 |
+
"step": 11600
|
| 1350 |
+
},
|
| 1351 |
+
{
|
| 1352 |
+
"epoch": 34.0,
|
| 1353 |
+
"eval_accuracy": 0.4195384615384615,
|
| 1354 |
+
"eval_loss": 6.621041774749756,
|
| 1355 |
+
"eval_runtime": 5.1712,
|
| 1356 |
+
"eval_samples_per_second": 96.689,
|
| 1357 |
+
"eval_steps_per_second": 12.183,
|
| 1358 |
+
"step": 11611
|
| 1359 |
+
},
|
| 1360 |
+
{
|
| 1361 |
+
"epoch": 34.0,
|
| 1362 |
+
"eval_bleu": 0.2088991994544206,
|
| 1363 |
+
"eval_exact_match": 0.0,
|
| 1364 |
+
"eval_prefix_exact_match": 0.02,
|
| 1365 |
+
"step": 11611
|
| 1366 |
+
},
|
| 1367 |
+
{
|
| 1368 |
+
"epoch": 34.2606149341142,
|
| 1369 |
+
"grad_norm": 0.7514725923538208,
|
| 1370 |
+
"learning_rate": 0.0001,
|
| 1371 |
+
"loss": 0.1199,
|
| 1372 |
+
"step": 11700
|
| 1373 |
+
},
|
| 1374 |
+
{
|
| 1375 |
+
"epoch": 34.55344070278184,
|
| 1376 |
+
"grad_norm": 0.9057220220565796,
|
| 1377 |
+
"learning_rate": 0.0001,
|
| 1378 |
+
"loss": 0.121,
|
| 1379 |
+
"step": 11800
|
| 1380 |
+
},
|
| 1381 |
+
{
|
| 1382 |
+
"epoch": 34.84626647144949,
|
| 1383 |
+
"grad_norm": 0.8322867155075073,
|
| 1384 |
+
"learning_rate": 0.0001,
|
| 1385 |
+
"loss": 0.1271,
|
| 1386 |
+
"step": 11900
|
| 1387 |
+
},
|
| 1388 |
+
{
|
| 1389 |
+
"epoch": 34.99853587115666,
|
| 1390 |
+
"eval_accuracy": 0.41846153846153844,
|
| 1391 |
+
"eval_loss": 6.704195976257324,
|
| 1392 |
+
"eval_runtime": 5.1841,
|
| 1393 |
+
"eval_samples_per_second": 96.448,
|
| 1394 |
+
"eval_steps_per_second": 12.153,
|
| 1395 |
+
"step": 11952
|
| 1396 |
+
},
|
| 1397 |
+
{
|
| 1398 |
+
"epoch": 34.99853587115666,
|
| 1399 |
+
"eval_bleu": 0.17311282679501735,
|
| 1400 |
+
"eval_exact_match": 0.0,
|
| 1401 |
+
"eval_prefix_exact_match": 0.022,
|
| 1402 |
+
"step": 11952
|
| 1403 |
+
},
|
| 1404 |
+
{
|
| 1405 |
+
"epoch": 35.13909224011713,
|
| 1406 |
+
"grad_norm": 0.9303777813911438,
|
| 1407 |
+
"learning_rate": 0.0001,
|
| 1408 |
+
"loss": 0.1231,
|
| 1409 |
+
"step": 12000
|
| 1410 |
+
},
|
| 1411 |
+
{
|
| 1412 |
+
"epoch": 35.43191800878477,
|
| 1413 |
+
"grad_norm": 0.8847092986106873,
|
| 1414 |
+
"learning_rate": 0.0001,
|
| 1415 |
+
"loss": 0.1201,
|
| 1416 |
+
"step": 12100
|
| 1417 |
+
},
|
| 1418 |
+
{
|
| 1419 |
+
"epoch": 35.72474377745242,
|
| 1420 |
+
"grad_norm": 1.0509458780288696,
|
| 1421 |
+
"learning_rate": 0.0001,
|
| 1422 |
+
"loss": 0.1239,
|
| 1423 |
+
"step": 12200
|
| 1424 |
+
},
|
| 1425 |
+
{
|
| 1426 |
+
"epoch": 36.0,
|
| 1427 |
+
"eval_accuracy": 0.42035897435897435,
|
| 1428 |
+
"eval_loss": 6.601570129394531,
|
| 1429 |
+
"eval_runtime": 4.9227,
|
| 1430 |
+
"eval_samples_per_second": 101.571,
|
| 1431 |
+
"eval_steps_per_second": 12.798,
|
| 1432 |
+
"step": 12294
|
| 1433 |
+
},
|
| 1434 |
+
{
|
| 1435 |
+
"epoch": 36.0,
|
| 1436 |
+
"eval_bleu": 0.19246146326233543,
|
| 1437 |
+
"eval_exact_match": 0.0,
|
| 1438 |
+
"eval_prefix_exact_match": 0.022,
|
| 1439 |
+
"step": 12294
|
| 1440 |
+
},
|
| 1441 |
+
{
|
| 1442 |
+
"epoch": 36.01756954612006,
|
| 1443 |
+
"grad_norm": 1.0885076522827148,
|
| 1444 |
+
"learning_rate": 0.0001,
|
| 1445 |
+
"loss": 0.1262,
|
| 1446 |
+
"step": 12300
|
| 1447 |
+
},
|
| 1448 |
+
{
|
| 1449 |
+
"epoch": 36.3103953147877,
|
| 1450 |
+
"grad_norm": 0.8121168613433838,
|
| 1451 |
+
"learning_rate": 0.0001,
|
| 1452 |
+
"loss": 0.1153,
|
| 1453 |
+
"step": 12400
|
| 1454 |
+
},
|
| 1455 |
+
{
|
| 1456 |
+
"epoch": 36.603221083455345,
|
| 1457 |
+
"grad_norm": 0.7476085424423218,
|
| 1458 |
+
"learning_rate": 0.0001,
|
| 1459 |
+
"loss": 0.1212,
|
| 1460 |
+
"step": 12500
|
| 1461 |
+
},
|
| 1462 |
+
{
|
| 1463 |
+
"epoch": 36.896046852122986,
|
| 1464 |
+
"grad_norm": 1.0950144529342651,
|
| 1465 |
+
"learning_rate": 0.0001,
|
| 1466 |
+
"loss": 0.1263,
|
| 1467 |
+
"step": 12600
|
| 1468 |
+
},
|
| 1469 |
+
{
|
| 1470 |
+
"epoch": 36.99853587115666,
|
| 1471 |
+
"eval_accuracy": 0.4195384615384615,
|
| 1472 |
+
"eval_loss": 6.573636531829834,
|
| 1473 |
+
"eval_runtime": 4.8706,
|
| 1474 |
+
"eval_samples_per_second": 102.657,
|
| 1475 |
+
"eval_steps_per_second": 12.935,
|
| 1476 |
+
"step": 12635
|
| 1477 |
+
},
|
| 1478 |
+
{
|
| 1479 |
+
"epoch": 36.99853587115666,
|
| 1480 |
+
"eval_bleu": 0.11460835767059586,
|
| 1481 |
+
"eval_exact_match": 0.0,
|
| 1482 |
+
"eval_prefix_exact_match": 0.022,
|
| 1483 |
+
"step": 12635
|
| 1484 |
+
},
|
| 1485 |
+
{
|
| 1486 |
+
"epoch": 37.18887262079063,
|
| 1487 |
+
"grad_norm": 1.199575662612915,
|
| 1488 |
+
"learning_rate": 0.0001,
|
| 1489 |
+
"loss": 0.1187,
|
| 1490 |
+
"step": 12700
|
| 1491 |
+
},
|
| 1492 |
+
{
|
| 1493 |
+
"epoch": 37.48169838945827,
|
| 1494 |
+
"grad_norm": 0.6631502509117126,
|
| 1495 |
+
"learning_rate": 0.0001,
|
| 1496 |
+
"loss": 0.1189,
|
| 1497 |
+
"step": 12800
|
| 1498 |
+
},
|
| 1499 |
+
{
|
| 1500 |
+
"epoch": 37.774524158125914,
|
| 1501 |
+
"grad_norm": 1.0126887559890747,
|
| 1502 |
+
"learning_rate": 0.0001,
|
| 1503 |
+
"loss": 0.1234,
|
| 1504 |
+
"step": 12900
|
| 1505 |
+
},
|
| 1506 |
+
{
|
| 1507 |
+
"epoch": 38.0,
|
| 1508 |
+
"eval_accuracy": 0.4169230769230769,
|
| 1509 |
+
"eval_loss": 6.609395503997803,
|
| 1510 |
+
"eval_runtime": 5.0108,
|
| 1511 |
+
"eval_samples_per_second": 99.784,
|
| 1512 |
+
"eval_steps_per_second": 12.573,
|
| 1513 |
+
"step": 12977
|
| 1514 |
+
},
|
| 1515 |
+
{
|
| 1516 |
+
"epoch": 38.0,
|
| 1517 |
+
"eval_bleu": 0.13428219944335265,
|
| 1518 |
+
"eval_exact_match": 0.002,
|
| 1519 |
+
"eval_prefix_exact_match": 0.024,
|
| 1520 |
+
"step": 12977
|
| 1521 |
+
}
|
| 1522 |
+
],
|
| 1523 |
+
"logging_steps": 100,
|
| 1524 |
+
"max_steps": 17050,
|
| 1525 |
+
"num_input_tokens_seen": 0,
|
| 1526 |
+
"num_train_epochs": 50,
|
| 1527 |
+
"save_steps": 500,
|
| 1528 |
+
"stateful_callbacks": {
|
| 1529 |
+
"TrainerControl": {
|
| 1530 |
+
"args": {
|
| 1531 |
+
"should_epoch_stop": false,
|
| 1532 |
+
"should_evaluate": false,
|
| 1533 |
+
"should_log": false,
|
| 1534 |
+
"should_save": true,
|
| 1535 |
+
"should_training_stop": false
|
| 1536 |
+
},
|
| 1537 |
+
"attributes": {}
|
| 1538 |
+
}
|
| 1539 |
+
},
|
| 1540 |
+
"total_flos": 1.114356674874835e+18,
|
| 1541 |
+
"train_batch_size": 1,
|
| 1542 |
+
"trial_name": null,
|
| 1543 |
+
"trial_params": null
|
| 1544 |
+
}
|
checkpoint-12977/training_args.bin
ADDED
|
@@ -0,0 +1,3 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
version https://git-lfs.github.com/spec/v1
|
| 2 |
+
oid sha256:7e63e74298c414f7364874e60281a380506bddfac466373e2426a740b4eb5594
|
| 3 |
+
size 5304
|
checkpoint-12977/vocab.json
ADDED
|
The diff for this file is too large to render.
See raw diff
|
|
|