Training in progress, epoch 42, checkpoint
Browse files- checkpoint-8062/README.md +9 -0
- checkpoint-8062/adapter_config.json +25 -0
- checkpoint-8062/adapter_model.safetensors +3 -0
- checkpoint-8062/added_tokens.json +5 -0
- checkpoint-8062/merges.txt +0 -0
- checkpoint-8062/optimizer.pt +3 -0
- checkpoint-8062/rng_state_0.pth +3 -0
- checkpoint-8062/rng_state_1.pth +3 -0
- checkpoint-8062/rng_state_2.pth +3 -0
- checkpoint-8062/rng_state_3.pth +3 -0
- checkpoint-8062/scheduler.pt +3 -0
- checkpoint-8062/special_tokens_map.json +14 -0
- checkpoint-8062/tokenizer.json +0 -0
- checkpoint-8062/tokenizer_config.json +43 -0
- checkpoint-8062/trainer_state.json +1238 -0
- checkpoint-8062/training_args.bin +3 -0
- checkpoint-8062/vocab.json +0 -0
checkpoint-8062/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-8062/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-8062/adapter_model.safetensors
ADDED
|
@@ -0,0 +1,3 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
version https://git-lfs.github.com/spec/v1
|
| 2 |
+
oid sha256:ce9d5aba0b1d610624f916f6e001968243fdcdfded226182792d4bc41c56c929
|
| 3 |
+
size 112130416
|
checkpoint-8062/added_tokens.json
ADDED
|
@@ -0,0 +1,5 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
{
|
| 2 |
+
"<|endoftext|>": 151643,
|
| 3 |
+
"<|im_end|>": 151645,
|
| 4 |
+
"<|im_start|>": 151644
|
| 5 |
+
}
|
checkpoint-8062/merges.txt
ADDED
|
The diff for this file is too large to render.
See raw diff
|
|
|
checkpoint-8062/optimizer.pt
ADDED
|
@@ -0,0 +1,3 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
version https://git-lfs.github.com/spec/v1
|
| 2 |
+
oid sha256:7098ce1bf5c6b748c9d4089f95fe70d0396c54c4384e6870e5edc85bc0d56f32
|
| 3 |
+
size 224537202
|
checkpoint-8062/rng_state_0.pth
ADDED
|
@@ -0,0 +1,3 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
version https://git-lfs.github.com/spec/v1
|
| 2 |
+
oid sha256:b40ef5cc760504b96f18639f0a8441a5fd29f862cf546014003e36e0cdea3024
|
| 3 |
+
size 15024
|
checkpoint-8062/rng_state_1.pth
ADDED
|
@@ -0,0 +1,3 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
version https://git-lfs.github.com/spec/v1
|
| 2 |
+
oid sha256:ac517313c8100db39b21072bb674b0c34b590fb47f4588497ec7f70ad77a5f46
|
| 3 |
+
size 15024
|
checkpoint-8062/rng_state_2.pth
ADDED
|
@@ -0,0 +1,3 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
version https://git-lfs.github.com/spec/v1
|
| 2 |
+
oid sha256:dff1b87f7f4c08a5af39e7f4caf8849a19172c18e3195e90a3660641b481192a
|
| 3 |
+
size 15024
|
checkpoint-8062/rng_state_3.pth
ADDED
|
@@ -0,0 +1,3 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
version https://git-lfs.github.com/spec/v1
|
| 2 |
+
oid sha256:e2170a5846e1f836bd88f64fde56635c6c389cda1e2b308962ad4ee0bba8d11b
|
| 3 |
+
size 15024
|
checkpoint-8062/scheduler.pt
ADDED
|
@@ -0,0 +1,3 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
version https://git-lfs.github.com/spec/v1
|
| 2 |
+
oid sha256:d06ff0a5f450667aa0c2754d468a57ab84d750add80e65ed0786a9f34bf93284
|
| 3 |
+
size 1064
|
checkpoint-8062/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-8062/tokenizer.json
ADDED
|
The diff for this file is too large to render.
See raw diff
|
|
|
checkpoint-8062/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-8062/trainer_state.json
ADDED
|
@@ -0,0 +1,1238 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
{
|
| 2 |
+
"best_metric": null,
|
| 3 |
+
"best_model_checkpoint": null,
|
| 4 |
+
"epoch": 42.99733333333333,
|
| 5 |
+
"eval_steps": 500,
|
| 6 |
+
"global_step": 8062,
|
| 7 |
+
"is_hyper_param_search": false,
|
| 8 |
+
"is_local_process_zero": true,
|
| 9 |
+
"is_world_process_zero": true,
|
| 10 |
+
"log_history": [
|
| 11 |
+
{
|
| 12 |
+
"epoch": 0.5333333333333333,
|
| 13 |
+
"grad_norm": 0.7527768015861511,
|
| 14 |
+
"learning_rate": 0.0005,
|
| 15 |
+
"loss": 1.7217,
|
| 16 |
+
"step": 100
|
| 17 |
+
},
|
| 18 |
+
{
|
| 19 |
+
"epoch": 0.9973333333333333,
|
| 20 |
+
"eval_accuracy": 0.5738974358974359,
|
| 21 |
+
"eval_loss": 1.6209890842437744,
|
| 22 |
+
"eval_runtime": 5.1723,
|
| 23 |
+
"eval_samples_per_second": 96.668,
|
| 24 |
+
"eval_steps_per_second": 12.18,
|
| 25 |
+
"step": 187
|
| 26 |
+
},
|
| 27 |
+
{
|
| 28 |
+
"epoch": 0.9973333333333333,
|
| 29 |
+
"eval_exact_match": 18.2,
|
| 30 |
+
"eval_f1": 27.000476190476196,
|
| 31 |
+
"step": 187
|
| 32 |
+
},
|
| 33 |
+
{
|
| 34 |
+
"epoch": 1.0666666666666667,
|
| 35 |
+
"grad_norm": 0.7723727226257324,
|
| 36 |
+
"learning_rate": 0.0005,
|
| 37 |
+
"loss": 1.615,
|
| 38 |
+
"step": 200
|
| 39 |
+
},
|
| 40 |
+
{
|
| 41 |
+
"epoch": 1.6,
|
| 42 |
+
"grad_norm": 0.8481244444847107,
|
| 43 |
+
"learning_rate": 0.0005,
|
| 44 |
+
"loss": 1.2605,
|
| 45 |
+
"step": 300
|
| 46 |
+
},
|
| 47 |
+
{
|
| 48 |
+
"epoch": 2.0,
|
| 49 |
+
"eval_accuracy": 0.5706666666666667,
|
| 50 |
+
"eval_loss": 1.6909353733062744,
|
| 51 |
+
"eval_runtime": 5.4194,
|
| 52 |
+
"eval_samples_per_second": 92.26,
|
| 53 |
+
"eval_steps_per_second": 11.625,
|
| 54 |
+
"step": 375
|
| 55 |
+
},
|
| 56 |
+
{
|
| 57 |
+
"epoch": 2.0,
|
| 58 |
+
"eval_exact_match": 17.0,
|
| 59 |
+
"eval_f1": 25.9057142857143,
|
| 60 |
+
"step": 375
|
| 61 |
+
},
|
| 62 |
+
{
|
| 63 |
+
"epoch": 2.1333333333333333,
|
| 64 |
+
"grad_norm": 0.810580313205719,
|
| 65 |
+
"learning_rate": 0.0005,
|
| 66 |
+
"loss": 1.1991,
|
| 67 |
+
"step": 400
|
| 68 |
+
},
|
| 69 |
+
{
|
| 70 |
+
"epoch": 2.6666666666666665,
|
| 71 |
+
"grad_norm": 0.8971414566040039,
|
| 72 |
+
"learning_rate": 0.0005,
|
| 73 |
+
"loss": 0.9141,
|
| 74 |
+
"step": 500
|
| 75 |
+
},
|
| 76 |
+
{
|
| 77 |
+
"epoch": 2.997333333333333,
|
| 78 |
+
"eval_accuracy": 0.5657435897435897,
|
| 79 |
+
"eval_loss": 1.831854224205017,
|
| 80 |
+
"eval_runtime": 4.9933,
|
| 81 |
+
"eval_samples_per_second": 100.134,
|
| 82 |
+
"eval_steps_per_second": 12.617,
|
| 83 |
+
"step": 562
|
| 84 |
+
},
|
| 85 |
+
{
|
| 86 |
+
"epoch": 2.997333333333333,
|
| 87 |
+
"eval_exact_match": 16.0,
|
| 88 |
+
"eval_f1": 23.940000000000012,
|
| 89 |
+
"step": 562
|
| 90 |
+
},
|
| 91 |
+
{
|
| 92 |
+
"epoch": 3.2,
|
| 93 |
+
"grad_norm": 0.7627168297767639,
|
| 94 |
+
"learning_rate": 0.0005,
|
| 95 |
+
"loss": 0.8634,
|
| 96 |
+
"step": 600
|
| 97 |
+
},
|
| 98 |
+
{
|
| 99 |
+
"epoch": 3.7333333333333334,
|
| 100 |
+
"grad_norm": 0.7384397387504578,
|
| 101 |
+
"learning_rate": 0.0005,
|
| 102 |
+
"loss": 0.7223,
|
| 103 |
+
"step": 700
|
| 104 |
+
},
|
| 105 |
+
{
|
| 106 |
+
"epoch": 4.0,
|
| 107 |
+
"eval_accuracy": 0.5637948717948718,
|
| 108 |
+
"eval_loss": 1.9418420791625977,
|
| 109 |
+
"eval_runtime": 5.3476,
|
| 110 |
+
"eval_samples_per_second": 93.499,
|
| 111 |
+
"eval_steps_per_second": 11.781,
|
| 112 |
+
"step": 750
|
| 113 |
+
},
|
| 114 |
+
{
|
| 115 |
+
"epoch": 4.0,
|
| 116 |
+
"eval_exact_match": 16.6,
|
| 117 |
+
"eval_f1": 24.14476190476191,
|
| 118 |
+
"step": 750
|
| 119 |
+
},
|
| 120 |
+
{
|
| 121 |
+
"epoch": 4.266666666666667,
|
| 122 |
+
"grad_norm": 0.7065069079399109,
|
| 123 |
+
"learning_rate": 0.0005,
|
| 124 |
+
"loss": 0.6786,
|
| 125 |
+
"step": 800
|
| 126 |
+
},
|
| 127 |
+
{
|
| 128 |
+
"epoch": 4.8,
|
| 129 |
+
"grad_norm": 0.7625908851623535,
|
| 130 |
+
"learning_rate": 0.0005,
|
| 131 |
+
"loss": 0.6434,
|
| 132 |
+
"step": 900
|
| 133 |
+
},
|
| 134 |
+
{
|
| 135 |
+
"epoch": 4.997333333333334,
|
| 136 |
+
"eval_accuracy": 0.5602051282051282,
|
| 137 |
+
"eval_loss": 2.093710422515869,
|
| 138 |
+
"eval_runtime": 5.1796,
|
| 139 |
+
"eval_samples_per_second": 96.532,
|
| 140 |
+
"eval_steps_per_second": 12.163,
|
| 141 |
+
"step": 937
|
| 142 |
+
},
|
| 143 |
+
{
|
| 144 |
+
"epoch": 4.997333333333334,
|
| 145 |
+
"eval_exact_match": 17.2,
|
| 146 |
+
"eval_f1": 24.822857142857156,
|
| 147 |
+
"step": 937
|
| 148 |
+
},
|
| 149 |
+
{
|
| 150 |
+
"epoch": 5.333333333333333,
|
| 151 |
+
"grad_norm": 0.6257635951042175,
|
| 152 |
+
"learning_rate": 0.0005,
|
| 153 |
+
"loss": 0.6022,
|
| 154 |
+
"step": 1000
|
| 155 |
+
},
|
| 156 |
+
{
|
| 157 |
+
"epoch": 5.866666666666667,
|
| 158 |
+
"grad_norm": 0.7330732345581055,
|
| 159 |
+
"learning_rate": 0.0005,
|
| 160 |
+
"loss": 0.5957,
|
| 161 |
+
"step": 1100
|
| 162 |
+
},
|
| 163 |
+
{
|
| 164 |
+
"epoch": 6.0,
|
| 165 |
+
"eval_accuracy": 0.5573333333333333,
|
| 166 |
+
"eval_loss": 2.17874813079834,
|
| 167 |
+
"eval_runtime": 5.3048,
|
| 168 |
+
"eval_samples_per_second": 94.255,
|
| 169 |
+
"eval_steps_per_second": 11.876,
|
| 170 |
+
"step": 1125
|
| 171 |
+
},
|
| 172 |
+
{
|
| 173 |
+
"epoch": 6.0,
|
| 174 |
+
"eval_exact_match": 15.2,
|
| 175 |
+
"eval_f1": 22.341269841269852,
|
| 176 |
+
"step": 1125
|
| 177 |
+
},
|
| 178 |
+
{
|
| 179 |
+
"epoch": 6.4,
|
| 180 |
+
"grad_norm": 0.5348079204559326,
|
| 181 |
+
"learning_rate": 0.0005,
|
| 182 |
+
"loss": 0.5583,
|
| 183 |
+
"step": 1200
|
| 184 |
+
},
|
| 185 |
+
{
|
| 186 |
+
"epoch": 6.933333333333334,
|
| 187 |
+
"grad_norm": 0.4842449426651001,
|
| 188 |
+
"learning_rate": 0.0005,
|
| 189 |
+
"loss": 0.5728,
|
| 190 |
+
"step": 1300
|
| 191 |
+
},
|
| 192 |
+
{
|
| 193 |
+
"epoch": 6.997333333333334,
|
| 194 |
+
"eval_accuracy": 0.558,
|
| 195 |
+
"eval_loss": 2.2602250576019287,
|
| 196 |
+
"eval_runtime": 5.068,
|
| 197 |
+
"eval_samples_per_second": 98.658,
|
| 198 |
+
"eval_steps_per_second": 12.431,
|
| 199 |
+
"step": 1312
|
| 200 |
+
},
|
| 201 |
+
{
|
| 202 |
+
"epoch": 6.997333333333334,
|
| 203 |
+
"eval_exact_match": 15.8,
|
| 204 |
+
"eval_f1": 22.874285714285733,
|
| 205 |
+
"step": 1312
|
| 206 |
+
},
|
| 207 |
+
{
|
| 208 |
+
"epoch": 7.466666666666667,
|
| 209 |
+
"grad_norm": 0.5476555824279785,
|
| 210 |
+
"learning_rate": 0.0005,
|
| 211 |
+
"loss": 0.5339,
|
| 212 |
+
"step": 1400
|
| 213 |
+
},
|
| 214 |
+
{
|
| 215 |
+
"epoch": 8.0,
|
| 216 |
+
"grad_norm": 0.8545846939086914,
|
| 217 |
+
"learning_rate": 0.0005,
|
| 218 |
+
"loss": 0.5596,
|
| 219 |
+
"step": 1500
|
| 220 |
+
},
|
| 221 |
+
{
|
| 222 |
+
"epoch": 8.0,
|
| 223 |
+
"eval_accuracy": 0.5559487179487179,
|
| 224 |
+
"eval_loss": 2.2222671508789062,
|
| 225 |
+
"eval_runtime": 5.0471,
|
| 226 |
+
"eval_samples_per_second": 99.067,
|
| 227 |
+
"eval_steps_per_second": 12.482,
|
| 228 |
+
"step": 1500
|
| 229 |
+
},
|
| 230 |
+
{
|
| 231 |
+
"epoch": 8.0,
|
| 232 |
+
"eval_exact_match": 15.2,
|
| 233 |
+
"eval_f1": 23.320000000000007,
|
| 234 |
+
"step": 1500
|
| 235 |
+
},
|
| 236 |
+
{
|
| 237 |
+
"epoch": 8.533333333333333,
|
| 238 |
+
"grad_norm": 0.5809034705162048,
|
| 239 |
+
"learning_rate": 0.0005,
|
| 240 |
+
"loss": 0.5218,
|
| 241 |
+
"step": 1600
|
| 242 |
+
},
|
| 243 |
+
{
|
| 244 |
+
"epoch": 8.997333333333334,
|
| 245 |
+
"eval_accuracy": 0.5538461538461539,
|
| 246 |
+
"eval_loss": 2.2796590328216553,
|
| 247 |
+
"eval_runtime": 5.0384,
|
| 248 |
+
"eval_samples_per_second": 99.238,
|
| 249 |
+
"eval_steps_per_second": 12.504,
|
| 250 |
+
"step": 1687
|
| 251 |
+
},
|
| 252 |
+
{
|
| 253 |
+
"epoch": 8.997333333333334,
|
| 254 |
+
"eval_exact_match": 13.2,
|
| 255 |
+
"eval_f1": 20.368109668109675,
|
| 256 |
+
"step": 1687
|
| 257 |
+
},
|
| 258 |
+
{
|
| 259 |
+
"epoch": 9.066666666666666,
|
| 260 |
+
"grad_norm": 0.48757097125053406,
|
| 261 |
+
"learning_rate": 0.0005,
|
| 262 |
+
"loss": 0.5444,
|
| 263 |
+
"step": 1700
|
| 264 |
+
},
|
| 265 |
+
{
|
| 266 |
+
"epoch": 9.6,
|
| 267 |
+
"grad_norm": 0.5225056409835815,
|
| 268 |
+
"learning_rate": 0.0005,
|
| 269 |
+
"loss": 0.5254,
|
| 270 |
+
"step": 1800
|
| 271 |
+
},
|
| 272 |
+
{
|
| 273 |
+
"epoch": 10.0,
|
| 274 |
+
"eval_accuracy": 0.5561025641025641,
|
| 275 |
+
"eval_loss": 2.275782823562622,
|
| 276 |
+
"eval_runtime": 5.158,
|
| 277 |
+
"eval_samples_per_second": 96.937,
|
| 278 |
+
"eval_steps_per_second": 12.214,
|
| 279 |
+
"step": 1875
|
| 280 |
+
},
|
| 281 |
+
{
|
| 282 |
+
"epoch": 10.0,
|
| 283 |
+
"eval_exact_match": 14.0,
|
| 284 |
+
"eval_f1": 21.65190476190477,
|
| 285 |
+
"step": 1875
|
| 286 |
+
},
|
| 287 |
+
{
|
| 288 |
+
"epoch": 10.133333333333333,
|
| 289 |
+
"grad_norm": 0.7123950719833374,
|
| 290 |
+
"learning_rate": 0.0005,
|
| 291 |
+
"loss": 0.5304,
|
| 292 |
+
"step": 1900
|
| 293 |
+
},
|
| 294 |
+
{
|
| 295 |
+
"epoch": 10.666666666666666,
|
| 296 |
+
"grad_norm": 0.48232755064964294,
|
| 297 |
+
"learning_rate": 0.0005,
|
| 298 |
+
"loss": 0.526,
|
| 299 |
+
"step": 2000
|
| 300 |
+
},
|
| 301 |
+
{
|
| 302 |
+
"epoch": 10.997333333333334,
|
| 303 |
+
"eval_accuracy": 0.5544102564102564,
|
| 304 |
+
"eval_loss": 2.2686431407928467,
|
| 305 |
+
"eval_runtime": 5.0517,
|
| 306 |
+
"eval_samples_per_second": 98.976,
|
| 307 |
+
"eval_steps_per_second": 12.471,
|
| 308 |
+
"step": 2062
|
| 309 |
+
},
|
| 310 |
+
{
|
| 311 |
+
"epoch": 10.997333333333334,
|
| 312 |
+
"eval_exact_match": 13.0,
|
| 313 |
+
"eval_f1": 20.387460317460324,
|
| 314 |
+
"step": 2062
|
| 315 |
+
},
|
| 316 |
+
{
|
| 317 |
+
"epoch": 11.2,
|
| 318 |
+
"grad_norm": 0.6355456113815308,
|
| 319 |
+
"learning_rate": 0.0005,
|
| 320 |
+
"loss": 0.5335,
|
| 321 |
+
"step": 2100
|
| 322 |
+
},
|
| 323 |
+
{
|
| 324 |
+
"epoch": 11.733333333333333,
|
| 325 |
+
"grad_norm": 0.7310680150985718,
|
| 326 |
+
"learning_rate": 0.0005,
|
| 327 |
+
"loss": 0.5396,
|
| 328 |
+
"step": 2200
|
| 329 |
+
},
|
| 330 |
+
{
|
| 331 |
+
"epoch": 12.0,
|
| 332 |
+
"eval_accuracy": 0.5533846153846154,
|
| 333 |
+
"eval_loss": 2.2792105674743652,
|
| 334 |
+
"eval_runtime": 5.1276,
|
| 335 |
+
"eval_samples_per_second": 97.511,
|
| 336 |
+
"eval_steps_per_second": 12.286,
|
| 337 |
+
"step": 2250
|
| 338 |
+
},
|
| 339 |
+
{
|
| 340 |
+
"epoch": 12.0,
|
| 341 |
+
"eval_exact_match": 13.4,
|
| 342 |
+
"eval_f1": 20.533650793650807,
|
| 343 |
+
"step": 2250
|
| 344 |
+
},
|
| 345 |
+
{
|
| 346 |
+
"epoch": 12.266666666666667,
|
| 347 |
+
"grad_norm": 0.8953878879547119,
|
| 348 |
+
"learning_rate": 0.0005,
|
| 349 |
+
"loss": 0.5383,
|
| 350 |
+
"step": 2300
|
| 351 |
+
},
|
| 352 |
+
{
|
| 353 |
+
"epoch": 12.8,
|
| 354 |
+
"grad_norm": 0.8268033266067505,
|
| 355 |
+
"learning_rate": 0.0005,
|
| 356 |
+
"loss": 0.5466,
|
| 357 |
+
"step": 2400
|
| 358 |
+
},
|
| 359 |
+
{
|
| 360 |
+
"epoch": 12.997333333333334,
|
| 361 |
+
"eval_accuracy": 0.5500512820512821,
|
| 362 |
+
"eval_loss": 2.2915709018707275,
|
| 363 |
+
"eval_runtime": 5.0459,
|
| 364 |
+
"eval_samples_per_second": 99.089,
|
| 365 |
+
"eval_steps_per_second": 12.485,
|
| 366 |
+
"step": 2437
|
| 367 |
+
},
|
| 368 |
+
{
|
| 369 |
+
"epoch": 12.997333333333334,
|
| 370 |
+
"eval_exact_match": 14.0,
|
| 371 |
+
"eval_f1": 21.203650793650798,
|
| 372 |
+
"step": 2437
|
| 373 |
+
},
|
| 374 |
+
{
|
| 375 |
+
"epoch": 13.333333333333334,
|
| 376 |
+
"grad_norm": 0.9471064209938049,
|
| 377 |
+
"learning_rate": 0.0005,
|
| 378 |
+
"loss": 0.5417,
|
| 379 |
+
"step": 2500
|
| 380 |
+
},
|
| 381 |
+
{
|
| 382 |
+
"epoch": 13.866666666666667,
|
| 383 |
+
"grad_norm": 1.0178123712539673,
|
| 384 |
+
"learning_rate": 0.0005,
|
| 385 |
+
"loss": 0.556,
|
| 386 |
+
"step": 2600
|
| 387 |
+
},
|
| 388 |
+
{
|
| 389 |
+
"epoch": 14.0,
|
| 390 |
+
"eval_accuracy": 0.5496923076923077,
|
| 391 |
+
"eval_loss": 2.3209521770477295,
|
| 392 |
+
"eval_runtime": 5.4611,
|
| 393 |
+
"eval_samples_per_second": 91.557,
|
| 394 |
+
"eval_steps_per_second": 11.536,
|
| 395 |
+
"step": 2625
|
| 396 |
+
},
|
| 397 |
+
{
|
| 398 |
+
"epoch": 14.0,
|
| 399 |
+
"eval_exact_match": 12.6,
|
| 400 |
+
"eval_f1": 20.174790764790767,
|
| 401 |
+
"step": 2625
|
| 402 |
+
},
|
| 403 |
+
{
|
| 404 |
+
"epoch": 14.4,
|
| 405 |
+
"grad_norm": 0.750542402267456,
|
| 406 |
+
"learning_rate": 0.0005,
|
| 407 |
+
"loss": 0.5322,
|
| 408 |
+
"step": 2700
|
| 409 |
+
},
|
| 410 |
+
{
|
| 411 |
+
"epoch": 14.933333333333334,
|
| 412 |
+
"grad_norm": 0.8345568776130676,
|
| 413 |
+
"learning_rate": 0.0005,
|
| 414 |
+
"loss": 0.5501,
|
| 415 |
+
"step": 2800
|
| 416 |
+
},
|
| 417 |
+
{
|
| 418 |
+
"epoch": 14.997333333333334,
|
| 419 |
+
"eval_accuracy": 0.5464615384615384,
|
| 420 |
+
"eval_loss": 2.3918418884277344,
|
| 421 |
+
"eval_runtime": 5.1243,
|
| 422 |
+
"eval_samples_per_second": 97.574,
|
| 423 |
+
"eval_steps_per_second": 12.294,
|
| 424 |
+
"step": 2812
|
| 425 |
+
},
|
| 426 |
+
{
|
| 427 |
+
"epoch": 14.997333333333334,
|
| 428 |
+
"eval_exact_match": 12.4,
|
| 429 |
+
"eval_f1": 19.851428571428578,
|
| 430 |
+
"step": 2812
|
| 431 |
+
},
|
| 432 |
+
{
|
| 433 |
+
"epoch": 15.466666666666667,
|
| 434 |
+
"grad_norm": 0.60535728931427,
|
| 435 |
+
"learning_rate": 0.0005,
|
| 436 |
+
"loss": 0.5174,
|
| 437 |
+
"step": 2900
|
| 438 |
+
},
|
| 439 |
+
{
|
| 440 |
+
"epoch": 16.0,
|
| 441 |
+
"grad_norm": 0.7003186941146851,
|
| 442 |
+
"learning_rate": 0.0005,
|
| 443 |
+
"loss": 0.5461,
|
| 444 |
+
"step": 3000
|
| 445 |
+
},
|
| 446 |
+
{
|
| 447 |
+
"epoch": 16.0,
|
| 448 |
+
"eval_accuracy": 0.5475897435897435,
|
| 449 |
+
"eval_loss": 2.3702445030212402,
|
| 450 |
+
"eval_runtime": 5.6883,
|
| 451 |
+
"eval_samples_per_second": 87.9,
|
| 452 |
+
"eval_steps_per_second": 11.075,
|
| 453 |
+
"step": 3000
|
| 454 |
+
},
|
| 455 |
+
{
|
| 456 |
+
"epoch": 16.0,
|
| 457 |
+
"eval_exact_match": 12.6,
|
| 458 |
+
"eval_f1": 19.352380952380955,
|
| 459 |
+
"step": 3000
|
| 460 |
+
},
|
| 461 |
+
{
|
| 462 |
+
"epoch": 16.533333333333335,
|
| 463 |
+
"grad_norm": 0.9538315534591675,
|
| 464 |
+
"learning_rate": 0.0005,
|
| 465 |
+
"loss": 0.5032,
|
| 466 |
+
"step": 3100
|
| 467 |
+
},
|
| 468 |
+
{
|
| 469 |
+
"epoch": 16.997333333333334,
|
| 470 |
+
"eval_accuracy": 0.5461025641025641,
|
| 471 |
+
"eval_loss": 2.443984270095825,
|
| 472 |
+
"eval_runtime": 5.1587,
|
| 473 |
+
"eval_samples_per_second": 96.923,
|
| 474 |
+
"eval_steps_per_second": 12.212,
|
| 475 |
+
"step": 3187
|
| 476 |
+
},
|
| 477 |
+
{
|
| 478 |
+
"epoch": 16.997333333333334,
|
| 479 |
+
"eval_exact_match": 11.6,
|
| 480 |
+
"eval_f1": 19.255238095238102,
|
| 481 |
+
"step": 3187
|
| 482 |
+
},
|
| 483 |
+
{
|
| 484 |
+
"epoch": 17.066666666666666,
|
| 485 |
+
"grad_norm": 0.7299863696098328,
|
| 486 |
+
"learning_rate": 0.0005,
|
| 487 |
+
"loss": 0.5267,
|
| 488 |
+
"step": 3200
|
| 489 |
+
},
|
| 490 |
+
{
|
| 491 |
+
"epoch": 17.6,
|
| 492 |
+
"grad_norm": 0.49411433935165405,
|
| 493 |
+
"learning_rate": 0.0005,
|
| 494 |
+
"loss": 0.501,
|
| 495 |
+
"step": 3300
|
| 496 |
+
},
|
| 497 |
+
{
|
| 498 |
+
"epoch": 18.0,
|
| 499 |
+
"eval_accuracy": 0.5468205128205128,
|
| 500 |
+
"eval_loss": 2.4240849018096924,
|
| 501 |
+
"eval_runtime": 5.0346,
|
| 502 |
+
"eval_samples_per_second": 99.313,
|
| 503 |
+
"eval_steps_per_second": 12.513,
|
| 504 |
+
"step": 3375
|
| 505 |
+
},
|
| 506 |
+
{
|
| 507 |
+
"epoch": 18.0,
|
| 508 |
+
"eval_exact_match": 12.8,
|
| 509 |
+
"eval_f1": 19.46714285714286,
|
| 510 |
+
"step": 3375
|
| 511 |
+
},
|
| 512 |
+
{
|
| 513 |
+
"epoch": 18.133333333333333,
|
| 514 |
+
"grad_norm": 0.5169008374214172,
|
| 515 |
+
"learning_rate": 0.0005,
|
| 516 |
+
"loss": 0.5077,
|
| 517 |
+
"step": 3400
|
| 518 |
+
},
|
| 519 |
+
{
|
| 520 |
+
"epoch": 18.666666666666668,
|
| 521 |
+
"grad_norm": 0.4064069092273712,
|
| 522 |
+
"learning_rate": 0.0005,
|
| 523 |
+
"loss": 0.4975,
|
| 524 |
+
"step": 3500
|
| 525 |
+
},
|
| 526 |
+
{
|
| 527 |
+
"epoch": 18.997333333333334,
|
| 528 |
+
"eval_accuracy": 0.5450769230769231,
|
| 529 |
+
"eval_loss": 2.4591097831726074,
|
| 530 |
+
"eval_runtime": 5.7434,
|
| 531 |
+
"eval_samples_per_second": 87.056,
|
| 532 |
+
"eval_steps_per_second": 10.969,
|
| 533 |
+
"step": 3562
|
| 534 |
+
},
|
| 535 |
+
{
|
| 536 |
+
"epoch": 18.997333333333334,
|
| 537 |
+
"eval_exact_match": 12.8,
|
| 538 |
+
"eval_f1": 19.07666666666667,
|
| 539 |
+
"step": 3562
|
| 540 |
+
},
|
| 541 |
+
{
|
| 542 |
+
"epoch": 19.2,
|
| 543 |
+
"grad_norm": 0.362119197845459,
|
| 544 |
+
"learning_rate": 0.0005,
|
| 545 |
+
"loss": 0.5028,
|
| 546 |
+
"step": 3600
|
| 547 |
+
},
|
| 548 |
+
{
|
| 549 |
+
"epoch": 19.733333333333334,
|
| 550 |
+
"grad_norm": 0.45963579416275024,
|
| 551 |
+
"learning_rate": 0.0005,
|
| 552 |
+
"loss": 0.4977,
|
| 553 |
+
"step": 3700
|
| 554 |
+
},
|
| 555 |
+
{
|
| 556 |
+
"epoch": 20.0,
|
| 557 |
+
"eval_accuracy": 0.545025641025641,
|
| 558 |
+
"eval_loss": 2.4378092288970947,
|
| 559 |
+
"eval_runtime": 5.0227,
|
| 560 |
+
"eval_samples_per_second": 99.548,
|
| 561 |
+
"eval_steps_per_second": 12.543,
|
| 562 |
+
"step": 3750
|
| 563 |
+
},
|
| 564 |
+
{
|
| 565 |
+
"epoch": 20.0,
|
| 566 |
+
"eval_exact_match": 12.6,
|
| 567 |
+
"eval_f1": 19.655873015873016,
|
| 568 |
+
"step": 3750
|
| 569 |
+
},
|
| 570 |
+
{
|
| 571 |
+
"epoch": 20.266666666666666,
|
| 572 |
+
"grad_norm": 0.6107614636421204,
|
| 573 |
+
"learning_rate": 0.0005,
|
| 574 |
+
"loss": 0.4964,
|
| 575 |
+
"step": 3800
|
| 576 |
+
},
|
| 577 |
+
{
|
| 578 |
+
"epoch": 20.8,
|
| 579 |
+
"grad_norm": 0.5394221544265747,
|
| 580 |
+
"learning_rate": 0.0005,
|
| 581 |
+
"loss": 0.501,
|
| 582 |
+
"step": 3900
|
| 583 |
+
},
|
| 584 |
+
{
|
| 585 |
+
"epoch": 20.997333333333334,
|
| 586 |
+
"eval_accuracy": 0.5438461538461539,
|
| 587 |
+
"eval_loss": 2.448828935623169,
|
| 588 |
+
"eval_runtime": 5.1176,
|
| 589 |
+
"eval_samples_per_second": 97.702,
|
| 590 |
+
"eval_steps_per_second": 12.31,
|
| 591 |
+
"step": 3937
|
| 592 |
+
},
|
| 593 |
+
{
|
| 594 |
+
"epoch": 20.997333333333334,
|
| 595 |
+
"eval_exact_match": 12.6,
|
| 596 |
+
"eval_f1": 18.367460317460317,
|
| 597 |
+
"step": 3937
|
| 598 |
+
},
|
| 599 |
+
{
|
| 600 |
+
"epoch": 21.333333333333332,
|
| 601 |
+
"grad_norm": 0.35812634229660034,
|
| 602 |
+
"learning_rate": 0.0005,
|
| 603 |
+
"loss": 0.4964,
|
| 604 |
+
"step": 4000
|
| 605 |
+
},
|
| 606 |
+
{
|
| 607 |
+
"epoch": 21.866666666666667,
|
| 608 |
+
"grad_norm": 0.4076957404613495,
|
| 609 |
+
"learning_rate": 0.0005,
|
| 610 |
+
"loss": 0.5133,
|
| 611 |
+
"step": 4100
|
| 612 |
+
},
|
| 613 |
+
{
|
| 614 |
+
"epoch": 22.0,
|
| 615 |
+
"eval_accuracy": 0.5428717948717948,
|
| 616 |
+
"eval_loss": 2.548297882080078,
|
| 617 |
+
"eval_runtime": 6.0053,
|
| 618 |
+
"eval_samples_per_second": 83.26,
|
| 619 |
+
"eval_steps_per_second": 10.491,
|
| 620 |
+
"step": 4125
|
| 621 |
+
},
|
| 622 |
+
{
|
| 623 |
+
"epoch": 22.0,
|
| 624 |
+
"eval_exact_match": 11.6,
|
| 625 |
+
"eval_f1": 17.79587301587302,
|
| 626 |
+
"step": 4125
|
| 627 |
+
},
|
| 628 |
+
{
|
| 629 |
+
"epoch": 22.4,
|
| 630 |
+
"grad_norm": 0.5148049592971802,
|
| 631 |
+
"learning_rate": 0.0005,
|
| 632 |
+
"loss": 0.4957,
|
| 633 |
+
"step": 4200
|
| 634 |
+
},
|
| 635 |
+
{
|
| 636 |
+
"epoch": 22.933333333333334,
|
| 637 |
+
"grad_norm": 0.7316868305206299,
|
| 638 |
+
"learning_rate": 0.0005,
|
| 639 |
+
"loss": 0.5154,
|
| 640 |
+
"step": 4300
|
| 641 |
+
},
|
| 642 |
+
{
|
| 643 |
+
"epoch": 22.997333333333334,
|
| 644 |
+
"eval_accuracy": 0.5442051282051282,
|
| 645 |
+
"eval_loss": 2.460996389389038,
|
| 646 |
+
"eval_runtime": 5.0409,
|
| 647 |
+
"eval_samples_per_second": 99.189,
|
| 648 |
+
"eval_steps_per_second": 12.498,
|
| 649 |
+
"step": 4312
|
| 650 |
+
},
|
| 651 |
+
{
|
| 652 |
+
"epoch": 22.997333333333334,
|
| 653 |
+
"eval_exact_match": 13.4,
|
| 654 |
+
"eval_f1": 19.424285714285713,
|
| 655 |
+
"step": 4312
|
| 656 |
+
},
|
| 657 |
+
{
|
| 658 |
+
"epoch": 23.466666666666665,
|
| 659 |
+
"grad_norm": 0.7490749955177307,
|
| 660 |
+
"learning_rate": 0.0005,
|
| 661 |
+
"loss": 0.5006,
|
| 662 |
+
"step": 4400
|
| 663 |
+
},
|
| 664 |
+
{
|
| 665 |
+
"epoch": 24.0,
|
| 666 |
+
"grad_norm": 0.7351626753807068,
|
| 667 |
+
"learning_rate": 0.0005,
|
| 668 |
+
"loss": 0.5257,
|
| 669 |
+
"step": 4500
|
| 670 |
+
},
|
| 671 |
+
{
|
| 672 |
+
"epoch": 24.0,
|
| 673 |
+
"eval_accuracy": 0.5448717948717948,
|
| 674 |
+
"eval_loss": 2.503119945526123,
|
| 675 |
+
"eval_runtime": 4.9846,
|
| 676 |
+
"eval_samples_per_second": 100.308,
|
| 677 |
+
"eval_steps_per_second": 12.639,
|
| 678 |
+
"step": 4500
|
| 679 |
+
},
|
| 680 |
+
{
|
| 681 |
+
"epoch": 24.0,
|
| 682 |
+
"eval_exact_match": 12.2,
|
| 683 |
+
"eval_f1": 18.472222222222225,
|
| 684 |
+
"step": 4500
|
| 685 |
+
},
|
| 686 |
+
{
|
| 687 |
+
"epoch": 24.533333333333335,
|
| 688 |
+
"grad_norm": 0.5879587531089783,
|
| 689 |
+
"learning_rate": 0.0005,
|
| 690 |
+
"loss": 0.5065,
|
| 691 |
+
"step": 4600
|
| 692 |
+
},
|
| 693 |
+
{
|
| 694 |
+
"epoch": 24.997333333333334,
|
| 695 |
+
"eval_accuracy": 0.5415384615384615,
|
| 696 |
+
"eval_loss": 2.510695695877075,
|
| 697 |
+
"eval_runtime": 5.7813,
|
| 698 |
+
"eval_samples_per_second": 86.485,
|
| 699 |
+
"eval_steps_per_second": 10.897,
|
| 700 |
+
"step": 4687
|
| 701 |
+
},
|
| 702 |
+
{
|
| 703 |
+
"epoch": 24.997333333333334,
|
| 704 |
+
"eval_exact_match": 13.2,
|
| 705 |
+
"eval_f1": 19.150952380952386,
|
| 706 |
+
"step": 4687
|
| 707 |
+
},
|
| 708 |
+
{
|
| 709 |
+
"epoch": 25.066666666666666,
|
| 710 |
+
"grad_norm": 0.61104816198349,
|
| 711 |
+
"learning_rate": 0.0005,
|
| 712 |
+
"loss": 0.5273,
|
| 713 |
+
"step": 4700
|
| 714 |
+
},
|
| 715 |
+
{
|
| 716 |
+
"epoch": 25.6,
|
| 717 |
+
"grad_norm": 1.318751335144043,
|
| 718 |
+
"learning_rate": 0.0005,
|
| 719 |
+
"loss": 0.5177,
|
| 720 |
+
"step": 4800
|
| 721 |
+
},
|
| 722 |
+
{
|
| 723 |
+
"epoch": 26.0,
|
| 724 |
+
"eval_accuracy": 0.5402051282051282,
|
| 725 |
+
"eval_loss": 2.5010507106781006,
|
| 726 |
+
"eval_runtime": 5.3385,
|
| 727 |
+
"eval_samples_per_second": 93.66,
|
| 728 |
+
"eval_steps_per_second": 11.801,
|
| 729 |
+
"step": 4875
|
| 730 |
+
},
|
| 731 |
+
{
|
| 732 |
+
"epoch": 26.0,
|
| 733 |
+
"eval_exact_match": 12.0,
|
| 734 |
+
"eval_f1": 18.683665223665226,
|
| 735 |
+
"step": 4875
|
| 736 |
+
},
|
| 737 |
+
{
|
| 738 |
+
"epoch": 26.133333333333333,
|
| 739 |
+
"grad_norm": 0.8964791893959045,
|
| 740 |
+
"learning_rate": 0.0005,
|
| 741 |
+
"loss": 0.5267,
|
| 742 |
+
"step": 4900
|
| 743 |
+
},
|
| 744 |
+
{
|
| 745 |
+
"epoch": 26.666666666666668,
|
| 746 |
+
"grad_norm": 0.6350087523460388,
|
| 747 |
+
"learning_rate": 0.0005,
|
| 748 |
+
"loss": 0.5115,
|
| 749 |
+
"step": 5000
|
| 750 |
+
},
|
| 751 |
+
{
|
| 752 |
+
"epoch": 26.997333333333334,
|
| 753 |
+
"eval_accuracy": 0.5407179487179488,
|
| 754 |
+
"eval_loss": 2.547609567642212,
|
| 755 |
+
"eval_runtime": 5.1686,
|
| 756 |
+
"eval_samples_per_second": 96.738,
|
| 757 |
+
"eval_steps_per_second": 12.189,
|
| 758 |
+
"step": 5062
|
| 759 |
+
},
|
| 760 |
+
{
|
| 761 |
+
"epoch": 26.997333333333334,
|
| 762 |
+
"eval_exact_match": 12.4,
|
| 763 |
+
"eval_f1": 18.38857142857143,
|
| 764 |
+
"step": 5062
|
| 765 |
+
},
|
| 766 |
+
{
|
| 767 |
+
"epoch": 27.2,
|
| 768 |
+
"grad_norm": 0.47840726375579834,
|
| 769 |
+
"learning_rate": 0.0005,
|
| 770 |
+
"loss": 0.5146,
|
| 771 |
+
"step": 5100
|
| 772 |
+
},
|
| 773 |
+
{
|
| 774 |
+
"epoch": 27.733333333333334,
|
| 775 |
+
"grad_norm": 0.474448561668396,
|
| 776 |
+
"learning_rate": 0.0005,
|
| 777 |
+
"loss": 0.5041,
|
| 778 |
+
"step": 5200
|
| 779 |
+
},
|
| 780 |
+
{
|
| 781 |
+
"epoch": 28.0,
|
| 782 |
+
"eval_accuracy": 0.5392820512820513,
|
| 783 |
+
"eval_loss": 2.578618049621582,
|
| 784 |
+
"eval_runtime": 5.4216,
|
| 785 |
+
"eval_samples_per_second": 92.224,
|
| 786 |
+
"eval_steps_per_second": 11.62,
|
| 787 |
+
"step": 5250
|
| 788 |
+
},
|
| 789 |
+
{
|
| 790 |
+
"epoch": 28.0,
|
| 791 |
+
"eval_exact_match": 12.8,
|
| 792 |
+
"eval_f1": 18.753809523809526,
|
| 793 |
+
"step": 5250
|
| 794 |
+
},
|
| 795 |
+
{
|
| 796 |
+
"epoch": 28.266666666666666,
|
| 797 |
+
"grad_norm": 0.4811856150627136,
|
| 798 |
+
"learning_rate": 0.0005,
|
| 799 |
+
"loss": 0.4985,
|
| 800 |
+
"step": 5300
|
| 801 |
+
},
|
| 802 |
+
{
|
| 803 |
+
"epoch": 28.8,
|
| 804 |
+
"grad_norm": 0.7236740589141846,
|
| 805 |
+
"learning_rate": 0.0005,
|
| 806 |
+
"loss": 0.5014,
|
| 807 |
+
"step": 5400
|
| 808 |
+
},
|
| 809 |
+
{
|
| 810 |
+
"epoch": 28.997333333333334,
|
| 811 |
+
"eval_accuracy": 0.5383589743589744,
|
| 812 |
+
"eval_loss": 2.687852621078491,
|
| 813 |
+
"eval_runtime": 5.147,
|
| 814 |
+
"eval_samples_per_second": 97.143,
|
| 815 |
+
"eval_steps_per_second": 12.24,
|
| 816 |
+
"step": 5437
|
| 817 |
+
},
|
| 818 |
+
{
|
| 819 |
+
"epoch": 28.997333333333334,
|
| 820 |
+
"eval_exact_match": 12.6,
|
| 821 |
+
"eval_f1": 18.35571428571428,
|
| 822 |
+
"step": 5437
|
| 823 |
+
},
|
| 824 |
+
{
|
| 825 |
+
"epoch": 29.333333333333332,
|
| 826 |
+
"grad_norm": 0.6178373098373413,
|
| 827 |
+
"learning_rate": 0.0005,
|
| 828 |
+
"loss": 0.4912,
|
| 829 |
+
"step": 5500
|
| 830 |
+
},
|
| 831 |
+
{
|
| 832 |
+
"epoch": 29.866666666666667,
|
| 833 |
+
"grad_norm": 0.6287767887115479,
|
| 834 |
+
"learning_rate": 0.0005,
|
| 835 |
+
"loss": 0.5005,
|
| 836 |
+
"step": 5600
|
| 837 |
+
},
|
| 838 |
+
{
|
| 839 |
+
"epoch": 30.0,
|
| 840 |
+
"eval_accuracy": 0.5392307692307692,
|
| 841 |
+
"eval_loss": 2.5662922859191895,
|
| 842 |
+
"eval_runtime": 5.4853,
|
| 843 |
+
"eval_samples_per_second": 91.153,
|
| 844 |
+
"eval_steps_per_second": 11.485,
|
| 845 |
+
"step": 5625
|
| 846 |
+
},
|
| 847 |
+
{
|
| 848 |
+
"epoch": 30.0,
|
| 849 |
+
"eval_exact_match": 11.4,
|
| 850 |
+
"eval_f1": 17.72714285714286,
|
| 851 |
+
"step": 5625
|
| 852 |
+
},
|
| 853 |
+
{
|
| 854 |
+
"epoch": 30.4,
|
| 855 |
+
"grad_norm": 0.45170456171035767,
|
| 856 |
+
"learning_rate": 0.0005,
|
| 857 |
+
"loss": 0.4899,
|
| 858 |
+
"step": 5700
|
| 859 |
+
},
|
| 860 |
+
{
|
| 861 |
+
"epoch": 30.933333333333334,
|
| 862 |
+
"grad_norm": 0.36589592695236206,
|
| 863 |
+
"learning_rate": 0.0005,
|
| 864 |
+
"loss": 0.5015,
|
| 865 |
+
"step": 5800
|
| 866 |
+
},
|
| 867 |
+
{
|
| 868 |
+
"epoch": 30.997333333333334,
|
| 869 |
+
"eval_accuracy": 0.5392307692307692,
|
| 870 |
+
"eval_loss": 2.6379120349884033,
|
| 871 |
+
"eval_runtime": 5.3104,
|
| 872 |
+
"eval_samples_per_second": 94.155,
|
| 873 |
+
"eval_steps_per_second": 11.864,
|
| 874 |
+
"step": 5812
|
| 875 |
+
},
|
| 876 |
+
{
|
| 877 |
+
"epoch": 30.997333333333334,
|
| 878 |
+
"eval_exact_match": 12.4,
|
| 879 |
+
"eval_f1": 18.726190476190478,
|
| 880 |
+
"step": 5812
|
| 881 |
+
},
|
| 882 |
+
{
|
| 883 |
+
"epoch": 31.466666666666665,
|
| 884 |
+
"grad_norm": 0.4986824095249176,
|
| 885 |
+
"learning_rate": 0.0005,
|
| 886 |
+
"loss": 0.4861,
|
| 887 |
+
"step": 5900
|
| 888 |
+
},
|
| 889 |
+
{
|
| 890 |
+
"epoch": 32.0,
|
| 891 |
+
"grad_norm": 0.5775201320648193,
|
| 892 |
+
"learning_rate": 0.0005,
|
| 893 |
+
"loss": 0.5061,
|
| 894 |
+
"step": 6000
|
| 895 |
+
},
|
| 896 |
+
{
|
| 897 |
+
"epoch": 32.0,
|
| 898 |
+
"eval_accuracy": 0.5353846153846153,
|
| 899 |
+
"eval_loss": 2.716935396194458,
|
| 900 |
+
"eval_runtime": 5.0073,
|
| 901 |
+
"eval_samples_per_second": 99.854,
|
| 902 |
+
"eval_steps_per_second": 12.582,
|
| 903 |
+
"step": 6000
|
| 904 |
+
},
|
| 905 |
+
{
|
| 906 |
+
"epoch": 32.0,
|
| 907 |
+
"eval_exact_match": 11.0,
|
| 908 |
+
"eval_f1": 16.892380952380954,
|
| 909 |
+
"step": 6000
|
| 910 |
+
},
|
| 911 |
+
{
|
| 912 |
+
"epoch": 32.53333333333333,
|
| 913 |
+
"grad_norm": 0.8057941794395447,
|
| 914 |
+
"learning_rate": 0.0005,
|
| 915 |
+
"loss": 0.4849,
|
| 916 |
+
"step": 6100
|
| 917 |
+
},
|
| 918 |
+
{
|
| 919 |
+
"epoch": 32.99733333333333,
|
| 920 |
+
"eval_accuracy": 0.5364615384615384,
|
| 921 |
+
"eval_loss": 2.627026081085205,
|
| 922 |
+
"eval_runtime": 5.057,
|
| 923 |
+
"eval_samples_per_second": 98.873,
|
| 924 |
+
"eval_steps_per_second": 12.458,
|
| 925 |
+
"step": 6187
|
| 926 |
+
},
|
| 927 |
+
{
|
| 928 |
+
"epoch": 32.99733333333333,
|
| 929 |
+
"eval_exact_match": 12.8,
|
| 930 |
+
"eval_f1": 18.03952380952381,
|
| 931 |
+
"step": 6187
|
| 932 |
+
},
|
| 933 |
+
{
|
| 934 |
+
"epoch": 33.06666666666667,
|
| 935 |
+
"grad_norm": 0.5468693375587463,
|
| 936 |
+
"learning_rate": 0.0005,
|
| 937 |
+
"loss": 0.5052,
|
| 938 |
+
"step": 6200
|
| 939 |
+
},
|
| 940 |
+
{
|
| 941 |
+
"epoch": 33.6,
|
| 942 |
+
"grad_norm": 0.4891034960746765,
|
| 943 |
+
"learning_rate": 0.0005,
|
| 944 |
+
"loss": 0.4969,
|
| 945 |
+
"step": 6300
|
| 946 |
+
},
|
| 947 |
+
{
|
| 948 |
+
"epoch": 34.0,
|
| 949 |
+
"eval_accuracy": 0.5351794871794872,
|
| 950 |
+
"eval_loss": 2.698223352432251,
|
| 951 |
+
"eval_runtime": 5.2797,
|
| 952 |
+
"eval_samples_per_second": 94.702,
|
| 953 |
+
"eval_steps_per_second": 11.932,
|
| 954 |
+
"step": 6375
|
| 955 |
+
},
|
| 956 |
+
{
|
| 957 |
+
"epoch": 34.0,
|
| 958 |
+
"eval_exact_match": 11.4,
|
| 959 |
+
"eval_f1": 17.50714285714286,
|
| 960 |
+
"step": 6375
|
| 961 |
+
},
|
| 962 |
+
{
|
| 963 |
+
"epoch": 34.13333333333333,
|
| 964 |
+
"grad_norm": 0.9730343818664551,
|
| 965 |
+
"learning_rate": 0.0005,
|
| 966 |
+
"loss": 0.507,
|
| 967 |
+
"step": 6400
|
| 968 |
+
},
|
| 969 |
+
{
|
| 970 |
+
"epoch": 34.666666666666664,
|
| 971 |
+
"grad_norm": 0.8066098093986511,
|
| 972 |
+
"learning_rate": 0.0005,
|
| 973 |
+
"loss": 0.5034,
|
| 974 |
+
"step": 6500
|
| 975 |
+
},
|
| 976 |
+
{
|
| 977 |
+
"epoch": 34.99733333333333,
|
| 978 |
+
"eval_accuracy": 0.5345128205128206,
|
| 979 |
+
"eval_loss": 2.6721694469451904,
|
| 980 |
+
"eval_runtime": 5.1095,
|
| 981 |
+
"eval_samples_per_second": 97.857,
|
| 982 |
+
"eval_steps_per_second": 12.33,
|
| 983 |
+
"step": 6562
|
| 984 |
+
},
|
| 985 |
+
{
|
| 986 |
+
"epoch": 34.99733333333333,
|
| 987 |
+
"eval_exact_match": 11.4,
|
| 988 |
+
"eval_f1": 16.180158730158734,
|
| 989 |
+
"step": 6562
|
| 990 |
+
},
|
| 991 |
+
{
|
| 992 |
+
"epoch": 35.2,
|
| 993 |
+
"grad_norm": 1.1951669454574585,
|
| 994 |
+
"learning_rate": 0.0005,
|
| 995 |
+
"loss": 0.5083,
|
| 996 |
+
"step": 6600
|
| 997 |
+
},
|
| 998 |
+
{
|
| 999 |
+
"epoch": 35.733333333333334,
|
| 1000 |
+
"grad_norm": 0.7211179137229919,
|
| 1001 |
+
"learning_rate": 0.0005,
|
| 1002 |
+
"loss": 0.5052,
|
| 1003 |
+
"step": 6700
|
| 1004 |
+
},
|
| 1005 |
+
{
|
| 1006 |
+
"epoch": 36.0,
|
| 1007 |
+
"eval_accuracy": 0.5344615384615384,
|
| 1008 |
+
"eval_loss": 2.68320894241333,
|
| 1009 |
+
"eval_runtime": 5.0034,
|
| 1010 |
+
"eval_samples_per_second": 99.931,
|
| 1011 |
+
"eval_steps_per_second": 12.591,
|
| 1012 |
+
"step": 6750
|
| 1013 |
+
},
|
| 1014 |
+
{
|
| 1015 |
+
"epoch": 36.0,
|
| 1016 |
+
"eval_exact_match": 10.8,
|
| 1017 |
+
"eval_f1": 16.037619047619053,
|
| 1018 |
+
"step": 6750
|
| 1019 |
+
},
|
| 1020 |
+
{
|
| 1021 |
+
"epoch": 36.266666666666666,
|
| 1022 |
+
"grad_norm": 0.4955157935619354,
|
| 1023 |
+
"learning_rate": 0.0005,
|
| 1024 |
+
"loss": 0.505,
|
| 1025 |
+
"step": 6800
|
| 1026 |
+
},
|
| 1027 |
+
{
|
| 1028 |
+
"epoch": 36.8,
|
| 1029 |
+
"grad_norm": 0.6588331460952759,
|
| 1030 |
+
"learning_rate": 0.0005,
|
| 1031 |
+
"loss": 0.5123,
|
| 1032 |
+
"step": 6900
|
| 1033 |
+
},
|
| 1034 |
+
{
|
| 1035 |
+
"epoch": 36.99733333333333,
|
| 1036 |
+
"eval_accuracy": 0.5363589743589744,
|
| 1037 |
+
"eval_loss": 2.6866376399993896,
|
| 1038 |
+
"eval_runtime": 5.2422,
|
| 1039 |
+
"eval_samples_per_second": 95.379,
|
| 1040 |
+
"eval_steps_per_second": 12.018,
|
| 1041 |
+
"step": 6937
|
| 1042 |
+
},
|
| 1043 |
+
{
|
| 1044 |
+
"epoch": 36.99733333333333,
|
| 1045 |
+
"eval_exact_match": 11.6,
|
| 1046 |
+
"eval_f1": 17.830476190476197,
|
| 1047 |
+
"step": 6937
|
| 1048 |
+
},
|
| 1049 |
+
{
|
| 1050 |
+
"epoch": 37.333333333333336,
|
| 1051 |
+
"grad_norm": 0.5928548574447632,
|
| 1052 |
+
"learning_rate": 0.0005,
|
| 1053 |
+
"loss": 0.4974,
|
| 1054 |
+
"step": 7000
|
| 1055 |
+
},
|
| 1056 |
+
{
|
| 1057 |
+
"epoch": 37.86666666666667,
|
| 1058 |
+
"grad_norm": 0.5979794263839722,
|
| 1059 |
+
"learning_rate": 0.0005,
|
| 1060 |
+
"loss": 0.5061,
|
| 1061 |
+
"step": 7100
|
| 1062 |
+
},
|
| 1063 |
+
{
|
| 1064 |
+
"epoch": 38.0,
|
| 1065 |
+
"eval_accuracy": 0.5329230769230769,
|
| 1066 |
+
"eval_loss": 2.657357931137085,
|
| 1067 |
+
"eval_runtime": 5.238,
|
| 1068 |
+
"eval_samples_per_second": 95.457,
|
| 1069 |
+
"eval_steps_per_second": 12.028,
|
| 1070 |
+
"step": 7125
|
| 1071 |
+
},
|
| 1072 |
+
{
|
| 1073 |
+
"epoch": 38.0,
|
| 1074 |
+
"eval_exact_match": 10.8,
|
| 1075 |
+
"eval_f1": 16.75142857142858,
|
| 1076 |
+
"step": 7125
|
| 1077 |
+
},
|
| 1078 |
+
{
|
| 1079 |
+
"epoch": 38.4,
|
| 1080 |
+
"grad_norm": 0.6051629185676575,
|
| 1081 |
+
"learning_rate": 0.0005,
|
| 1082 |
+
"loss": 0.4896,
|
| 1083 |
+
"step": 7200
|
| 1084 |
+
},
|
| 1085 |
+
{
|
| 1086 |
+
"epoch": 38.93333333333333,
|
| 1087 |
+
"grad_norm": 0.7862063646316528,
|
| 1088 |
+
"learning_rate": 0.0005,
|
| 1089 |
+
"loss": 0.5032,
|
| 1090 |
+
"step": 7300
|
| 1091 |
+
},
|
| 1092 |
+
{
|
| 1093 |
+
"epoch": 38.99733333333333,
|
| 1094 |
+
"eval_accuracy": 0.5366666666666666,
|
| 1095 |
+
"eval_loss": 2.6846814155578613,
|
| 1096 |
+
"eval_runtime": 5.3323,
|
| 1097 |
+
"eval_samples_per_second": 93.768,
|
| 1098 |
+
"eval_steps_per_second": 11.815,
|
| 1099 |
+
"step": 7312
|
| 1100 |
+
},
|
| 1101 |
+
{
|
| 1102 |
+
"epoch": 38.99733333333333,
|
| 1103 |
+
"eval_exact_match": 10.0,
|
| 1104 |
+
"eval_f1": 15.647301587301591,
|
| 1105 |
+
"step": 7312
|
| 1106 |
+
},
|
| 1107 |
+
{
|
| 1108 |
+
"epoch": 39.46666666666667,
|
| 1109 |
+
"grad_norm": 0.5463705658912659,
|
| 1110 |
+
"learning_rate": 0.0005,
|
| 1111 |
+
"loss": 0.4797,
|
| 1112 |
+
"step": 7400
|
| 1113 |
+
},
|
| 1114 |
+
{
|
| 1115 |
+
"epoch": 40.0,
|
| 1116 |
+
"grad_norm": 0.6618899703025818,
|
| 1117 |
+
"learning_rate": 0.0005,
|
| 1118 |
+
"loss": 0.5051,
|
| 1119 |
+
"step": 7500
|
| 1120 |
+
},
|
| 1121 |
+
{
|
| 1122 |
+
"epoch": 40.0,
|
| 1123 |
+
"eval_accuracy": 0.536,
|
| 1124 |
+
"eval_loss": 2.660195827484131,
|
| 1125 |
+
"eval_runtime": 5.0572,
|
| 1126 |
+
"eval_samples_per_second": 98.869,
|
| 1127 |
+
"eval_steps_per_second": 12.457,
|
| 1128 |
+
"step": 7500
|
| 1129 |
+
},
|
| 1130 |
+
{
|
| 1131 |
+
"epoch": 40.0,
|
| 1132 |
+
"eval_exact_match": 10.6,
|
| 1133 |
+
"eval_f1": 16.340476190476195,
|
| 1134 |
+
"step": 7500
|
| 1135 |
+
},
|
| 1136 |
+
{
|
| 1137 |
+
"epoch": 40.53333333333333,
|
| 1138 |
+
"grad_norm": 0.8436341285705566,
|
| 1139 |
+
"learning_rate": 0.0005,
|
| 1140 |
+
"loss": 0.4813,
|
| 1141 |
+
"step": 7600
|
| 1142 |
+
},
|
| 1143 |
+
{
|
| 1144 |
+
"epoch": 40.99733333333333,
|
| 1145 |
+
"eval_accuracy": 0.5350769230769231,
|
| 1146 |
+
"eval_loss": 2.715898036956787,
|
| 1147 |
+
"eval_runtime": 5.0842,
|
| 1148 |
+
"eval_samples_per_second": 98.344,
|
| 1149 |
+
"eval_steps_per_second": 12.391,
|
| 1150 |
+
"step": 7687
|
| 1151 |
+
},
|
| 1152 |
+
{
|
| 1153 |
+
"epoch": 40.99733333333333,
|
| 1154 |
+
"eval_exact_match": 11.4,
|
| 1155 |
+
"eval_f1": 16.93714285714286,
|
| 1156 |
+
"step": 7687
|
| 1157 |
+
},
|
| 1158 |
+
{
|
| 1159 |
+
"epoch": 41.06666666666667,
|
| 1160 |
+
"grad_norm": 0.5301070809364319,
|
| 1161 |
+
"learning_rate": 0.0005,
|
| 1162 |
+
"loss": 0.4961,
|
| 1163 |
+
"step": 7700
|
| 1164 |
+
},
|
| 1165 |
+
{
|
| 1166 |
+
"epoch": 41.6,
|
| 1167 |
+
"grad_norm": 0.503410816192627,
|
| 1168 |
+
"learning_rate": 0.0005,
|
| 1169 |
+
"loss": 0.4827,
|
| 1170 |
+
"step": 7800
|
| 1171 |
+
},
|
| 1172 |
+
{
|
| 1173 |
+
"epoch": 42.0,
|
| 1174 |
+
"eval_accuracy": 0.5348717948717949,
|
| 1175 |
+
"eval_loss": 2.748167037963867,
|
| 1176 |
+
"eval_runtime": 4.9833,
|
| 1177 |
+
"eval_samples_per_second": 100.335,
|
| 1178 |
+
"eval_steps_per_second": 12.642,
|
| 1179 |
+
"step": 7875
|
| 1180 |
+
},
|
| 1181 |
+
{
|
| 1182 |
+
"epoch": 42.0,
|
| 1183 |
+
"eval_exact_match": 12.2,
|
| 1184 |
+
"eval_f1": 18.29142857142858,
|
| 1185 |
+
"step": 7875
|
| 1186 |
+
},
|
| 1187 |
+
{
|
| 1188 |
+
"epoch": 42.13333333333333,
|
| 1189 |
+
"grad_norm": 0.5958962440490723,
|
| 1190 |
+
"learning_rate": 0.0005,
|
| 1191 |
+
"loss": 0.4967,
|
| 1192 |
+
"step": 7900
|
| 1193 |
+
},
|
| 1194 |
+
{
|
| 1195 |
+
"epoch": 42.666666666666664,
|
| 1196 |
+
"grad_norm": 0.345262736082077,
|
| 1197 |
+
"learning_rate": 0.0005,
|
| 1198 |
+
"loss": 0.492,
|
| 1199 |
+
"step": 8000
|
| 1200 |
+
},
|
| 1201 |
+
{
|
| 1202 |
+
"epoch": 42.99733333333333,
|
| 1203 |
+
"eval_accuracy": 0.5344615384615384,
|
| 1204 |
+
"eval_loss": 2.728513240814209,
|
| 1205 |
+
"eval_runtime": 5.2446,
|
| 1206 |
+
"eval_samples_per_second": 95.336,
|
| 1207 |
+
"eval_steps_per_second": 12.012,
|
| 1208 |
+
"step": 8062
|
| 1209 |
+
},
|
| 1210 |
+
{
|
| 1211 |
+
"epoch": 42.99733333333333,
|
| 1212 |
+
"eval_exact_match": 10.6,
|
| 1213 |
+
"eval_f1": 15.787619047619046,
|
| 1214 |
+
"step": 8062
|
| 1215 |
+
}
|
| 1216 |
+
],
|
| 1217 |
+
"logging_steps": 100,
|
| 1218 |
+
"max_steps": 9350,
|
| 1219 |
+
"num_input_tokens_seen": 0,
|
| 1220 |
+
"num_train_epochs": 50,
|
| 1221 |
+
"save_steps": 500,
|
| 1222 |
+
"stateful_callbacks": {
|
| 1223 |
+
"TrainerControl": {
|
| 1224 |
+
"args": {
|
| 1225 |
+
"should_epoch_stop": false,
|
| 1226 |
+
"should_evaluate": false,
|
| 1227 |
+
"should_log": false,
|
| 1228 |
+
"should_save": true,
|
| 1229 |
+
"should_training_stop": false
|
| 1230 |
+
},
|
| 1231 |
+
"attributes": {}
|
| 1232 |
+
}
|
| 1233 |
+
},
|
| 1234 |
+
"total_flos": 1.4041487384982323e+17,
|
| 1235 |
+
"train_batch_size": 1,
|
| 1236 |
+
"trial_name": null,
|
| 1237 |
+
"trial_params": null
|
| 1238 |
+
}
|
checkpoint-8062/training_args.bin
ADDED
|
@@ -0,0 +1,3 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
version https://git-lfs.github.com/spec/v1
|
| 2 |
+
oid sha256:2776615b5656a131d464a15a3d5283d9bdf82687bc8ffc8d243c778d1158123d
|
| 3 |
+
size 5304
|
checkpoint-8062/vocab.json
ADDED
|
The diff for this file is too large to render.
See raw diff
|
|
|