Commit
•
c6dd428
1
Parent(s):
33f3da9
update 28 tasks lora adapters
Browse filesThis view is limited to 50 files because it contains too many changes.
See raw diff
- llama2_13b_peft/linguistics_puzzles/README.md +74 -0
- llama2_13b_peft/linguistics_puzzles/adapter_config.json +34 -0
- llama2_13b_peft/linguistics_puzzles/adapter_model.safetensors +3 -0
- llama2_13b_peft/linguistics_puzzles/all_results.json +12 -0
- llama2_13b_peft/linguistics_puzzles/eval_results.json +7 -0
- llama2_13b_peft/linguistics_puzzles/special_tokens_map.json +24 -0
- llama2_13b_peft/linguistics_puzzles/tokenizer.model +3 -0
- llama2_13b_peft/linguistics_puzzles/tokenizer_config.json +45 -0
- llama2_13b_peft/linguistics_puzzles/train_results.json +8 -0
- llama2_13b_peft/linguistics_puzzles/trainer_log.jsonl +106 -0
- llama2_13b_peft/linguistics_puzzles/trainer_state.json +767 -0
- llama2_13b_peft/linguistics_puzzles/training_args.bin +3 -0
- llama2_13b_peft/linguistics_puzzles/training_eval_loss.png +0 -0
- llama2_13b_peft/linguistics_puzzles/training_loss.png +0 -0
- llama2_13b_peft/news_commentary_de/README.md +85 -0
- llama2_13b_peft/news_commentary_de/adapter_config.json +33 -0
- llama2_13b_peft/news_commentary_de/adapter_model.safetensors +3 -0
- llama2_13b_peft/news_commentary_de/all_results.json +11 -0
- llama2_13b_peft/news_commentary_de/eval_results.json +7 -0
- llama2_13b_peft/news_commentary_de/special_tokens_map.json +24 -0
- llama2_13b_peft/news_commentary_de/tokenizer.model +3 -0
- llama2_13b_peft/news_commentary_de/tokenizer_config.json +45 -0
- llama2_13b_peft/news_commentary_de/train_results.json +7 -0
- llama2_13b_peft/news_commentary_de/trainer_log.jsonl +0 -0
- llama2_13b_peft/news_commentary_de/trainer_state.json +2990 -0
- llama2_13b_peft/news_commentary_de/training_args.bin +3 -0
- llama2_13b_peft/news_commentary_de/training_eval_loss.png +0 -0
- llama2_13b_peft/news_commentary_de/training_loss.png +0 -0
- llama2_13b_peft/news_commentary_it/README.md +67 -0
- llama2_13b_peft/news_commentary_it/adapter_config.json +34 -0
- llama2_13b_peft/news_commentary_it/adapter_model.safetensors +3 -0
- llama2_13b_peft/news_commentary_it/all_results.json +12 -0
- llama2_13b_peft/news_commentary_it/eval_results.json +7 -0
- llama2_13b_peft/news_commentary_it/special_tokens_map.json +24 -0
- llama2_13b_peft/news_commentary_it/tokenizer.model +3 -0
- llama2_13b_peft/news_commentary_it/tokenizer_config.json +45 -0
- llama2_13b_peft/news_commentary_it/train_results.json +8 -0
- llama2_13b_peft/news_commentary_it/trainer_log.jsonl +0 -0
- llama2_13b_peft/news_commentary_it/trainer_state.json +0 -0
- llama2_13b_peft/news_commentary_it/training_args.bin +3 -0
- llama2_13b_peft/news_commentary_it/training_eval_loss.png +0 -0
- llama2_13b_peft/news_commentary_it/training_loss.png +0 -0
- llama2_13b_peft/topical_chat/README.md +89 -0
- llama2_13b_peft/topical_chat/adapter_config.json +34 -0
- llama2_13b_peft/topical_chat/adapter_model.safetensors +3 -0
- llama2_13b_peft/topical_chat/all_results.json +12 -0
- llama2_13b_peft/topical_chat/eval_results.json +7 -0
- llama2_13b_peft/topical_chat/special_tokens_map.json +24 -0
- llama2_13b_peft/topical_chat/tokenizer.model +3 -0
- llama2_13b_peft/topical_chat/tokenizer_config.json +45 -0
llama2_13b_peft/linguistics_puzzles/README.md
ADDED
|
@@ -0,0 +1,74 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
---
|
| 2 |
+
license: other
|
| 3 |
+
library_name: peft
|
| 4 |
+
tags:
|
| 5 |
+
- llama-factory
|
| 6 |
+
- lora
|
| 7 |
+
- generated_from_trainer
|
| 8 |
+
base_model: /data1/model/llama2/meta-llama/Llama2-13b
|
| 9 |
+
model-index:
|
| 10 |
+
- name: linguistics_puzzles_no_sys
|
| 11 |
+
results: []
|
| 12 |
+
---
|
| 13 |
+
|
| 14 |
+
<!-- This model card has been generated automatically according to the information the Trainer had access to. You
|
| 15 |
+
should probably proofread and complete it, then remove this comment. -->
|
| 16 |
+
|
| 17 |
+
# linguistics_puzzles_no_sys
|
| 18 |
+
|
| 19 |
+
This model is a fine-tuned version of [/data1/model/llama2/meta-llama/Llama2-13b](https://huggingface.co//data1/model/llama2/meta-llama/Llama2-13b) on the linguistics_puzzles_no_sys dataset.
|
| 20 |
+
It achieves the following results on the evaluation set:
|
| 21 |
+
- Loss: 0.5924
|
| 22 |
+
|
| 23 |
+
## Model description
|
| 24 |
+
|
| 25 |
+
More information needed
|
| 26 |
+
|
| 27 |
+
## Intended uses & limitations
|
| 28 |
+
|
| 29 |
+
More information needed
|
| 30 |
+
|
| 31 |
+
## Training and evaluation data
|
| 32 |
+
|
| 33 |
+
More information needed
|
| 34 |
+
|
| 35 |
+
## Training procedure
|
| 36 |
+
|
| 37 |
+
### Training hyperparameters
|
| 38 |
+
|
| 39 |
+
The following hyperparameters were used during training:
|
| 40 |
+
- learning_rate: 5e-05
|
| 41 |
+
- train_batch_size: 4
|
| 42 |
+
- eval_batch_size: 4
|
| 43 |
+
- seed: 42
|
| 44 |
+
- distributed_type: multi-GPU
|
| 45 |
+
- num_devices: 2
|
| 46 |
+
- total_train_batch_size: 8
|
| 47 |
+
- total_eval_batch_size: 8
|
| 48 |
+
- optimizer: Adam with betas=(0.9,0.999) and epsilon=1e-08
|
| 49 |
+
- lr_scheduler_type: cosine
|
| 50 |
+
- lr_scheduler_warmup_steps: 20
|
| 51 |
+
- num_epochs: 5.0
|
| 52 |
+
|
| 53 |
+
### Training results
|
| 54 |
+
|
| 55 |
+
| Training Loss | Epoch | Step | Validation Loss |
|
| 56 |
+
|:-------------:|:------:|:----:|:---------------:|
|
| 57 |
+
| 1.1276 | 0.5263 | 100 | 1.0876 |
|
| 58 |
+
| 0.8128 | 1.0526 | 200 | 0.8153 |
|
| 59 |
+
| 0.6705 | 1.5789 | 300 | 0.6892 |
|
| 60 |
+
| 0.4876 | 2.1053 | 400 | 0.6225 |
|
| 61 |
+
| 0.4435 | 2.6316 | 500 | 0.5924 |
|
| 62 |
+
| 0.2743 | 3.1579 | 600 | 0.6151 |
|
| 63 |
+
| 0.2846 | 3.6842 | 700 | 0.6084 |
|
| 64 |
+
| 0.2069 | 4.2105 | 800 | 0.6427 |
|
| 65 |
+
| 0.172 | 4.7368 | 900 | 0.6495 |
|
| 66 |
+
|
| 67 |
+
|
| 68 |
+
### Framework versions
|
| 69 |
+
|
| 70 |
+
- PEFT 0.10.0
|
| 71 |
+
- Transformers 4.40.0
|
| 72 |
+
- Pytorch 2.2.1
|
| 73 |
+
- Datasets 2.18.0
|
| 74 |
+
- Tokenizers 0.19.1
|
llama2_13b_peft/linguistics_puzzles/adapter_config.json
ADDED
|
@@ -0,0 +1,34 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
{
|
| 2 |
+
"alpha_pattern": {},
|
| 3 |
+
"auto_mapping": null,
|
| 4 |
+
"base_model_name_or_path": "/data1/model/llama2/meta-llama/Llama2-13b",
|
| 5 |
+
"bias": "none",
|
| 6 |
+
"fan_in_fan_out": false,
|
| 7 |
+
"inference_mode": true,
|
| 8 |
+
"init_lora_weights": true,
|
| 9 |
+
"layer_replication": null,
|
| 10 |
+
"layers_pattern": null,
|
| 11 |
+
"layers_to_transform": null,
|
| 12 |
+
"loftq_config": {},
|
| 13 |
+
"lora_alpha": 16,
|
| 14 |
+
"lora_dropout": 0.0,
|
| 15 |
+
"megatron_config": null,
|
| 16 |
+
"megatron_core": "megatron.core",
|
| 17 |
+
"modules_to_save": null,
|
| 18 |
+
"peft_type": "LORA",
|
| 19 |
+
"r": 8,
|
| 20 |
+
"rank_pattern": {},
|
| 21 |
+
"revision": null,
|
| 22 |
+
"target_modules": [
|
| 23 |
+
"o_proj",
|
| 24 |
+
"k_proj",
|
| 25 |
+
"q_proj",
|
| 26 |
+
"up_proj",
|
| 27 |
+
"gate_proj",
|
| 28 |
+
"v_proj",
|
| 29 |
+
"down_proj"
|
| 30 |
+
],
|
| 31 |
+
"task_type": "CAUSAL_LM",
|
| 32 |
+
"use_dora": false,
|
| 33 |
+
"use_rslora": false
|
| 34 |
+
}
|
llama2_13b_peft/linguistics_puzzles/adapter_model.safetensors
ADDED
|
@@ -0,0 +1,3 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
version https://git-lfs.github.com/spec/v1
|
| 2 |
+
oid sha256:ef6dff2a000b0e4ef6a4db9d169cbaa257c1423e45fae8ef4f428ba9852e00f5
|
| 3 |
+
size 125248064
|
llama2_13b_peft/linguistics_puzzles/all_results.json
ADDED
|
@@ -0,0 +1,12 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
{
|
| 2 |
+
"epoch": 5.0,
|
| 3 |
+
"eval_loss": 0.5924356579780579,
|
| 4 |
+
"eval_runtime": 1.9025,
|
| 5 |
+
"eval_samples_per_second": 42.05,
|
| 6 |
+
"eval_steps_per_second": 5.256,
|
| 7 |
+
"total_flos": 2.0275085174217114e+17,
|
| 8 |
+
"train_loss": 0.5822552880487945,
|
| 9 |
+
"train_runtime": 660.0352,
|
| 10 |
+
"train_samples_per_second": 11.515,
|
| 11 |
+
"train_steps_per_second": 1.439
|
| 12 |
+
}
|
llama2_13b_peft/linguistics_puzzles/eval_results.json
ADDED
|
@@ -0,0 +1,7 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
{
|
| 2 |
+
"epoch": 5.0,
|
| 3 |
+
"eval_loss": 0.5924356579780579,
|
| 4 |
+
"eval_runtime": 1.9025,
|
| 5 |
+
"eval_samples_per_second": 42.05,
|
| 6 |
+
"eval_steps_per_second": 5.256
|
| 7 |
+
}
|
llama2_13b_peft/linguistics_puzzles/special_tokens_map.json
ADDED
|
@@ -0,0 +1,24 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
{
|
| 2 |
+
"bos_token": {
|
| 3 |
+
"content": "<s>",
|
| 4 |
+
"lstrip": false,
|
| 5 |
+
"normalized": false,
|
| 6 |
+
"rstrip": false,
|
| 7 |
+
"single_word": false
|
| 8 |
+
},
|
| 9 |
+
"eos_token": {
|
| 10 |
+
"content": "</s>",
|
| 11 |
+
"lstrip": false,
|
| 12 |
+
"normalized": false,
|
| 13 |
+
"rstrip": false,
|
| 14 |
+
"single_word": false
|
| 15 |
+
},
|
| 16 |
+
"pad_token": "</s>",
|
| 17 |
+
"unk_token": {
|
| 18 |
+
"content": "<unk>",
|
| 19 |
+
"lstrip": false,
|
| 20 |
+
"normalized": false,
|
| 21 |
+
"rstrip": false,
|
| 22 |
+
"single_word": false
|
| 23 |
+
}
|
| 24 |
+
}
|
llama2_13b_peft/linguistics_puzzles/tokenizer.model
ADDED
|
@@ -0,0 +1,3 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
version https://git-lfs.github.com/spec/v1
|
| 2 |
+
oid sha256:9e556afd44213b6bd1be2b850ebbbd98f5481437a8021afaf58ee7fb1818d347
|
| 3 |
+
size 499723
|
llama2_13b_peft/linguistics_puzzles/tokenizer_config.json
ADDED
|
@@ -0,0 +1,45 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
{
|
| 2 |
+
"add_bos_token": true,
|
| 3 |
+
"add_eos_token": false,
|
| 4 |
+
"add_prefix_space": true,
|
| 5 |
+
"added_tokens_decoder": {
|
| 6 |
+
"0": {
|
| 7 |
+
"content": "<unk>",
|
| 8 |
+
"lstrip": false,
|
| 9 |
+
"normalized": false,
|
| 10 |
+
"rstrip": false,
|
| 11 |
+
"single_word": false,
|
| 12 |
+
"special": true
|
| 13 |
+
},
|
| 14 |
+
"1": {
|
| 15 |
+
"content": "<s>",
|
| 16 |
+
"lstrip": false,
|
| 17 |
+
"normalized": false,
|
| 18 |
+
"rstrip": false,
|
| 19 |
+
"single_word": false,
|
| 20 |
+
"special": true
|
| 21 |
+
},
|
| 22 |
+
"2": {
|
| 23 |
+
"content": "</s>",
|
| 24 |
+
"lstrip": false,
|
| 25 |
+
"normalized": false,
|
| 26 |
+
"rstrip": false,
|
| 27 |
+
"single_word": false,
|
| 28 |
+
"special": true
|
| 29 |
+
}
|
| 30 |
+
},
|
| 31 |
+
"bos_token": "<s>",
|
| 32 |
+
"chat_template": "{% if messages[0]['role'] == 'system' %}{% set system_message = messages[0]['content'] %}{% endif %}{% if system_message is defined %}{{ system_message }}{% endif %}{% for message in messages %}{% set content = message['content'] %}{% if message['role'] == 'user' %}{{ content }}{% elif message['role'] == 'assistant' %}{{ content + '\\n' }}{% endif %}{% endfor %}",
|
| 33 |
+
"clean_up_tokenization_spaces": false,
|
| 34 |
+
"eos_token": "</s>",
|
| 35 |
+
"legacy": true,
|
| 36 |
+
"model_max_length": 1000000000000000019884624838656,
|
| 37 |
+
"pad_token": "</s>",
|
| 38 |
+
"padding_side": "right",
|
| 39 |
+
"sp_model_kwargs": {},
|
| 40 |
+
"spaces_between_special_tokens": false,
|
| 41 |
+
"split_special_tokens": false,
|
| 42 |
+
"tokenizer_class": "LlamaTokenizer",
|
| 43 |
+
"unk_token": "<unk>",
|
| 44 |
+
"use_default_system_prompt": false
|
| 45 |
+
}
|
llama2_13b_peft/linguistics_puzzles/train_results.json
ADDED
|
@@ -0,0 +1,8 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
{
|
| 2 |
+
"epoch": 5.0,
|
| 3 |
+
"total_flos": 2.0275085174217114e+17,
|
| 4 |
+
"train_loss": 0.5822552880487945,
|
| 5 |
+
"train_runtime": 660.0352,
|
| 6 |
+
"train_samples_per_second": 11.515,
|
| 7 |
+
"train_steps_per_second": 1.439
|
| 8 |
+
}
|
llama2_13b_peft/linguistics_puzzles/trainer_log.jsonl
ADDED
|
@@ -0,0 +1,106 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
{"current_steps": 10, "total_steps": 950, "loss": 2.5922, "eval_loss": null, "predict_loss": null, "reward": null, "accuracy": null, "learning_rate": 2.5e-05, "epoch": 0.05263157894736842, "percentage": 1.05, "elapsed_time": "0:00:07", "remaining_time": "0:12:31"}
|
| 2 |
+
{"current_steps": 20, "total_steps": 950, "loss": 2.3206, "eval_loss": null, "predict_loss": null, "reward": null, "accuracy": null, "learning_rate": 5e-05, "epoch": 0.10526315789473684, "percentage": 2.11, "elapsed_time": "0:00:14", "remaining_time": "0:11:00"}
|
| 3 |
+
{"current_steps": 30, "total_steps": 950, "loss": 1.7229, "eval_loss": null, "predict_loss": null, "reward": null, "accuracy": null, "learning_rate": 4.998573727324295e-05, "epoch": 0.15789473684210525, "percentage": 3.16, "elapsed_time": "0:00:20", "remaining_time": "0:10:32"}
|
| 4 |
+
{"current_steps": 40, "total_steps": 950, "loss": 1.3729, "eval_loss": null, "predict_loss": null, "reward": null, "accuracy": null, "learning_rate": 4.994296536700177e-05, "epoch": 0.21052631578947367, "percentage": 4.21, "elapsed_time": "0:00:26", "remaining_time": "0:10:14"}
|
| 5 |
+
{"current_steps": 50, "total_steps": 950, "loss": 1.3635, "eval_loss": null, "predict_loss": null, "reward": null, "accuracy": null, "learning_rate": 4.987173308479738e-05, "epoch": 0.2631578947368421, "percentage": 5.26, "elapsed_time": "0:00:33", "remaining_time": "0:10:10"}
|
| 6 |
+
{"current_steps": 60, "total_steps": 950, "loss": 1.3315, "eval_loss": null, "predict_loss": null, "reward": null, "accuracy": null, "learning_rate": 4.977212170395598e-05, "epoch": 0.3157894736842105, "percentage": 6.32, "elapsed_time": "0:00:40", "remaining_time": "0:10:00"}
|
| 7 |
+
{"current_steps": 70, "total_steps": 950, "loss": 1.2515, "eval_loss": null, "predict_loss": null, "reward": null, "accuracy": null, "learning_rate": 4.964424488287009e-05, "epoch": 0.3684210526315789, "percentage": 7.37, "elapsed_time": "0:00:46", "remaining_time": "0:09:49"}
|
| 8 |
+
{"current_steps": 80, "total_steps": 950, "loss": 1.1872, "eval_loss": null, "predict_loss": null, "reward": null, "accuracy": null, "learning_rate": 4.948824853131236e-05, "epoch": 0.42105263157894735, "percentage": 8.42, "elapsed_time": "0:00:53", "remaining_time": "0:09:43"}
|
| 9 |
+
{"current_steps": 90, "total_steps": 950, "loss": 1.1552, "eval_loss": null, "predict_loss": null, "reward": null, "accuracy": null, "learning_rate": 4.930431064394977e-05, "epoch": 0.47368421052631576, "percentage": 9.47, "elapsed_time": "0:01:00", "remaining_time": "0:09:34"}
|
| 10 |
+
{"current_steps": 100, "total_steps": 950, "loss": 1.1276, "eval_loss": null, "predict_loss": null, "reward": null, "accuracy": null, "learning_rate": 4.909264109724853e-05, "epoch": 0.5263157894736842, "percentage": 10.53, "elapsed_time": "0:01:06", "remaining_time": "0:09:27"}
|
| 11 |
+
{"current_steps": 100, "total_steps": 950, "loss": null, "eval_loss": 1.0876480340957642, "predict_loss": null, "reward": null, "accuracy": null, "learning_rate": null, "epoch": 0.5263157894736842, "percentage": 10.53, "elapsed_time": "0:01:06", "remaining_time": "0:09:27"}
|
| 12 |
+
{"current_steps": 110, "total_steps": 950, "loss": 1.1756, "eval_loss": null, "predict_loss": null, "reward": null, "accuracy": null, "learning_rate": 4.885348141000122e-05, "epoch": 0.5789473684210527, "percentage": 11.58, "elapsed_time": "0:01:15", "remaining_time": "0:09:37"}
|
| 13 |
+
{"current_steps": 120, "total_steps": 950, "loss": 1.1106, "eval_loss": null, "predict_loss": null, "reward": null, "accuracy": null, "learning_rate": 4.858710446774951e-05, "epoch": 0.631578947368421, "percentage": 12.63, "elapsed_time": "0:01:22", "remaining_time": "0:09:29"}
|
| 14 |
+
{"current_steps": 130, "total_steps": 950, "loss": 1.0175, "eval_loss": null, "predict_loss": null, "reward": null, "accuracy": null, "learning_rate": 4.829381421141671e-05, "epoch": 0.6842105263157895, "percentage": 13.68, "elapsed_time": "0:01:28", "remaining_time": "0:09:19"}
|
| 15 |
+
{"current_steps": 140, "total_steps": 950, "loss": 0.9733, "eval_loss": null, "predict_loss": null, "reward": null, "accuracy": null, "learning_rate": 4.7973945290505766e-05, "epoch": 0.7368421052631579, "percentage": 14.74, "elapsed_time": "0:01:35", "remaining_time": "0:09:10"}
|
| 16 |
+
{"current_steps": 150, "total_steps": 950, "loss": 0.9907, "eval_loss": null, "predict_loss": null, "reward": null, "accuracy": null, "learning_rate": 4.7627862681258037e-05, "epoch": 0.7894736842105263, "percentage": 15.79, "elapsed_time": "0:01:41", "remaining_time": "0:09:03"}
|
| 17 |
+
{"current_steps": 160, "total_steps": 950, "loss": 0.9312, "eval_loss": null, "predict_loss": null, "reward": null, "accuracy": null, "learning_rate": 4.725596127020879e-05, "epoch": 0.8421052631578947, "percentage": 16.84, "elapsed_time": "0:01:48", "remaining_time": "0:08:54"}
|
| 18 |
+
{"current_steps": 170, "total_steps": 950, "loss": 0.9586, "eval_loss": null, "predict_loss": null, "reward": null, "accuracy": null, "learning_rate": 4.685866540361456e-05, "epoch": 0.8947368421052632, "percentage": 17.89, "elapsed_time": "0:01:55", "remaining_time": "0:08:47"}
|
| 19 |
+
{"current_steps": 180, "total_steps": 950, "loss": 0.9595, "eval_loss": null, "predict_loss": null, "reward": null, "accuracy": null, "learning_rate": 4.643642840326627e-05, "epoch": 0.9473684210526315, "percentage": 18.95, "elapsed_time": "0:02:01", "remaining_time": "0:08:39"}
|
| 20 |
+
{"current_steps": 190, "total_steps": 950, "loss": 0.8331, "eval_loss": null, "predict_loss": null, "reward": null, "accuracy": null, "learning_rate": 4.598973204924097e-05, "epoch": 1.0, "percentage": 20.0, "elapsed_time": "0:02:08", "remaining_time": "0:08:32"}
|
| 21 |
+
{"current_steps": 200, "total_steps": 950, "loss": 0.8128, "eval_loss": null, "predict_loss": null, "reward": null, "accuracy": null, "learning_rate": 4.551908603018191e-05, "epoch": 1.0526315789473684, "percentage": 21.05, "elapsed_time": "0:02:14", "remaining_time": "0:08:25"}
|
| 22 |
+
{"current_steps": 200, "total_steps": 950, "loss": null, "eval_loss": 0.8153461217880249, "predict_loss": null, "reward": null, "accuracy": null, "learning_rate": null, "epoch": 1.0526315789473684, "percentage": 21.05, "elapsed_time": "0:02:14", "remaining_time": "0:08:25"}
|
| 23 |
+
{"current_steps": 210, "total_steps": 950, "loss": 0.8186, "eval_loss": null, "predict_loss": null, "reward": null, "accuracy": null, "learning_rate": 4.502502736173462e-05, "epoch": 1.1052631578947367, "percentage": 22.11, "elapsed_time": "0:02:23", "remaining_time": "0:08:26"}
|
| 24 |
+
{"current_steps": 220, "total_steps": 950, "loss": 0.6895, "eval_loss": null, "predict_loss": null, "reward": null, "accuracy": null, "learning_rate": 4.45081197738023e-05, "epoch": 1.1578947368421053, "percentage": 23.16, "elapsed_time": "0:02:30", "remaining_time": "0:08:18"}
|
| 25 |
+
{"current_steps": 230, "total_steps": 950, "loss": 0.7901, "eval_loss": null, "predict_loss": null, "reward": null, "accuracy": null, "learning_rate": 4.3968953067319777e-05, "epoch": 1.2105263157894737, "percentage": 24.21, "elapsed_time": "0:02:36", "remaining_time": "0:08:10"}
|
| 26 |
+
{"current_steps": 240, "total_steps": 950, "loss": 0.704, "eval_loss": null, "predict_loss": null, "reward": null, "accuracy": null, "learning_rate": 4.340814244127993e-05, "epoch": 1.263157894736842, "percentage": 25.26, "elapsed_time": "0:02:43", "remaining_time": "0:08:03"}
|
| 27 |
+
{"current_steps": 250, "total_steps": 950, "loss": 0.6879, "eval_loss": null, "predict_loss": null, "reward": null, "accuracy": null, "learning_rate": 4.282632779078051e-05, "epoch": 1.3157894736842106, "percentage": 26.32, "elapsed_time": "0:02:49", "remaining_time": "0:07:55"}
|
| 28 |
+
{"current_steps": 260, "total_steps": 950, "loss": 0.7563, "eval_loss": null, "predict_loss": null, "reward": null, "accuracy": null, "learning_rate": 4.222417297689217e-05, "epoch": 1.368421052631579, "percentage": 27.37, "elapsed_time": "0:02:56", "remaining_time": "0:07:48"}
|
| 29 |
+
{"current_steps": 270, "total_steps": 950, "loss": 0.6846, "eval_loss": null, "predict_loss": null, "reward": null, "accuracy": null, "learning_rate": 4.160236506918098e-05, "epoch": 1.4210526315789473, "percentage": 28.42, "elapsed_time": "0:03:03", "remaining_time": "0:07:41"}
|
| 30 |
+
{"current_steps": 280, "total_steps": 950, "loss": 0.7155, "eval_loss": null, "predict_loss": null, "reward": null, "accuracy": null, "learning_rate": 4.096161356174959e-05, "epoch": 1.4736842105263157, "percentage": 29.47, "elapsed_time": "0:03:09", "remaining_time": "0:07:34"}
|
| 31 |
+
{"current_steps": 290, "total_steps": 950, "loss": 0.8037, "eval_loss": null, "predict_loss": null, "reward": null, "accuracy": null, "learning_rate": 4.030264956369157e-05, "epoch": 1.526315789473684, "percentage": 30.53, "elapsed_time": "0:03:16", "remaining_time": "0:07:26"}
|
| 32 |
+
{"current_steps": 300, "total_steps": 950, "loss": 0.6705, "eval_loss": null, "predict_loss": null, "reward": null, "accuracy": null, "learning_rate": 3.962622496488269e-05, "epoch": 1.5789473684210527, "percentage": 31.58, "elapsed_time": "0:03:22", "remaining_time": "0:07:19"}
|
| 33 |
+
{"current_steps": 300, "total_steps": 950, "loss": null, "eval_loss": 0.6891714930534363, "predict_loss": null, "reward": null, "accuracy": null, "learning_rate": null, "epoch": 1.5789473684210527, "percentage": 31.58, "elapsed_time": "0:03:22", "remaining_time": "0:07:19"}
|
| 34 |
+
{"current_steps": 310, "total_steps": 950, "loss": 0.6389, "eval_loss": null, "predict_loss": null, "reward": null, "accuracy": null, "learning_rate": 3.893311157806091e-05, "epoch": 1.631578947368421, "percentage": 32.63, "elapsed_time": "0:03:31", "remaining_time": "0:07:17"}
|
| 35 |
+
{"current_steps": 320, "total_steps": 950, "loss": 0.7223, "eval_loss": null, "predict_loss": null, "reward": null, "accuracy": null, "learning_rate": 3.822410025817406e-05, "epoch": 1.6842105263157894, "percentage": 33.68, "elapsed_time": "0:03:38", "remaining_time": "0:07:10"}
|
| 36 |
+
{"current_steps": 330, "total_steps": 950, "loss": 0.6948, "eval_loss": null, "predict_loss": null, "reward": null, "accuracy": null, "learning_rate": 3.7500000000000003e-05, "epoch": 1.736842105263158, "percentage": 34.74, "elapsed_time": "0:03:45", "remaining_time": "0:07:03"}
|
| 37 |
+
{"current_steps": 340, "total_steps": 950, "loss": 0.6658, "eval_loss": null, "predict_loss": null, "reward": null, "accuracy": null, "learning_rate": 3.67616370150689e-05, "epoch": 1.7894736842105263, "percentage": 35.79, "elapsed_time": "0:03:51", "remaining_time": "0:06:55"}
|
| 38 |
+
{"current_steps": 350, "total_steps": 950, "loss": 0.643, "eval_loss": null, "predict_loss": null, "reward": null, "accuracy": null, "learning_rate": 3.600985378894086e-05, "epoch": 1.8421052631578947, "percentage": 36.84, "elapsed_time": "0:03:58", "remaining_time": "0:06:48"}
|
| 39 |
+
{"current_steps": 360, "total_steps": 950, "loss": 0.6537, "eval_loss": null, "predict_loss": null, "reward": null, "accuracy": null, "learning_rate": 3.5245508119914687e-05, "epoch": 1.8947368421052633, "percentage": 37.89, "elapsed_time": "0:04:05", "remaining_time": "0:06:41"}
|
| 40 |
+
{"current_steps": 370, "total_steps": 950, "loss": 0.641, "eval_loss": null, "predict_loss": null, "reward": null, "accuracy": null, "learning_rate": 3.44694721402644e-05, "epoch": 1.9473684210526314, "percentage": 38.95, "elapsed_time": "0:04:12", "remaining_time": "0:06:35"}
|
| 41 |
+
{"current_steps": 380, "total_steps": 950, "loss": 0.6708, "eval_loss": null, "predict_loss": null, "reward": null, "accuracy": null, "learning_rate": 3.3682631321120504e-05, "epoch": 2.0, "percentage": 40.0, "elapsed_time": "0:04:18", "remaining_time": "0:06:27"}
|
| 42 |
+
{"current_steps": 390, "total_steps": 950, "loss": 0.5061, "eval_loss": null, "predict_loss": null, "reward": null, "accuracy": null, "learning_rate": 3.2885883462131394e-05, "epoch": 2.0526315789473686, "percentage": 41.05, "elapsed_time": "0:04:24", "remaining_time": "0:06:20"}
|
| 43 |
+
{"current_steps": 400, "total_steps": 950, "loss": 0.4876, "eval_loss": null, "predict_loss": null, "reward": null, "accuracy": null, "learning_rate": 3.2080137667057595e-05, "epoch": 2.1052631578947367, "percentage": 42.11, "elapsed_time": "0:04:31", "remaining_time": "0:06:13"}
|
| 44 |
+
{"current_steps": 400, "total_steps": 950, "loss": null, "eval_loss": 0.6224929690361023, "predict_loss": null, "reward": null, "accuracy": null, "learning_rate": null, "epoch": 2.1052631578947367, "percentage": 42.11, "elapsed_time": "0:04:31", "remaining_time": "0:06:13"}
|
| 45 |
+
{"current_steps": 410, "total_steps": 950, "loss": 0.485, "eval_loss": null, "predict_loss": null, "reward": null, "accuracy": null, "learning_rate": 3.126631330646802e-05, "epoch": 2.1578947368421053, "percentage": 43.16, "elapsed_time": "0:04:40", "remaining_time": "0:06:09"}
|
| 46 |
+
{"current_steps": 420, "total_steps": 950, "loss": 0.536, "eval_loss": null, "predict_loss": null, "reward": null, "accuracy": null, "learning_rate": 3.0445338968721287e-05, "epoch": 2.2105263157894735, "percentage": 44.21, "elapsed_time": "0:04:47", "remaining_time": "0:06:02"}
|
| 47 |
+
{"current_steps": 430, "total_steps": 950, "loss": 0.4493, "eval_loss": null, "predict_loss": null, "reward": null, "accuracy": null, "learning_rate": 2.961815140042974e-05, "epoch": 2.263157894736842, "percentage": 45.26, "elapsed_time": "0:04:53", "remaining_time": "0:05:55"}
|
| 48 |
+
{"current_steps": 440, "total_steps": 950, "loss": 0.4806, "eval_loss": null, "predict_loss": null, "reward": null, "accuracy": null, "learning_rate": 2.878569443761442e-05, "epoch": 2.3157894736842106, "percentage": 46.32, "elapsed_time": "0:05:00", "remaining_time": "0:05:48"}
|
| 49 |
+
{"current_steps": 450, "total_steps": 950, "loss": 0.4642, "eval_loss": null, "predict_loss": null, "reward": null, "accuracy": null, "learning_rate": 2.7948917928771158e-05, "epoch": 2.3684210526315788, "percentage": 47.37, "elapsed_time": "0:05:07", "remaining_time": "0:05:41"}
|
| 50 |
+
{"current_steps": 460, "total_steps": 950, "loss": 0.4857, "eval_loss": null, "predict_loss": null, "reward": null, "accuracy": null, "learning_rate": 2.7108776651076118e-05, "epoch": 2.4210526315789473, "percentage": 48.42, "elapsed_time": "0:05:13", "remaining_time": "0:05:34"}
|
| 51 |
+
{"current_steps": 470, "total_steps": 950, "loss": 0.4604, "eval_loss": null, "predict_loss": null, "reward": null, "accuracy": null, "learning_rate": 2.6266229220967818e-05, "epoch": 2.473684210526316, "percentage": 49.47, "elapsed_time": "0:05:20", "remaining_time": "0:05:27"}
|
| 52 |
+
{"current_steps": 480, "total_steps": 950, "loss": 0.4294, "eval_loss": null, "predict_loss": null, "reward": null, "accuracy": null, "learning_rate": 2.5422237000348276e-05, "epoch": 2.526315789473684, "percentage": 50.53, "elapsed_time": "0:05:27", "remaining_time": "0:05:20"}
|
| 53 |
+
{"current_steps": 490, "total_steps": 950, "loss": 0.436, "eval_loss": null, "predict_loss": null, "reward": null, "accuracy": null, "learning_rate": 2.4577762999651726e-05, "epoch": 2.5789473684210527, "percentage": 51.58, "elapsed_time": "0:05:34", "remaining_time": "0:05:13"}
|
| 54 |
+
{"current_steps": 500, "total_steps": 950, "loss": 0.4435, "eval_loss": null, "predict_loss": null, "reward": null, "accuracy": null, "learning_rate": 2.3733770779032184e-05, "epoch": 2.6315789473684212, "percentage": 52.63, "elapsed_time": "0:05:40", "remaining_time": "0:05:06"}
|
| 55 |
+
{"current_steps": 500, "total_steps": 950, "loss": null, "eval_loss": 0.5924356579780579, "predict_loss": null, "reward": null, "accuracy": null, "learning_rate": null, "epoch": 2.6315789473684212, "percentage": 52.63, "elapsed_time": "0:05:40", "remaining_time": "0:05:06"}
|
| 56 |
+
{"current_steps": 510, "total_steps": 950, "loss": 0.4128, "eval_loss": null, "predict_loss": null, "reward": null, "accuracy": null, "learning_rate": 2.2891223348923884e-05, "epoch": 2.6842105263157894, "percentage": 53.68, "elapsed_time": "0:05:49", "remaining_time": "0:05:01"}
|
| 57 |
+
{"current_steps": 520, "total_steps": 950, "loss": 0.4201, "eval_loss": null, "predict_loss": null, "reward": null, "accuracy": null, "learning_rate": 2.2051082071228854e-05, "epoch": 2.736842105263158, "percentage": 54.74, "elapsed_time": "0:05:55", "remaining_time": "0:04:54"}
|
| 58 |
+
{"current_steps": 530, "total_steps": 950, "loss": 0.4144, "eval_loss": null, "predict_loss": null, "reward": null, "accuracy": null, "learning_rate": 2.1214305562385592e-05, "epoch": 2.7894736842105265, "percentage": 55.79, "elapsed_time": "0:06:02", "remaining_time": "0:04:47"}
|
| 59 |
+
{"current_steps": 540, "total_steps": 950, "loss": 0.4325, "eval_loss": null, "predict_loss": null, "reward": null, "accuracy": null, "learning_rate": 2.0381848599570276e-05, "epoch": 2.8421052631578947, "percentage": 56.84, "elapsed_time": "0:06:09", "remaining_time": "0:04:40"}
|
| 60 |
+
{"current_steps": 550, "total_steps": 950, "loss": 0.4539, "eval_loss": null, "predict_loss": null, "reward": null, "accuracy": null, "learning_rate": 1.9554661031278712e-05, "epoch": 2.8947368421052633, "percentage": 57.89, "elapsed_time": "0:06:15", "remaining_time": "0:04:33"}
|
| 61 |
+
{"current_steps": 560, "total_steps": 950, "loss": 0.3898, "eval_loss": null, "predict_loss": null, "reward": null, "accuracy": null, "learning_rate": 1.8733686693531985e-05, "epoch": 2.9473684210526314, "percentage": 58.95, "elapsed_time": "0:06:22", "remaining_time": "0:04:26"}
|
| 62 |
+
{"current_steps": 570, "total_steps": 950, "loss": 0.4347, "eval_loss": null, "predict_loss": null, "reward": null, "accuracy": null, "learning_rate": 1.79198623329424e-05, "epoch": 3.0, "percentage": 60.0, "elapsed_time": "0:06:28", "remaining_time": "0:04:19"}
|
| 63 |
+
{"current_steps": 580, "total_steps": 950, "loss": 0.2771, "eval_loss": null, "predict_loss": null, "reward": null, "accuracy": null, "learning_rate": 1.711411653786861e-05, "epoch": 3.0526315789473686, "percentage": 61.05, "elapsed_time": "0:06:35", "remaining_time": "0:04:12"}
|
| 64 |
+
{"current_steps": 590, "total_steps": 950, "loss": 0.2786, "eval_loss": null, "predict_loss": null, "reward": null, "accuracy": null, "learning_rate": 1.6317368678879495e-05, "epoch": 3.1052631578947367, "percentage": 62.11, "elapsed_time": "0:06:41", "remaining_time": "0:04:05"}
|
| 65 |
+
{"current_steps": 600, "total_steps": 950, "loss": 0.2743, "eval_loss": null, "predict_loss": null, "reward": null, "accuracy": null, "learning_rate": 1.55305278597356e-05, "epoch": 3.1578947368421053, "percentage": 63.16, "elapsed_time": "0:06:48", "remaining_time": "0:03:58"}
|
| 66 |
+
{"current_steps": 600, "total_steps": 950, "loss": null, "eval_loss": 0.6151354908943176, "predict_loss": null, "reward": null, "accuracy": null, "learning_rate": null, "epoch": 3.1578947368421053, "percentage": 63.16, "elapsed_time": "0:06:48", "remaining_time": "0:03:58"}
|
| 67 |
+
{"current_steps": 610, "total_steps": 950, "loss": 0.2611, "eval_loss": null, "predict_loss": null, "reward": null, "accuracy": null, "learning_rate": 1.475449188008532e-05, "epoch": 3.2105263157894735, "percentage": 64.21, "elapsed_time": "0:06:57", "remaining_time": "0:03:52"}
|
| 68 |
+
{"current_steps": 620, "total_steps": 950, "loss": 0.237, "eval_loss": null, "predict_loss": null, "reward": null, "accuracy": null, "learning_rate": 1.399014621105914e-05, "epoch": 3.263157894736842, "percentage": 65.26, "elapsed_time": "0:07:03", "remaining_time": "0:03:45"}
|
| 69 |
+
{"current_steps": 630, "total_steps": 950, "loss": 0.2319, "eval_loss": null, "predict_loss": null, "reward": null, "accuracy": null, "learning_rate": 1.3238362984931113e-05, "epoch": 3.3157894736842106, "percentage": 66.32, "elapsed_time": "0:07:10", "remaining_time": "0:03:38"}
|
| 70 |
+
{"current_steps": 640, "total_steps": 950, "loss": 0.2785, "eval_loss": null, "predict_loss": null, "reward": null, "accuracy": null, "learning_rate": 1.2500000000000006e-05, "epoch": 3.3684210526315788, "percentage": 67.37, "elapsed_time": "0:07:17", "remaining_time": "0:03:31"}
|
| 71 |
+
{"current_steps": 650, "total_steps": 950, "loss": 0.3323, "eval_loss": null, "predict_loss": null, "reward": null, "accuracy": null, "learning_rate": 1.1775899741825947e-05, "epoch": 3.4210526315789473, "percentage": 68.42, "elapsed_time": "0:07:23", "remaining_time": "0:03:24"}
|
| 72 |
+
{"current_steps": 660, "total_steps": 950, "loss": 0.2762, "eval_loss": null, "predict_loss": null, "reward": null, "accuracy": null, "learning_rate": 1.1066888421939093e-05, "epoch": 3.473684210526316, "percentage": 69.47, "elapsed_time": "0:07:30", "remaining_time": "0:03:17"}
|
| 73 |
+
{"current_steps": 670, "total_steps": 950, "loss": 0.2982, "eval_loss": null, "predict_loss": null, "reward": null, "accuracy": null, "learning_rate": 1.0373775035117305e-05, "epoch": 3.526315789473684, "percentage": 70.53, "elapsed_time": "0:07:37", "remaining_time": "0:03:11"}
|
| 74 |
+
{"current_steps": 680, "total_steps": 950, "loss": 0.2338, "eval_loss": null, "predict_loss": null, "reward": null, "accuracy": null, "learning_rate": 9.697350436308427e-06, "epoch": 3.5789473684210527, "percentage": 71.58, "elapsed_time": "0:07:43", "remaining_time": "0:03:04"}
|
| 75 |
+
{"current_steps": 690, "total_steps": 950, "loss": 0.2962, "eval_loss": null, "predict_loss": null, "reward": null, "accuracy": null, "learning_rate": 9.038386438250415e-06, "epoch": 3.6315789473684212, "percentage": 72.63, "elapsed_time": "0:07:50", "remaining_time": "0:02:57"}
|
| 76 |
+
{"current_steps": 700, "total_steps": 950, "loss": 0.2846, "eval_loss": null, "predict_loss": null, "reward": null, "accuracy": null, "learning_rate": 8.397634930819021e-06, "epoch": 3.6842105263157894, "percentage": 73.68, "elapsed_time": "0:07:57", "remaining_time": "0:02:50"}
|
| 77 |
+
{"current_steps": 700, "total_steps": 950, "loss": null, "eval_loss": 0.6083844900131226, "predict_loss": null, "reward": null, "accuracy": null, "learning_rate": null, "epoch": 3.6842105263157894, "percentage": 73.68, "elapsed_time": "0:07:57", "remaining_time": "0:02:50"}
|
| 78 |
+
{"current_steps": 710, "total_steps": 950, "loss": 0.2895, "eval_loss": null, "predict_loss": null, "reward": null, "accuracy": null, "learning_rate": 7.775827023107835e-06, "epoch": 3.736842105263158, "percentage": 74.74, "elapsed_time": "0:08:06", "remaining_time": "0:02:44"}
|
| 79 |
+
{"current_steps": 720, "total_steps": 950, "loss": 0.3261, "eval_loss": null, "predict_loss": null, "reward": null, "accuracy": null, "learning_rate": 7.173672209219495e-06, "epoch": 3.7894736842105265, "percentage": 75.79, "elapsed_time": "0:08:13", "remaining_time": "0:02:37"}
|
| 80 |
+
{"current_steps": 730, "total_steps": 950, "loss": 0.2358, "eval_loss": null, "predict_loss": null, "reward": null, "accuracy": null, "learning_rate": 6.591857558720071e-06, "epoch": 3.8421052631578947, "percentage": 76.84, "elapsed_time": "0:08:20", "remaining_time": "0:02:30"}
|
| 81 |
+
{"current_steps": 740, "total_steps": 950, "loss": 0.2723, "eval_loss": null, "predict_loss": null, "reward": null, "accuracy": null, "learning_rate": 6.031046932680229e-06, "epoch": 3.8947368421052633, "percentage": 77.89, "elapsed_time": "0:08:26", "remaining_time": "0:02:23"}
|
| 82 |
+
{"current_steps": 750, "total_steps": 950, "loss": 0.2941, "eval_loss": null, "predict_loss": null, "reward": null, "accuracy": null, "learning_rate": 5.491880226197707e-06, "epoch": 3.9473684210526314, "percentage": 78.95, "elapsed_time": "0:08:33", "remaining_time": "0:02:16"}
|
| 83 |
+
{"current_steps": 760, "total_steps": 950, "loss": 0.2721, "eval_loss": null, "predict_loss": null, "reward": null, "accuracy": null, "learning_rate": 4.9749726382653905e-06, "epoch": 4.0, "percentage": 80.0, "elapsed_time": "0:08:39", "remaining_time": "0:02:09"}
|
| 84 |
+
{"current_steps": 770, "total_steps": 950, "loss": 0.1677, "eval_loss": null, "predict_loss": null, "reward": null, "accuracy": null, "learning_rate": 4.480913969818098e-06, "epoch": 4.052631578947368, "percentage": 81.05, "elapsed_time": "0:08:46", "remaining_time": "0:02:03"}
|
| 85 |
+
{"current_steps": 780, "total_steps": 950, "loss": 0.2291, "eval_loss": null, "predict_loss": null, "reward": null, "accuracy": null, "learning_rate": 4.010267950759025e-06, "epoch": 4.105263157894737, "percentage": 82.11, "elapsed_time": "0:08:53", "remaining_time": "0:01:56"}
|
| 86 |
+
{"current_steps": 790, "total_steps": 950, "loss": 0.1991, "eval_loss": null, "predict_loss": null, "reward": null, "accuracy": null, "learning_rate": 3.5635715967337223e-06, "epoch": 4.157894736842105, "percentage": 83.16, "elapsed_time": "0:09:00", "remaining_time": "0:01:49"}
|
| 87 |
+
{"current_steps": 800, "total_steps": 950, "loss": 0.2069, "eval_loss": null, "predict_loss": null, "reward": null, "accuracy": null, "learning_rate": 3.141334596385448e-06, "epoch": 4.2105263157894735, "percentage": 84.21, "elapsed_time": "0:09:06", "remaining_time": "0:01:42"}
|
| 88 |
+
{"current_steps": 800, "total_steps": 950, "loss": null, "eval_loss": 0.6427180767059326, "predict_loss": null, "reward": null, "accuracy": null, "learning_rate": null, "epoch": 4.2105263157894735, "percentage": 84.21, "elapsed_time": "0:09:06", "remaining_time": "0:01:42"}
|
| 89 |
+
{"current_steps": 810, "total_steps": 950, "loss": 0.2213, "eval_loss": null, "predict_loss": null, "reward": null, "accuracy": null, "learning_rate": 2.7440387297912123e-06, "epoch": 4.2631578947368425, "percentage": 85.26, "elapsed_time": "0:09:15", "remaining_time": "0:01:36"}
|
| 90 |
+
{"current_steps": 820, "total_steps": 950, "loss": 0.2008, "eval_loss": null, "predict_loss": null, "reward": null, "accuracy": null, "learning_rate": 2.372137318741968e-06, "epoch": 4.315789473684211, "percentage": 86.32, "elapsed_time": "0:09:22", "remaining_time": "0:01:29"}
|
| 91 |
+
{"current_steps": 830, "total_steps": 950, "loss": 0.2178, "eval_loss": null, "predict_loss": null, "reward": null, "accuracy": null, "learning_rate": 2.026054709494235e-06, "epoch": 4.368421052631579, "percentage": 87.37, "elapsed_time": "0:09:28", "remaining_time": "0:01:22"}
|
| 92 |
+
{"current_steps": 840, "total_steps": 950, "loss": 0.1878, "eval_loss": null, "predict_loss": null, "reward": null, "accuracy": null, "learning_rate": 1.7061857885832893e-06, "epoch": 4.421052631578947, "percentage": 88.42, "elapsed_time": "0:09:35", "remaining_time": "0:01:15"}
|
| 93 |
+
{"current_steps": 850, "total_steps": 950, "loss": 0.1733, "eval_loss": null, "predict_loss": null, "reward": null, "accuracy": null, "learning_rate": 1.4128955322504966e-06, "epoch": 4.473684210526316, "percentage": 89.47, "elapsed_time": "0:09:41", "remaining_time": "0:01:08"}
|
| 94 |
+
{"current_steps": 860, "total_steps": 950, "loss": 0.193, "eval_loss": null, "predict_loss": null, "reward": null, "accuracy": null, "learning_rate": 1.1465185899987797e-06, "epoch": 4.526315789473684, "percentage": 90.53, "elapsed_time": "0:09:48", "remaining_time": "0:01:01"}
|
| 95 |
+
{"current_steps": 870, "total_steps": 950, "loss": 0.1802, "eval_loss": null, "predict_loss": null, "reward": null, "accuracy": null, "learning_rate": 9.073589027514789e-07, "epoch": 4.578947368421053, "percentage": 91.58, "elapsed_time": "0:09:55", "remaining_time": "0:00:54"}
|
| 96 |
+
{"current_steps": 880, "total_steps": 950, "loss": 0.1736, "eval_loss": null, "predict_loss": null, "reward": null, "accuracy": null, "learning_rate": 6.956893560502359e-07, "epoch": 4.631578947368421, "percentage": 92.63, "elapsed_time": "0:10:01", "remaining_time": "0:00:47"}
|
| 97 |
+
{"current_steps": 890, "total_steps": 950, "loss": 0.1761, "eval_loss": null, "predict_loss": null, "reward": null, "accuracy": null, "learning_rate": 5.117514686876379e-07, "epoch": 4.684210526315789, "percentage": 93.68, "elapsed_time": "0:10:08", "remaining_time": "0:00:41"}
|
| 98 |
+
{"current_steps": 900, "total_steps": 950, "loss": 0.172, "eval_loss": null, "predict_loss": null, "reward": null, "accuracy": null, "learning_rate": 3.557551171299051e-07, "epoch": 4.7368421052631575, "percentage": 94.74, "elapsed_time": "0:10:14", "remaining_time": "0:00:34"}
|
| 99 |
+
{"current_steps": 900, "total_steps": 950, "loss": null, "eval_loss": 0.6494551301002502, "predict_loss": null, "reward": null, "accuracy": null, "learning_rate": null, "epoch": 4.7368421052631575, "percentage": 94.74, "elapsed_time": "0:10:14", "remaining_time": "0:00:34"}
|
| 100 |
+
{"current_steps": 910, "total_steps": 950, "loss": 0.1734, "eval_loss": null, "predict_loss": null, "reward": null, "accuracy": null, "learning_rate": 2.27878296044029e-07, "epoch": 4.7894736842105265, "percentage": 95.79, "elapsed_time": "0:10:23", "remaining_time": "0:00:27"}
|
| 101 |
+
{"current_steps": 920, "total_steps": 950, "loss": 0.1954, "eval_loss": null, "predict_loss": null, "reward": null, "accuracy": null, "learning_rate": 1.2826691520262114e-07, "epoch": 4.842105263157895, "percentage": 96.84, "elapsed_time": "0:10:30", "remaining_time": "0:00:20"}
|
| 102 |
+
{"current_steps": 930, "total_steps": 950, "loss": 0.1744, "eval_loss": null, "predict_loss": null, "reward": null, "accuracy": null, "learning_rate": 5.7034632998231865e-08, "epoch": 4.894736842105263, "percentage": 97.89, "elapsed_time": "0:10:37", "remaining_time": "0:00:13"}
|
| 103 |
+
{"current_steps": 940, "total_steps": 950, "loss": 0.1778, "eval_loss": null, "predict_loss": null, "reward": null, "accuracy": null, "learning_rate": 1.4262726757049982e-08, "epoch": 4.947368421052632, "percentage": 98.95, "elapsed_time": "0:10:43", "remaining_time": "0:00:06"}
|
| 104 |
+
{"current_steps": 950, "total_steps": 950, "loss": 0.1836, "eval_loss": null, "predict_loss": null, "reward": null, "accuracy": null, "learning_rate": 0.0, "epoch": 5.0, "percentage": 100.0, "elapsed_time": "0:10:50", "remaining_time": "0:00:00"}
|
| 105 |
+
{"current_steps": 950, "total_steps": 950, "loss": null, "eval_loss": null, "predict_loss": null, "reward": null, "accuracy": null, "learning_rate": null, "epoch": 5.0, "percentage": 100.0, "elapsed_time": "0:10:50", "remaining_time": "0:00:00"}
|
| 106 |
+
{"current_steps": 10, "total_steps": 10, "loss": null, "eval_loss": 0.5924356579780579, "predict_loss": null, "reward": null, "accuracy": null, "learning_rate": null, "epoch": 5.0, "percentage": 100.0, "elapsed_time": "0:11:00", "remaining_time": "0:00:00"}
|
llama2_13b_peft/linguistics_puzzles/trainer_state.json
ADDED
|
@@ -0,0 +1,767 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
{
|
| 2 |
+
"best_metric": 0.5924356579780579,
|
| 3 |
+
"best_model_checkpoint": "ckpt/llama2_13b_other/linguistics_puzzles_no_sys/checkpoint-500",
|
| 4 |
+
"epoch": 5.0,
|
| 5 |
+
"eval_steps": 100,
|
| 6 |
+
"global_step": 950,
|
| 7 |
+
"is_hyper_param_search": false,
|
| 8 |
+
"is_local_process_zero": true,
|
| 9 |
+
"is_world_process_zero": true,
|
| 10 |
+
"log_history": [
|
| 11 |
+
{
|
| 12 |
+
"epoch": 0.05263157894736842,
|
| 13 |
+
"grad_norm": 1.5050264596939087,
|
| 14 |
+
"learning_rate": 2.5e-05,
|
| 15 |
+
"loss": 2.5922,
|
| 16 |
+
"step": 10
|
| 17 |
+
},
|
| 18 |
+
{
|
| 19 |
+
"epoch": 0.10526315789473684,
|
| 20 |
+
"grad_norm": 1.5525988340377808,
|
| 21 |
+
"learning_rate": 5e-05,
|
| 22 |
+
"loss": 2.3206,
|
| 23 |
+
"step": 20
|
| 24 |
+
},
|
| 25 |
+
{
|
| 26 |
+
"epoch": 0.15789473684210525,
|
| 27 |
+
"grad_norm": 1.7404705286026,
|
| 28 |
+
"learning_rate": 4.998573727324295e-05,
|
| 29 |
+
"loss": 1.7229,
|
| 30 |
+
"step": 30
|
| 31 |
+
},
|
| 32 |
+
{
|
| 33 |
+
"epoch": 0.21052631578947367,
|
| 34 |
+
"grad_norm": 1.8962088823318481,
|
| 35 |
+
"learning_rate": 4.994296536700177e-05,
|
| 36 |
+
"loss": 1.3729,
|
| 37 |
+
"step": 40
|
| 38 |
+
},
|
| 39 |
+
{
|
| 40 |
+
"epoch": 0.2631578947368421,
|
| 41 |
+
"grad_norm": 1.776729941368103,
|
| 42 |
+
"learning_rate": 4.987173308479738e-05,
|
| 43 |
+
"loss": 1.3635,
|
| 44 |
+
"step": 50
|
| 45 |
+
},
|
| 46 |
+
{
|
| 47 |
+
"epoch": 0.3157894736842105,
|
| 48 |
+
"grad_norm": 11.020795822143555,
|
| 49 |
+
"learning_rate": 4.977212170395598e-05,
|
| 50 |
+
"loss": 1.3315,
|
| 51 |
+
"step": 60
|
| 52 |
+
},
|
| 53 |
+
{
|
| 54 |
+
"epoch": 0.3684210526315789,
|
| 55 |
+
"grad_norm": 2.192176580429077,
|
| 56 |
+
"learning_rate": 4.964424488287009e-05,
|
| 57 |
+
"loss": 1.2515,
|
| 58 |
+
"step": 70
|
| 59 |
+
},
|
| 60 |
+
{
|
| 61 |
+
"epoch": 0.42105263157894735,
|
| 62 |
+
"grad_norm": 2.4063496589660645,
|
| 63 |
+
"learning_rate": 4.948824853131236e-05,
|
| 64 |
+
"loss": 1.1872,
|
| 65 |
+
"step": 80
|
| 66 |
+
},
|
| 67 |
+
{
|
| 68 |
+
"epoch": 0.47368421052631576,
|
| 69 |
+
"grad_norm": 2.7862613201141357,
|
| 70 |
+
"learning_rate": 4.930431064394977e-05,
|
| 71 |
+
"loss": 1.1552,
|
| 72 |
+
"step": 90
|
| 73 |
+
},
|
| 74 |
+
{
|
| 75 |
+
"epoch": 0.5263157894736842,
|
| 76 |
+
"grad_norm": 3.5330026149749756,
|
| 77 |
+
"learning_rate": 4.909264109724853e-05,
|
| 78 |
+
"loss": 1.1276,
|
| 79 |
+
"step": 100
|
| 80 |
+
},
|
| 81 |
+
{
|
| 82 |
+
"epoch": 0.5263157894736842,
|
| 83 |
+
"eval_loss": 1.0876480340957642,
|
| 84 |
+
"eval_runtime": 1.9022,
|
| 85 |
+
"eval_samples_per_second": 42.057,
|
| 86 |
+
"eval_steps_per_second": 5.257,
|
| 87 |
+
"step": 100
|
| 88 |
+
},
|
| 89 |
+
{
|
| 90 |
+
"epoch": 0.5789473684210527,
|
| 91 |
+
"grad_norm": 2.4774415493011475,
|
| 92 |
+
"learning_rate": 4.885348141000122e-05,
|
| 93 |
+
"loss": 1.1756,
|
| 94 |
+
"step": 110
|
| 95 |
+
},
|
| 96 |
+
{
|
| 97 |
+
"epoch": 0.631578947368421,
|
| 98 |
+
"grad_norm": 2.380500555038452,
|
| 99 |
+
"learning_rate": 4.858710446774951e-05,
|
| 100 |
+
"loss": 1.1106,
|
| 101 |
+
"step": 120
|
| 102 |
+
},
|
| 103 |
+
{
|
| 104 |
+
"epoch": 0.6842105263157895,
|
| 105 |
+
"grad_norm": 3.0656540393829346,
|
| 106 |
+
"learning_rate": 4.829381421141671e-05,
|
| 107 |
+
"loss": 1.0175,
|
| 108 |
+
"step": 130
|
| 109 |
+
},
|
| 110 |
+
{
|
| 111 |
+
"epoch": 0.7368421052631579,
|
| 112 |
+
"grad_norm": 5.256251811981201,
|
| 113 |
+
"learning_rate": 4.7973945290505766e-05,
|
| 114 |
+
"loss": 0.9733,
|
| 115 |
+
"step": 140
|
| 116 |
+
},
|
| 117 |
+
{
|
| 118 |
+
"epoch": 0.7894736842105263,
|
| 119 |
+
"grad_norm": 2.674135446548462,
|
| 120 |
+
"learning_rate": 4.7627862681258037e-05,
|
| 121 |
+
"loss": 0.9907,
|
| 122 |
+
"step": 150
|
| 123 |
+
},
|
| 124 |
+
{
|
| 125 |
+
"epoch": 0.8421052631578947,
|
| 126 |
+
"grad_norm": 3.5206069946289062,
|
| 127 |
+
"learning_rate": 4.725596127020879e-05,
|
| 128 |
+
"loss": 0.9312,
|
| 129 |
+
"step": 160
|
| 130 |
+
},
|
| 131 |
+
{
|
| 132 |
+
"epoch": 0.8947368421052632,
|
| 133 |
+
"grad_norm": 3.4086978435516357,
|
| 134 |
+
"learning_rate": 4.685866540361456e-05,
|
| 135 |
+
"loss": 0.9586,
|
| 136 |
+
"step": 170
|
| 137 |
+
},
|
| 138 |
+
{
|
| 139 |
+
"epoch": 0.9473684210526315,
|
| 140 |
+
"grad_norm": 4.591642379760742,
|
| 141 |
+
"learning_rate": 4.643642840326627e-05,
|
| 142 |
+
"loss": 0.9595,
|
| 143 |
+
"step": 180
|
| 144 |
+
},
|
| 145 |
+
{
|
| 146 |
+
"epoch": 1.0,
|
| 147 |
+
"grad_norm": 2.8823249340057373,
|
| 148 |
+
"learning_rate": 4.598973204924097e-05,
|
| 149 |
+
"loss": 0.8331,
|
| 150 |
+
"step": 190
|
| 151 |
+
},
|
| 152 |
+
{
|
| 153 |
+
"epoch": 1.0526315789473684,
|
| 154 |
+
"grad_norm": 3.7064428329467773,
|
| 155 |
+
"learning_rate": 4.551908603018191e-05,
|
| 156 |
+
"loss": 0.8128,
|
| 157 |
+
"step": 200
|
| 158 |
+
},
|
| 159 |
+
{
|
| 160 |
+
"epoch": 1.0526315789473684,
|
| 161 |
+
"eval_loss": 0.8153461217880249,
|
| 162 |
+
"eval_runtime": 1.9192,
|
| 163 |
+
"eval_samples_per_second": 41.684,
|
| 164 |
+
"eval_steps_per_second": 5.21,
|
| 165 |
+
"step": 200
|
| 166 |
+
},
|
| 167 |
+
{
|
| 168 |
+
"epoch": 1.1052631578947367,
|
| 169 |
+
"grad_norm": 4.2386274337768555,
|
| 170 |
+
"learning_rate": 4.502502736173462e-05,
|
| 171 |
+
"loss": 0.8186,
|
| 172 |
+
"step": 210
|
| 173 |
+
},
|
| 174 |
+
{
|
| 175 |
+
"epoch": 1.1578947368421053,
|
| 176 |
+
"grad_norm": 3.1767256259918213,
|
| 177 |
+
"learning_rate": 4.45081197738023e-05,
|
| 178 |
+
"loss": 0.6895,
|
| 179 |
+
"step": 220
|
| 180 |
+
},
|
| 181 |
+
{
|
| 182 |
+
"epoch": 1.2105263157894737,
|
| 183 |
+
"grad_norm": 3.748518466949463,
|
| 184 |
+
"learning_rate": 4.3968953067319777e-05,
|
| 185 |
+
"loss": 0.7901,
|
| 186 |
+
"step": 230
|
| 187 |
+
},
|
| 188 |
+
{
|
| 189 |
+
"epoch": 1.263157894736842,
|
| 190 |
+
"grad_norm": 3.807053565979004,
|
| 191 |
+
"learning_rate": 4.340814244127993e-05,
|
| 192 |
+
"loss": 0.704,
|
| 193 |
+
"step": 240
|
| 194 |
+
},
|
| 195 |
+
{
|
| 196 |
+
"epoch": 1.3157894736842106,
|
| 197 |
+
"grad_norm": 5.013542175292969,
|
| 198 |
+
"learning_rate": 4.282632779078051e-05,
|
| 199 |
+
"loss": 0.6879,
|
| 200 |
+
"step": 250
|
| 201 |
+
},
|
| 202 |
+
{
|
| 203 |
+
"epoch": 1.368421052631579,
|
| 204 |
+
"grad_norm": 4.752715110778809,
|
| 205 |
+
"learning_rate": 4.222417297689217e-05,
|
| 206 |
+
"loss": 0.7563,
|
| 207 |
+
"step": 260
|
| 208 |
+
},
|
| 209 |
+
{
|
| 210 |
+
"epoch": 1.4210526315789473,
|
| 211 |
+
"grad_norm": 3.6476950645446777,
|
| 212 |
+
"learning_rate": 4.160236506918098e-05,
|
| 213 |
+
"loss": 0.6846,
|
| 214 |
+
"step": 270
|
| 215 |
+
},
|
| 216 |
+
{
|
| 217 |
+
"epoch": 1.4736842105263157,
|
| 218 |
+
"grad_norm": 3.8758108615875244,
|
| 219 |
+
"learning_rate": 4.096161356174959e-05,
|
| 220 |
+
"loss": 0.7155,
|
| 221 |
+
"step": 280
|
| 222 |
+
},
|
| 223 |
+
{
|
| 224 |
+
"epoch": 1.526315789473684,
|
| 225 |
+
"grad_norm": 4.166601657867432,
|
| 226 |
+
"learning_rate": 4.030264956369157e-05,
|
| 227 |
+
"loss": 0.8037,
|
| 228 |
+
"step": 290
|
| 229 |
+
},
|
| 230 |
+
{
|
| 231 |
+
"epoch": 1.5789473684210527,
|
| 232 |
+
"grad_norm": 4.603171348571777,
|
| 233 |
+
"learning_rate": 3.962622496488269e-05,
|
| 234 |
+
"loss": 0.6705,
|
| 235 |
+
"step": 300
|
| 236 |
+
},
|
| 237 |
+
{
|
| 238 |
+
"epoch": 1.5789473684210527,
|
| 239 |
+
"eval_loss": 0.6891714930534363,
|
| 240 |
+
"eval_runtime": 1.9174,
|
| 241 |
+
"eval_samples_per_second": 41.724,
|
| 242 |
+
"eval_steps_per_second": 5.216,
|
| 243 |
+
"step": 300
|
| 244 |
+
},
|
| 245 |
+
{
|
| 246 |
+
"epoch": 1.631578947368421,
|
| 247 |
+
"grad_norm": 3.820142984390259,
|
| 248 |
+
"learning_rate": 3.893311157806091e-05,
|
| 249 |
+
"loss": 0.6389,
|
| 250 |
+
"step": 310
|
| 251 |
+
},
|
| 252 |
+
{
|
| 253 |
+
"epoch": 1.6842105263157894,
|
| 254 |
+
"grad_norm": 5.900814533233643,
|
| 255 |
+
"learning_rate": 3.822410025817406e-05,
|
| 256 |
+
"loss": 0.7223,
|
| 257 |
+
"step": 320
|
| 258 |
+
},
|
| 259 |
+
{
|
| 260 |
+
"epoch": 1.736842105263158,
|
| 261 |
+
"grad_norm": 4.315140724182129,
|
| 262 |
+
"learning_rate": 3.7500000000000003e-05,
|
| 263 |
+
"loss": 0.6948,
|
| 264 |
+
"step": 330
|
| 265 |
+
},
|
| 266 |
+
{
|
| 267 |
+
"epoch": 1.7894736842105263,
|
| 268 |
+
"grad_norm": 4.747324466705322,
|
| 269 |
+
"learning_rate": 3.67616370150689e-05,
|
| 270 |
+
"loss": 0.6658,
|
| 271 |
+
"step": 340
|
| 272 |
+
},
|
| 273 |
+
{
|
| 274 |
+
"epoch": 1.8421052631578947,
|
| 275 |
+
"grad_norm": 3.504014492034912,
|
| 276 |
+
"learning_rate": 3.600985378894086e-05,
|
| 277 |
+
"loss": 0.643,
|
| 278 |
+
"step": 350
|
| 279 |
+
},
|
| 280 |
+
{
|
| 281 |
+
"epoch": 1.8947368421052633,
|
| 282 |
+
"grad_norm": 5.181077480316162,
|
| 283 |
+
"learning_rate": 3.5245508119914687e-05,
|
| 284 |
+
"loss": 0.6537,
|
| 285 |
+
"step": 360
|
| 286 |
+
},
|
| 287 |
+
{
|
| 288 |
+
"epoch": 1.9473684210526314,
|
| 289 |
+
"grad_norm": 5.073149681091309,
|
| 290 |
+
"learning_rate": 3.44694721402644e-05,
|
| 291 |
+
"loss": 0.641,
|
| 292 |
+
"step": 370
|
| 293 |
+
},
|
| 294 |
+
{
|
| 295 |
+
"epoch": 2.0,
|
| 296 |
+
"grad_norm": 5.070895671844482,
|
| 297 |
+
"learning_rate": 3.3682631321120504e-05,
|
| 298 |
+
"loss": 0.6708,
|
| 299 |
+
"step": 380
|
| 300 |
+
},
|
| 301 |
+
{
|
| 302 |
+
"epoch": 2.0526315789473686,
|
| 303 |
+
"grad_norm": 5.305852890014648,
|
| 304 |
+
"learning_rate": 3.2885883462131394e-05,
|
| 305 |
+
"loss": 0.5061,
|
| 306 |
+
"step": 390
|
| 307 |
+
},
|
| 308 |
+
{
|
| 309 |
+
"epoch": 2.1052631578947367,
|
| 310 |
+
"grad_norm": 6.452213287353516,
|
| 311 |
+
"learning_rate": 3.2080137667057595e-05,
|
| 312 |
+
"loss": 0.4876,
|
| 313 |
+
"step": 400
|
| 314 |
+
},
|
| 315 |
+
{
|
| 316 |
+
"epoch": 2.1052631578947367,
|
| 317 |
+
"eval_loss": 0.6224929690361023,
|
| 318 |
+
"eval_runtime": 1.9167,
|
| 319 |
+
"eval_samples_per_second": 41.739,
|
| 320 |
+
"eval_steps_per_second": 5.217,
|
| 321 |
+
"step": 400
|
| 322 |
+
},
|
| 323 |
+
{
|
| 324 |
+
"epoch": 2.1578947368421053,
|
| 325 |
+
"grad_norm": 3.6080775260925293,
|
| 326 |
+
"learning_rate": 3.126631330646802e-05,
|
| 327 |
+
"loss": 0.485,
|
| 328 |
+
"step": 410
|
| 329 |
+
},
|
| 330 |
+
{
|
| 331 |
+
"epoch": 2.2105263157894735,
|
| 332 |
+
"grad_norm": 2.2630574703216553,
|
| 333 |
+
"learning_rate": 3.0445338968721287e-05,
|
| 334 |
+
"loss": 0.536,
|
| 335 |
+
"step": 420
|
| 336 |
+
},
|
| 337 |
+
{
|
| 338 |
+
"epoch": 2.263157894736842,
|
| 339 |
+
"grad_norm": 4.616273880004883,
|
| 340 |
+
"learning_rate": 2.961815140042974e-05,
|
| 341 |
+
"loss": 0.4493,
|
| 342 |
+
"step": 430
|
| 343 |
+
},
|
| 344 |
+
{
|
| 345 |
+
"epoch": 2.3157894736842106,
|
| 346 |
+
"grad_norm": 4.5297956466674805,
|
| 347 |
+
"learning_rate": 2.878569443761442e-05,
|
| 348 |
+
"loss": 0.4806,
|
| 349 |
+
"step": 440
|
| 350 |
+
},
|
| 351 |
+
{
|
| 352 |
+
"epoch": 2.3684210526315788,
|
| 353 |
+
"grad_norm": 4.910376071929932,
|
| 354 |
+
"learning_rate": 2.7948917928771158e-05,
|
| 355 |
+
"loss": 0.4642,
|
| 356 |
+
"step": 450
|
| 357 |
+
},
|
| 358 |
+
{
|
| 359 |
+
"epoch": 2.4210526315789473,
|
| 360 |
+
"grad_norm": 4.3276801109313965,
|
| 361 |
+
"learning_rate": 2.7108776651076118e-05,
|
| 362 |
+
"loss": 0.4857,
|
| 363 |
+
"step": 460
|
| 364 |
+
},
|
| 365 |
+
{
|
| 366 |
+
"epoch": 2.473684210526316,
|
| 367 |
+
"grad_norm": 3.657116413116455,
|
| 368 |
+
"learning_rate": 2.6266229220967818e-05,
|
| 369 |
+
"loss": 0.4604,
|
| 370 |
+
"step": 470
|
| 371 |
+
},
|
| 372 |
+
{
|
| 373 |
+
"epoch": 2.526315789473684,
|
| 374 |
+
"grad_norm": 4.7539896965026855,
|
| 375 |
+
"learning_rate": 2.5422237000348276e-05,
|
| 376 |
+
"loss": 0.4294,
|
| 377 |
+
"step": 480
|
| 378 |
+
},
|
| 379 |
+
{
|
| 380 |
+
"epoch": 2.5789473684210527,
|
| 381 |
+
"grad_norm": 4.227921962738037,
|
| 382 |
+
"learning_rate": 2.4577762999651726e-05,
|
| 383 |
+
"loss": 0.436,
|
| 384 |
+
"step": 490
|
| 385 |
+
},
|
| 386 |
+
{
|
| 387 |
+
"epoch": 2.6315789473684212,
|
| 388 |
+
"grad_norm": 6.821872234344482,
|
| 389 |
+
"learning_rate": 2.3733770779032184e-05,
|
| 390 |
+
"loss": 0.4435,
|
| 391 |
+
"step": 500
|
| 392 |
+
},
|
| 393 |
+
{
|
| 394 |
+
"epoch": 2.6315789473684212,
|
| 395 |
+
"eval_loss": 0.5924356579780579,
|
| 396 |
+
"eval_runtime": 1.9193,
|
| 397 |
+
"eval_samples_per_second": 41.683,
|
| 398 |
+
"eval_steps_per_second": 5.21,
|
| 399 |
+
"step": 500
|
| 400 |
+
},
|
| 401 |
+
{
|
| 402 |
+
"epoch": 2.6842105263157894,
|
| 403 |
+
"grad_norm": 4.023755073547363,
|
| 404 |
+
"learning_rate": 2.2891223348923884e-05,
|
| 405 |
+
"loss": 0.4128,
|
| 406 |
+
"step": 510
|
| 407 |
+
},
|
| 408 |
+
{
|
| 409 |
+
"epoch": 2.736842105263158,
|
| 410 |
+
"grad_norm": 4.245009899139404,
|
| 411 |
+
"learning_rate": 2.2051082071228854e-05,
|
| 412 |
+
"loss": 0.4201,
|
| 413 |
+
"step": 520
|
| 414 |
+
},
|
| 415 |
+
{
|
| 416 |
+
"epoch": 2.7894736842105265,
|
| 417 |
+
"grad_norm": 7.485212326049805,
|
| 418 |
+
"learning_rate": 2.1214305562385592e-05,
|
| 419 |
+
"loss": 0.4144,
|
| 420 |
+
"step": 530
|
| 421 |
+
},
|
| 422 |
+
{
|
| 423 |
+
"epoch": 2.8421052631578947,
|
| 424 |
+
"grad_norm": 3.890044689178467,
|
| 425 |
+
"learning_rate": 2.0381848599570276e-05,
|
| 426 |
+
"loss": 0.4325,
|
| 427 |
+
"step": 540
|
| 428 |
+
},
|
| 429 |
+
{
|
| 430 |
+
"epoch": 2.8947368421052633,
|
| 431 |
+
"grad_norm": 5.785126686096191,
|
| 432 |
+
"learning_rate": 1.9554661031278712e-05,
|
| 433 |
+
"loss": 0.4539,
|
| 434 |
+
"step": 550
|
| 435 |
+
},
|
| 436 |
+
{
|
| 437 |
+
"epoch": 2.9473684210526314,
|
| 438 |
+
"grad_norm": 3.959681272506714,
|
| 439 |
+
"learning_rate": 1.8733686693531985e-05,
|
| 440 |
+
"loss": 0.3898,
|
| 441 |
+
"step": 560
|
| 442 |
+
},
|
| 443 |
+
{
|
| 444 |
+
"epoch": 3.0,
|
| 445 |
+
"grad_norm": 6.1470160484313965,
|
| 446 |
+
"learning_rate": 1.79198623329424e-05,
|
| 447 |
+
"loss": 0.4347,
|
| 448 |
+
"step": 570
|
| 449 |
+
},
|
| 450 |
+
{
|
| 451 |
+
"epoch": 3.0526315789473686,
|
| 452 |
+
"grad_norm": 6.080893039703369,
|
| 453 |
+
"learning_rate": 1.711411653786861e-05,
|
| 454 |
+
"loss": 0.2771,
|
| 455 |
+
"step": 580
|
| 456 |
+
},
|
| 457 |
+
{
|
| 458 |
+
"epoch": 3.1052631578947367,
|
| 459 |
+
"grad_norm": 3.995936155319214,
|
| 460 |
+
"learning_rate": 1.6317368678879495e-05,
|
| 461 |
+
"loss": 0.2786,
|
| 462 |
+
"step": 590
|
| 463 |
+
},
|
| 464 |
+
{
|
| 465 |
+
"epoch": 3.1578947368421053,
|
| 466 |
+
"grad_norm": 4.9943084716796875,
|
| 467 |
+
"learning_rate": 1.55305278597356e-05,
|
| 468 |
+
"loss": 0.2743,
|
| 469 |
+
"step": 600
|
| 470 |
+
},
|
| 471 |
+
{
|
| 472 |
+
"epoch": 3.1578947368421053,
|
| 473 |
+
"eval_loss": 0.6151354908943176,
|
| 474 |
+
"eval_runtime": 1.9185,
|
| 475 |
+
"eval_samples_per_second": 41.7,
|
| 476 |
+
"eval_steps_per_second": 5.212,
|
| 477 |
+
"step": 600
|
| 478 |
+
},
|
| 479 |
+
{
|
| 480 |
+
"epoch": 3.2105263157894735,
|
| 481 |
+
"grad_norm": 3.650193452835083,
|
| 482 |
+
"learning_rate": 1.475449188008532e-05,
|
| 483 |
+
"loss": 0.2611,
|
| 484 |
+
"step": 610
|
| 485 |
+
},
|
| 486 |
+
{
|
| 487 |
+
"epoch": 3.263157894736842,
|
| 488 |
+
"grad_norm": 3.5425643920898438,
|
| 489 |
+
"learning_rate": 1.399014621105914e-05,
|
| 490 |
+
"loss": 0.237,
|
| 491 |
+
"step": 620
|
| 492 |
+
},
|
| 493 |
+
{
|
| 494 |
+
"epoch": 3.3157894736842106,
|
| 495 |
+
"grad_norm": 4.187167644500732,
|
| 496 |
+
"learning_rate": 1.3238362984931113e-05,
|
| 497 |
+
"loss": 0.2319,
|
| 498 |
+
"step": 630
|
| 499 |
+
},
|
| 500 |
+
{
|
| 501 |
+
"epoch": 3.3684210526315788,
|
| 502 |
+
"grad_norm": 3.7174108028411865,
|
| 503 |
+
"learning_rate": 1.2500000000000006e-05,
|
| 504 |
+
"loss": 0.2785,
|
| 505 |
+
"step": 640
|
| 506 |
+
},
|
| 507 |
+
{
|
| 508 |
+
"epoch": 3.4210526315789473,
|
| 509 |
+
"grad_norm": 4.665218353271484,
|
| 510 |
+
"learning_rate": 1.1775899741825947e-05,
|
| 511 |
+
"loss": 0.3323,
|
| 512 |
+
"step": 650
|
| 513 |
+
},
|
| 514 |
+
{
|
| 515 |
+
"epoch": 3.473684210526316,
|
| 516 |
+
"grad_norm": 6.711315631866455,
|
| 517 |
+
"learning_rate": 1.1066888421939093e-05,
|
| 518 |
+
"loss": 0.2762,
|
| 519 |
+
"step": 660
|
| 520 |
+
},
|
| 521 |
+
{
|
| 522 |
+
"epoch": 3.526315789473684,
|
| 523 |
+
"grad_norm": 4.101406097412109,
|
| 524 |
+
"learning_rate": 1.0373775035117305e-05,
|
| 525 |
+
"loss": 0.2982,
|
| 526 |
+
"step": 670
|
| 527 |
+
},
|
| 528 |
+
{
|
| 529 |
+
"epoch": 3.5789473684210527,
|
| 530 |
+
"grad_norm": 3.3571157455444336,
|
| 531 |
+
"learning_rate": 9.697350436308427e-06,
|
| 532 |
+
"loss": 0.2338,
|
| 533 |
+
"step": 680
|
| 534 |
+
},
|
| 535 |
+
{
|
| 536 |
+
"epoch": 3.6315789473684212,
|
| 537 |
+
"grad_norm": 7.152629852294922,
|
| 538 |
+
"learning_rate": 9.038386438250415e-06,
|
| 539 |
+
"loss": 0.2962,
|
| 540 |
+
"step": 690
|
| 541 |
+
},
|
| 542 |
+
{
|
| 543 |
+
"epoch": 3.6842105263157894,
|
| 544 |
+
"grad_norm": 5.147871971130371,
|
| 545 |
+
"learning_rate": 8.397634930819021e-06,
|
| 546 |
+
"loss": 0.2846,
|
| 547 |
+
"step": 700
|
| 548 |
+
},
|
| 549 |
+
{
|
| 550 |
+
"epoch": 3.6842105263157894,
|
| 551 |
+
"eval_loss": 0.6083844900131226,
|
| 552 |
+
"eval_runtime": 1.9199,
|
| 553 |
+
"eval_samples_per_second": 41.67,
|
| 554 |
+
"eval_steps_per_second": 5.209,
|
| 555 |
+
"step": 700
|
| 556 |
+
},
|
| 557 |
+
{
|
| 558 |
+
"epoch": 3.736842105263158,
|
| 559 |
+
"grad_norm": 3.984264373779297,
|
| 560 |
+
"learning_rate": 7.775827023107835e-06,
|
| 561 |
+
"loss": 0.2895,
|
| 562 |
+
"step": 710
|
| 563 |
+
},
|
| 564 |
+
{
|
| 565 |
+
"epoch": 3.7894736842105265,
|
| 566 |
+
"grad_norm": 6.230710983276367,
|
| 567 |
+
"learning_rate": 7.173672209219495e-06,
|
| 568 |
+
"loss": 0.3261,
|
| 569 |
+
"step": 720
|
| 570 |
+
},
|
| 571 |
+
{
|
| 572 |
+
"epoch": 3.8421052631578947,
|
| 573 |
+
"grad_norm": 3.685063362121582,
|
| 574 |
+
"learning_rate": 6.591857558720071e-06,
|
| 575 |
+
"loss": 0.2358,
|
| 576 |
+
"step": 730
|
| 577 |
+
},
|
| 578 |
+
{
|
| 579 |
+
"epoch": 3.8947368421052633,
|
| 580 |
+
"grad_norm": 4.337435245513916,
|
| 581 |
+
"learning_rate": 6.031046932680229e-06,
|
| 582 |
+
"loss": 0.2723,
|
| 583 |
+
"step": 740
|
| 584 |
+
},
|
| 585 |
+
{
|
| 586 |
+
"epoch": 3.9473684210526314,
|
| 587 |
+
"grad_norm": 4.504445552825928,
|
| 588 |
+
"learning_rate": 5.491880226197707e-06,
|
| 589 |
+
"loss": 0.2941,
|
| 590 |
+
"step": 750
|
| 591 |
+
},
|
| 592 |
+
{
|
| 593 |
+
"epoch": 4.0,
|
| 594 |
+
"grad_norm": 4.7959442138671875,
|
| 595 |
+
"learning_rate": 4.9749726382653905e-06,
|
| 596 |
+
"loss": 0.2721,
|
| 597 |
+
"step": 760
|
| 598 |
+
},
|
| 599 |
+
{
|
| 600 |
+
"epoch": 4.052631578947368,
|
| 601 |
+
"grad_norm": 2.663322925567627,
|
| 602 |
+
"learning_rate": 4.480913969818098e-06,
|
| 603 |
+
"loss": 0.1677,
|
| 604 |
+
"step": 770
|
| 605 |
+
},
|
| 606 |
+
{
|
| 607 |
+
"epoch": 4.105263157894737,
|
| 608 |
+
"grad_norm": 5.704188346862793,
|
| 609 |
+
"learning_rate": 4.010267950759025e-06,
|
| 610 |
+
"loss": 0.2291,
|
| 611 |
+
"step": 780
|
| 612 |
+
},
|
| 613 |
+
{
|
| 614 |
+
"epoch": 4.157894736842105,
|
| 615 |
+
"grad_norm": 4.857370853424072,
|
| 616 |
+
"learning_rate": 3.5635715967337223e-06,
|
| 617 |
+
"loss": 0.1991,
|
| 618 |
+
"step": 790
|
| 619 |
+
},
|
| 620 |
+
{
|
| 621 |
+
"epoch": 4.2105263157894735,
|
| 622 |
+
"grad_norm": 2.6290528774261475,
|
| 623 |
+
"learning_rate": 3.141334596385448e-06,
|
| 624 |
+
"loss": 0.2069,
|
| 625 |
+
"step": 800
|
| 626 |
+
},
|
| 627 |
+
{
|
| 628 |
+
"epoch": 4.2105263157894735,
|
| 629 |
+
"eval_loss": 0.6427180767059326,
|
| 630 |
+
"eval_runtime": 1.9195,
|
| 631 |
+
"eval_samples_per_second": 41.677,
|
| 632 |
+
"eval_steps_per_second": 5.21,
|
| 633 |
+
"step": 800
|
| 634 |
+
},
|
| 635 |
+
{
|
| 636 |
+
"epoch": 4.2631578947368425,
|
| 637 |
+
"grad_norm": 6.7939558029174805,
|
| 638 |
+
"learning_rate": 2.7440387297912123e-06,
|
| 639 |
+
"loss": 0.2213,
|
| 640 |
+
"step": 810
|
| 641 |
+
},
|
| 642 |
+
{
|
| 643 |
+
"epoch": 4.315789473684211,
|
| 644 |
+
"grad_norm": 5.425328731536865,
|
| 645 |
+
"learning_rate": 2.372137318741968e-06,
|
| 646 |
+
"loss": 0.2008,
|
| 647 |
+
"step": 820
|
| 648 |
+
},
|
| 649 |
+
{
|
| 650 |
+
"epoch": 4.368421052631579,
|
| 651 |
+
"grad_norm": 3.0159809589385986,
|
| 652 |
+
"learning_rate": 2.026054709494235e-06,
|
| 653 |
+
"loss": 0.2178,
|
| 654 |
+
"step": 830
|
| 655 |
+
},
|
| 656 |
+
{
|
| 657 |
+
"epoch": 4.421052631578947,
|
| 658 |
+
"grad_norm": 4.54276704788208,
|
| 659 |
+
"learning_rate": 1.7061857885832893e-06,
|
| 660 |
+
"loss": 0.1878,
|
| 661 |
+
"step": 840
|
| 662 |
+
},
|
| 663 |
+
{
|
| 664 |
+
"epoch": 4.473684210526316,
|
| 665 |
+
"grad_norm": 4.1157755851745605,
|
| 666 |
+
"learning_rate": 1.4128955322504966e-06,
|
| 667 |
+
"loss": 0.1733,
|
| 668 |
+
"step": 850
|
| 669 |
+
},
|
| 670 |
+
{
|
| 671 |
+
"epoch": 4.526315789473684,
|
| 672 |
+
"grad_norm": 4.860106945037842,
|
| 673 |
+
"learning_rate": 1.1465185899987797e-06,
|
| 674 |
+
"loss": 0.193,
|
| 675 |
+
"step": 860
|
| 676 |
+
},
|
| 677 |
+
{
|
| 678 |
+
"epoch": 4.578947368421053,
|
| 679 |
+
"grad_norm": 4.945047378540039,
|
| 680 |
+
"learning_rate": 9.073589027514789e-07,
|
| 681 |
+
"loss": 0.1802,
|
| 682 |
+
"step": 870
|
| 683 |
+
},
|
| 684 |
+
{
|
| 685 |
+
"epoch": 4.631578947368421,
|
| 686 |
+
"grad_norm": 2.316741943359375,
|
| 687 |
+
"learning_rate": 6.956893560502359e-07,
|
| 688 |
+
"loss": 0.1736,
|
| 689 |
+
"step": 880
|
| 690 |
+
},
|
| 691 |
+
{
|
| 692 |
+
"epoch": 4.684210526315789,
|
| 693 |
+
"grad_norm": 4.012813091278076,
|
| 694 |
+
"learning_rate": 5.117514686876379e-07,
|
| 695 |
+
"loss": 0.1761,
|
| 696 |
+
"step": 890
|
| 697 |
+
},
|
| 698 |
+
{
|
| 699 |
+
"epoch": 4.7368421052631575,
|
| 700 |
+
"grad_norm": 5.301681995391846,
|
| 701 |
+
"learning_rate": 3.557551171299051e-07,
|
| 702 |
+
"loss": 0.172,
|
| 703 |
+
"step": 900
|
| 704 |
+
},
|
| 705 |
+
{
|
| 706 |
+
"epoch": 4.7368421052631575,
|
| 707 |
+
"eval_loss": 0.6494551301002502,
|
| 708 |
+
"eval_runtime": 1.9201,
|
| 709 |
+
"eval_samples_per_second": 41.665,
|
| 710 |
+
"eval_steps_per_second": 5.208,
|
| 711 |
+
"step": 900
|
| 712 |
+
},
|
| 713 |
+
{
|
| 714 |
+
"epoch": 4.7894736842105265,
|
| 715 |
+
"grad_norm": 3.559140205383301,
|
| 716 |
+
"learning_rate": 2.27878296044029e-07,
|
| 717 |
+
"loss": 0.1734,
|
| 718 |
+
"step": 910
|
| 719 |
+
},
|
| 720 |
+
{
|
| 721 |
+
"epoch": 4.842105263157895,
|
| 722 |
+
"grad_norm": 7.743849277496338,
|
| 723 |
+
"learning_rate": 1.2826691520262114e-07,
|
| 724 |
+
"loss": 0.1954,
|
| 725 |
+
"step": 920
|
| 726 |
+
},
|
| 727 |
+
{
|
| 728 |
+
"epoch": 4.894736842105263,
|
| 729 |
+
"grad_norm": 3.5408854484558105,
|
| 730 |
+
"learning_rate": 5.7034632998231865e-08,
|
| 731 |
+
"loss": 0.1744,
|
| 732 |
+
"step": 930
|
| 733 |
+
},
|
| 734 |
+
{
|
| 735 |
+
"epoch": 4.947368421052632,
|
| 736 |
+
"grad_norm": 2.413121461868286,
|
| 737 |
+
"learning_rate": 1.4262726757049982e-08,
|
| 738 |
+
"loss": 0.1778,
|
| 739 |
+
"step": 940
|
| 740 |
+
},
|
| 741 |
+
{
|
| 742 |
+
"epoch": 5.0,
|
| 743 |
+
"grad_norm": 2.56962513923645,
|
| 744 |
+
"learning_rate": 0.0,
|
| 745 |
+
"loss": 0.1836,
|
| 746 |
+
"step": 950
|
| 747 |
+
},
|
| 748 |
+
{
|
| 749 |
+
"epoch": 5.0,
|
| 750 |
+
"step": 950,
|
| 751 |
+
"total_flos": 2.0275085174217114e+17,
|
| 752 |
+
"train_loss": 0.5822552880487945,
|
| 753 |
+
"train_runtime": 660.0352,
|
| 754 |
+
"train_samples_per_second": 11.515,
|
| 755 |
+
"train_steps_per_second": 1.439
|
| 756 |
+
}
|
| 757 |
+
],
|
| 758 |
+
"logging_steps": 10,
|
| 759 |
+
"max_steps": 950,
|
| 760 |
+
"num_input_tokens_seen": 0,
|
| 761 |
+
"num_train_epochs": 5,
|
| 762 |
+
"save_steps": 100,
|
| 763 |
+
"total_flos": 2.0275085174217114e+17,
|
| 764 |
+
"train_batch_size": 4,
|
| 765 |
+
"trial_name": null,
|
| 766 |
+
"trial_params": null
|
| 767 |
+
}
|
llama2_13b_peft/linguistics_puzzles/training_args.bin
ADDED
|
@@ -0,0 +1,3 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
version https://git-lfs.github.com/spec/v1
|
| 2 |
+
oid sha256:67ac58d8b967dcc701c74de72e5e18349db160299022d297808b6aa2f75860a0
|
| 3 |
+
size 5176
|
llama2_13b_peft/linguistics_puzzles/training_eval_loss.png
ADDED
|
llama2_13b_peft/linguistics_puzzles/training_loss.png
ADDED
|
llama2_13b_peft/news_commentary_de/README.md
ADDED
|
@@ -0,0 +1,85 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
---
|
| 2 |
+
license: other
|
| 3 |
+
library_name: peft
|
| 4 |
+
tags:
|
| 5 |
+
- llama-factory
|
| 6 |
+
- lora
|
| 7 |
+
- generated_from_trainer
|
| 8 |
+
base_model: /data1/model/llama2/meta-llama/Llama2-13b
|
| 9 |
+
model-index:
|
| 10 |
+
- name: news_commentary_de_no_sys
|
| 11 |
+
results: []
|
| 12 |
+
---
|
| 13 |
+
|
| 14 |
+
<!-- This model card has been generated automatically according to the information the Trainer had access to. You
|
| 15 |
+
should probably proofread and complete it, then remove this comment. -->
|
| 16 |
+
|
| 17 |
+
# news_commentary_de_no_sys
|
| 18 |
+
|
| 19 |
+
This model is a fine-tuned version of [/data1/model/llama2/meta-llama/Llama2-13b](https://huggingface.co//data1/model/llama2/meta-llama/Llama2-13b) on the news_commentary_de_no_sys dataset.
|
| 20 |
+
It achieves the following results on the evaluation set:
|
| 21 |
+
- Loss: 0.6944
|
| 22 |
+
|
| 23 |
+
## Model description
|
| 24 |
+
|
| 25 |
+
More information needed
|
| 26 |
+
|
| 27 |
+
## Intended uses & limitations
|
| 28 |
+
|
| 29 |
+
More information needed
|
| 30 |
+
|
| 31 |
+
## Training and evaluation data
|
| 32 |
+
|
| 33 |
+
More information needed
|
| 34 |
+
|
| 35 |
+
## Training procedure
|
| 36 |
+
|
| 37 |
+
### Training hyperparameters
|
| 38 |
+
|
| 39 |
+
The following hyperparameters were used during training:
|
| 40 |
+
- learning_rate: 1e-05
|
| 41 |
+
- train_batch_size: 8
|
| 42 |
+
- eval_batch_size: 8
|
| 43 |
+
- seed: 42
|
| 44 |
+
- distributed_type: multi-GPU
|
| 45 |
+
- num_devices: 2
|
| 46 |
+
- total_train_batch_size: 16
|
| 47 |
+
- total_eval_batch_size: 16
|
| 48 |
+
- optimizer: Adam with betas=(0.9,0.999) and epsilon=1e-08
|
| 49 |
+
- lr_scheduler_type: cosine
|
| 50 |
+
- lr_scheduler_warmup_steps: 20
|
| 51 |
+
- num_epochs: 10.0
|
| 52 |
+
|
| 53 |
+
### Training results
|
| 54 |
+
|
| 55 |
+
| Training Loss | Epoch | Step | Validation Loss |
|
| 56 |
+
|:-------------:|:-----:|:----:|:---------------:|
|
| 57 |
+
| 0.7429 | 0.13 | 200 | 0.7712 |
|
| 58 |
+
| 0.7549 | 0.25 | 400 | 0.7434 |
|
| 59 |
+
| 0.7552 | 0.38 | 600 | 0.7330 |
|
| 60 |
+
| 0.7265 | 0.5 | 800 | 0.7256 |
|
| 61 |
+
| 0.7524 | 0.63 | 1000 | 0.7200 |
|
| 62 |
+
| 0.6976 | 0.75 | 1200 | 0.7151 |
|
| 63 |
+
| 0.7408 | 0.88 | 1400 | 0.7116 |
|
| 64 |
+
| 0.701 | 1.0 | 1600 | 0.7085 |
|
| 65 |
+
| 0.7084 | 1.13 | 1800 | 0.7059 |
|
| 66 |
+
| 0.6999 | 1.25 | 2000 | 0.7040 |
|
| 67 |
+
| 0.7182 | 1.38 | 2200 | 0.7022 |
|
| 68 |
+
| 0.7267 | 1.51 | 2400 | 0.6994 |
|
| 69 |
+
| 0.6912 | 1.63 | 2600 | 0.6972 |
|
| 70 |
+
| 0.6821 | 1.76 | 2800 | 0.6954 |
|
| 71 |
+
| 0.7104 | 1.88 | 3000 | 0.6944 |
|
| 72 |
+
| 0.6222 | 2.01 | 3200 | 0.6934 |
|
| 73 |
+
| 0.6383 | 2.13 | 3400 | 0.6974 |
|
| 74 |
+
| 0.6436 | 2.26 | 3600 | 0.6981 |
|
| 75 |
+
| 0.6444 | 2.38 | 3800 | 0.6968 |
|
| 76 |
+
| 0.6368 | 2.51 | 4000 | 0.6987 |
|
| 77 |
+
|
| 78 |
+
|
| 79 |
+
### Framework versions
|
| 80 |
+
|
| 81 |
+
- PEFT 0.9.0
|
| 82 |
+
- Transformers 4.38.2
|
| 83 |
+
- Pytorch 2.2.1
|
| 84 |
+
- Datasets 2.18.0
|
| 85 |
+
- Tokenizers 0.15.2
|
llama2_13b_peft/news_commentary_de/adapter_config.json
ADDED
|
@@ -0,0 +1,33 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
{
|
| 2 |
+
"alpha_pattern": {},
|
| 3 |
+
"auto_mapping": null,
|
| 4 |
+
"base_model_name_or_path": "/data1/model/llama2/meta-llama/Llama2-13b",
|
| 5 |
+
"bias": "none",
|
| 6 |
+
"fan_in_fan_out": false,
|
| 7 |
+
"inference_mode": true,
|
| 8 |
+
"init_lora_weights": true,
|
| 9 |
+
"layers_pattern": null,
|
| 10 |
+
"layers_to_transform": null,
|
| 11 |
+
"loftq_config": {},
|
| 12 |
+
"lora_alpha": 16,
|
| 13 |
+
"lora_dropout": 0.0,
|
| 14 |
+
"megatron_config": null,
|
| 15 |
+
"megatron_core": "megatron.core",
|
| 16 |
+
"modules_to_save": null,
|
| 17 |
+
"peft_type": "LORA",
|
| 18 |
+
"r": 8,
|
| 19 |
+
"rank_pattern": {},
|
| 20 |
+
"revision": null,
|
| 21 |
+
"target_modules": [
|
| 22 |
+
"up_proj",
|
| 23 |
+
"down_proj",
|
| 24 |
+
"v_proj",
|
| 25 |
+
"k_proj",
|
| 26 |
+
"o_proj",
|
| 27 |
+
"gate_proj",
|
| 28 |
+
"q_proj"
|
| 29 |
+
],
|
| 30 |
+
"task_type": "CAUSAL_LM",
|
| 31 |
+
"use_dora": false,
|
| 32 |
+
"use_rslora": false
|
| 33 |
+
}
|
llama2_13b_peft/news_commentary_de/adapter_model.safetensors
ADDED
|
@@ -0,0 +1,3 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
version https://git-lfs.github.com/spec/v1
|
| 2 |
+
oid sha256:44a2422055e9542643288a7443b823001443ae5a402e2cff85e691f7121a6398
|
| 3 |
+
size 125248064
|
llama2_13b_peft/news_commentary_de/all_results.json
ADDED
|
@@ -0,0 +1,11 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
{
|
| 2 |
+
"epoch": 2.51,
|
| 3 |
+
"eval_loss": 0.6943792104721069,
|
| 4 |
+
"eval_runtime": 64.8294,
|
| 5 |
+
"eval_samples_per_second": 69.413,
|
| 6 |
+
"eval_steps_per_second": 4.35,
|
| 7 |
+
"train_loss": 0.7081527805328369,
|
| 8 |
+
"train_runtime": 4312.5386,
|
| 9 |
+
"train_samples_per_second": 59.13,
|
| 10 |
+
"train_steps_per_second": 3.696
|
| 11 |
+
}
|
llama2_13b_peft/news_commentary_de/eval_results.json
ADDED
|
@@ -0,0 +1,7 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
{
|
| 2 |
+
"epoch": 2.51,
|
| 3 |
+
"eval_loss": 0.6943792104721069,
|
| 4 |
+
"eval_runtime": 64.8294,
|
| 5 |
+
"eval_samples_per_second": 69.413,
|
| 6 |
+
"eval_steps_per_second": 4.35
|
| 7 |
+
}
|
llama2_13b_peft/news_commentary_de/special_tokens_map.json
ADDED
|
@@ -0,0 +1,24 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
{
|
| 2 |
+
"bos_token": {
|
| 3 |
+
"content": "<s>",
|
| 4 |
+
"lstrip": false,
|
| 5 |
+
"normalized": false,
|
| 6 |
+
"rstrip": false,
|
| 7 |
+
"single_word": false
|
| 8 |
+
},
|
| 9 |
+
"eos_token": {
|
| 10 |
+
"content": "</s>",
|
| 11 |
+
"lstrip": false,
|
| 12 |
+
"normalized": false,
|
| 13 |
+
"rstrip": false,
|
| 14 |
+
"single_word": false
|
| 15 |
+
},
|
| 16 |
+
"pad_token": "</s>",
|
| 17 |
+
"unk_token": {
|
| 18 |
+
"content": "<unk>",
|
| 19 |
+
"lstrip": false,
|
| 20 |
+
"normalized": false,
|
| 21 |
+
"rstrip": false,
|
| 22 |
+
"single_word": false
|
| 23 |
+
}
|
| 24 |
+
}
|
llama2_13b_peft/news_commentary_de/tokenizer.model
ADDED
|
@@ -0,0 +1,3 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
version https://git-lfs.github.com/spec/v1
|
| 2 |
+
oid sha256:9e556afd44213b6bd1be2b850ebbbd98f5481437a8021afaf58ee7fb1818d347
|
| 3 |
+
size 499723
|
llama2_13b_peft/news_commentary_de/tokenizer_config.json
ADDED
|
@@ -0,0 +1,45 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
{
|
| 2 |
+
"add_bos_token": true,
|
| 3 |
+
"add_eos_token": false,
|
| 4 |
+
"add_prefix_space": true,
|
| 5 |
+
"added_tokens_decoder": {
|
| 6 |
+
"0": {
|
| 7 |
+
"content": "<unk>",
|
| 8 |
+
"lstrip": false,
|
| 9 |
+
"normalized": false,
|
| 10 |
+
"rstrip": false,
|
| 11 |
+
"single_word": false,
|
| 12 |
+
"special": true
|
| 13 |
+
},
|
| 14 |
+
"1": {
|
| 15 |
+
"content": "<s>",
|
| 16 |
+
"lstrip": false,
|
| 17 |
+
"normalized": false,
|
| 18 |
+
"rstrip": false,
|
| 19 |
+
"single_word": false,
|
| 20 |
+
"special": true
|
| 21 |
+
},
|
| 22 |
+
"2": {
|
| 23 |
+
"content": "</s>",
|
| 24 |
+
"lstrip": false,
|
| 25 |
+
"normalized": false,
|
| 26 |
+
"rstrip": false,
|
| 27 |
+
"single_word": false,
|
| 28 |
+
"special": true
|
| 29 |
+
}
|
| 30 |
+
},
|
| 31 |
+
"bos_token": "<s>",
|
| 32 |
+
"chat_template": "{% if messages[0]['role'] == 'system' %}{% set system_message = messages[0]['content'] %}{% endif %}{% if system_message is defined %}{{ system_message }}{% endif %}{% for message in messages %}{% set content = message['content'] %}{% if message['role'] == 'user' %}{{ content }}{% elif message['role'] == 'assistant' %}{{ content + '\\n' }}{% endif %}{% endfor %}",
|
| 33 |
+
"clean_up_tokenization_spaces": false,
|
| 34 |
+
"eos_token": "</s>",
|
| 35 |
+
"legacy": true,
|
| 36 |
+
"model_max_length": 1000000000000000019884624838656,
|
| 37 |
+
"pad_token": "</s>",
|
| 38 |
+
"padding_side": "right",
|
| 39 |
+
"sp_model_kwargs": {},
|
| 40 |
+
"spaces_between_special_tokens": false,
|
| 41 |
+
"split_special_tokens": false,
|
| 42 |
+
"tokenizer_class": "LlamaTokenizer",
|
| 43 |
+
"unk_token": "<unk>",
|
| 44 |
+
"use_default_system_prompt": false
|
| 45 |
+
}
|
llama2_13b_peft/news_commentary_de/train_results.json
ADDED
|
@@ -0,0 +1,7 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
{
|
| 2 |
+
"epoch": 2.51,
|
| 3 |
+
"train_loss": 0.7081527805328369,
|
| 4 |
+
"train_runtime": 4312.5386,
|
| 5 |
+
"train_samples_per_second": 59.13,
|
| 6 |
+
"train_steps_per_second": 3.696
|
| 7 |
+
}
|
llama2_13b_peft/news_commentary_de/trainer_log.jsonl
ADDED
|
The diff for this file is too large to render.
See raw diff
|
|
|
llama2_13b_peft/news_commentary_de/trainer_state.json
ADDED
|
@@ -0,0 +1,2990 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
{
|
| 2 |
+
"best_metric": 0.6943792104721069,
|
| 3 |
+
"best_model_checkpoint": "ckpt/llama2_13b_fuze30_no_sys/news_commentary_de_no_sys/checkpoint-3000",
|
| 4 |
+
"epoch": 2.509410288582183,
|
| 5 |
+
"eval_steps": 200,
|
| 6 |
+
"global_step": 4000,
|
| 7 |
+
"is_hyper_param_search": false,
|
| 8 |
+
"is_local_process_zero": true,
|
| 9 |
+
"is_world_process_zero": true,
|
| 10 |
+
"log_history": [
|
| 11 |
+
{
|
| 12 |
+
"epoch": 0.01,
|
| 13 |
+
"grad_norm": 0.5409977436065674,
|
| 14 |
+
"learning_rate": 5e-06,
|
| 15 |
+
"loss": 1.3994,
|
| 16 |
+
"step": 10
|
| 17 |
+
},
|
| 18 |
+
{
|
| 19 |
+
"epoch": 0.01,
|
| 20 |
+
"grad_norm": 0.850004255771637,
|
| 21 |
+
"learning_rate": 1e-05,
|
| 22 |
+
"loss": 1.4561,
|
| 23 |
+
"step": 20
|
| 24 |
+
},
|
| 25 |
+
{
|
| 26 |
+
"epoch": 0.02,
|
| 27 |
+
"grad_norm": 0.8501812219619751,
|
| 28 |
+
"learning_rate": 9.999990264607035e-06,
|
| 29 |
+
"loss": 1.3697,
|
| 30 |
+
"step": 30
|
| 31 |
+
},
|
| 32 |
+
{
|
| 33 |
+
"epoch": 0.03,
|
| 34 |
+
"grad_norm": 0.6338475346565247,
|
| 35 |
+
"learning_rate": 9.999961058466052e-06,
|
| 36 |
+
"loss": 1.3627,
|
| 37 |
+
"step": 40
|
| 38 |
+
},
|
| 39 |
+
{
|
| 40 |
+
"epoch": 0.03,
|
| 41 |
+
"grad_norm": 0.7430967688560486,
|
| 42 |
+
"learning_rate": 9.999912381690781e-06,
|
| 43 |
+
"loss": 1.1155,
|
| 44 |
+
"step": 50
|
| 45 |
+
},
|
| 46 |
+
{
|
| 47 |
+
"epoch": 0.04,
|
| 48 |
+
"grad_norm": 0.5487976670265198,
|
| 49 |
+
"learning_rate": 9.999844234470782e-06,
|
| 50 |
+
"loss": 0.9492,
|
| 51 |
+
"step": 60
|
| 52 |
+
},
|
| 53 |
+
{
|
| 54 |
+
"epoch": 0.04,
|
| 55 |
+
"grad_norm": 0.3653506934642792,
|
| 56 |
+
"learning_rate": 9.999756617071427e-06,
|
| 57 |
+
"loss": 0.9067,
|
| 58 |
+
"step": 70
|
| 59 |
+
},
|
| 60 |
+
{
|
| 61 |
+
"epoch": 0.05,
|
| 62 |
+
"grad_norm": 0.38920339941978455,
|
| 63 |
+
"learning_rate": 9.999649529833915e-06,
|
| 64 |
+
"loss": 0.8848,
|
| 65 |
+
"step": 80
|
| 66 |
+
},
|
| 67 |
+
{
|
| 68 |
+
"epoch": 0.06,
|
| 69 |
+
"grad_norm": 0.4155251979827881,
|
| 70 |
+
"learning_rate": 9.999522973175257e-06,
|
| 71 |
+
"loss": 0.798,
|
| 72 |
+
"step": 90
|
| 73 |
+
},
|
| 74 |
+
{
|
| 75 |
+
"epoch": 0.06,
|
| 76 |
+
"grad_norm": 0.4156494438648224,
|
| 77 |
+
"learning_rate": 9.999376947588288e-06,
|
| 78 |
+
"loss": 0.8782,
|
| 79 |
+
"step": 100
|
| 80 |
+
},
|
| 81 |
+
{
|
| 82 |
+
"epoch": 0.07,
|
| 83 |
+
"grad_norm": 0.4306489825248718,
|
| 84 |
+
"learning_rate": 9.99921145364165e-06,
|
| 85 |
+
"loss": 0.8124,
|
| 86 |
+
"step": 110
|
| 87 |
+
},
|
| 88 |
+
{
|
| 89 |
+
"epoch": 0.08,
|
| 90 |
+
"grad_norm": 0.39355072379112244,
|
| 91 |
+
"learning_rate": 9.999026491979809e-06,
|
| 92 |
+
"loss": 0.838,
|
| 93 |
+
"step": 120
|
| 94 |
+
},
|
| 95 |
+
{
|
| 96 |
+
"epoch": 0.08,
|
| 97 |
+
"grad_norm": 0.4246688783168793,
|
| 98 |
+
"learning_rate": 9.99882206332303e-06,
|
| 99 |
+
"loss": 0.8383,
|
| 100 |
+
"step": 130
|
| 101 |
+
},
|
| 102 |
+
{
|
| 103 |
+
"epoch": 0.09,
|
| 104 |
+
"grad_norm": 0.47585156559944153,
|
| 105 |
+
"learning_rate": 9.99859816846739e-06,
|
| 106 |
+
"loss": 0.8705,
|
| 107 |
+
"step": 140
|
| 108 |
+
},
|
| 109 |
+
{
|
| 110 |
+
"epoch": 0.09,
|
| 111 |
+
"grad_norm": 0.48569419980049133,
|
| 112 |
+
"learning_rate": 9.998354808284774e-06,
|
| 113 |
+
"loss": 0.7872,
|
| 114 |
+
"step": 150
|
| 115 |
+
},
|
| 116 |
+
{
|
| 117 |
+
"epoch": 0.1,
|
| 118 |
+
"grad_norm": 0.5107733011245728,
|
| 119 |
+
"learning_rate": 9.998091983722862e-06,
|
| 120 |
+
"loss": 0.789,
|
| 121 |
+
"step": 160
|
| 122 |
+
},
|
| 123 |
+
{
|
| 124 |
+
"epoch": 0.11,
|
| 125 |
+
"grad_norm": 0.5669977068901062,
|
| 126 |
+
"learning_rate": 9.997809695805136e-06,
|
| 127 |
+
"loss": 0.7749,
|
| 128 |
+
"step": 170
|
| 129 |
+
},
|
| 130 |
+
{
|
| 131 |
+
"epoch": 0.11,
|
| 132 |
+
"grad_norm": 0.49600809812545776,
|
| 133 |
+
"learning_rate": 9.99750794563087e-06,
|
| 134 |
+
"loss": 0.7935,
|
| 135 |
+
"step": 180
|
| 136 |
+
},
|
| 137 |
+
{
|
| 138 |
+
"epoch": 0.12,
|
| 139 |
+
"grad_norm": 0.45251163840293884,
|
| 140 |
+
"learning_rate": 9.997186734375124e-06,
|
| 141 |
+
"loss": 0.7817,
|
| 142 |
+
"step": 190
|
| 143 |
+
},
|
| 144 |
+
{
|
| 145 |
+
"epoch": 0.13,
|
| 146 |
+
"grad_norm": 0.46742165088653564,
|
| 147 |
+
"learning_rate": 9.996846063288746e-06,
|
| 148 |
+
"loss": 0.7429,
|
| 149 |
+
"step": 200
|
| 150 |
+
},
|
| 151 |
+
{
|
| 152 |
+
"epoch": 0.13,
|
| 153 |
+
"eval_loss": 0.7712445855140686,
|
| 154 |
+
"eval_runtime": 64.6782,
|
| 155 |
+
"eval_samples_per_second": 69.575,
|
| 156 |
+
"eval_steps_per_second": 4.36,
|
| 157 |
+
"step": 200
|
| 158 |
+
},
|
| 159 |
+
{
|
| 160 |
+
"epoch": 0.13,
|
| 161 |
+
"grad_norm": 0.5643576383590698,
|
| 162 |
+
"learning_rate": 9.996485933698364e-06,
|
| 163 |
+
"loss": 0.7636,
|
| 164 |
+
"step": 210
|
| 165 |
+
},
|
| 166 |
+
{
|
| 167 |
+
"epoch": 0.14,
|
| 168 |
+
"grad_norm": 0.4915783405303955,
|
| 169 |
+
"learning_rate": 9.996106347006378e-06,
|
| 170 |
+
"loss": 0.7856,
|
| 171 |
+
"step": 220
|
| 172 |
+
},
|
| 173 |
+
{
|
| 174 |
+
"epoch": 0.14,
|
| 175 |
+
"grad_norm": 0.3926757574081421,
|
| 176 |
+
"learning_rate": 9.99570730469096e-06,
|
| 177 |
+
"loss": 0.7529,
|
| 178 |
+
"step": 230
|
| 179 |
+
},
|
| 180 |
+
{
|
| 181 |
+
"epoch": 0.15,
|
| 182 |
+
"grad_norm": 0.3297576606273651,
|
| 183 |
+
"learning_rate": 9.995288808306041e-06,
|
| 184 |
+
"loss": 0.7671,
|
| 185 |
+
"step": 240
|
| 186 |
+
},
|
| 187 |
+
{
|
| 188 |
+
"epoch": 0.16,
|
| 189 |
+
"grad_norm": 0.45379459857940674,
|
| 190 |
+
"learning_rate": 9.994850859481312e-06,
|
| 191 |
+
"loss": 0.7231,
|
| 192 |
+
"step": 250
|
| 193 |
+
},
|
| 194 |
+
{
|
| 195 |
+
"epoch": 0.16,
|
| 196 |
+
"grad_norm": 0.5688673853874207,
|
| 197 |
+
"learning_rate": 9.994393459922219e-06,
|
| 198 |
+
"loss": 0.7694,
|
| 199 |
+
"step": 260
|
| 200 |
+
},
|
| 201 |
+
{
|
| 202 |
+
"epoch": 0.17,
|
| 203 |
+
"grad_norm": 0.6590914130210876,
|
| 204 |
+
"learning_rate": 9.993916611409941e-06,
|
| 205 |
+
"loss": 0.7661,
|
| 206 |
+
"step": 270
|
| 207 |
+
},
|
| 208 |
+
{
|
| 209 |
+
"epoch": 0.18,
|
| 210 |
+
"grad_norm": 0.4207383692264557,
|
| 211 |
+
"learning_rate": 9.993420315801406e-06,
|
| 212 |
+
"loss": 0.7952,
|
| 213 |
+
"step": 280
|
| 214 |
+
},
|
| 215 |
+
{
|
| 216 |
+
"epoch": 0.18,
|
| 217 |
+
"grad_norm": 0.47460174560546875,
|
| 218 |
+
"learning_rate": 9.992904575029265e-06,
|
| 219 |
+
"loss": 0.7966,
|
| 220 |
+
"step": 290
|
| 221 |
+
},
|
| 222 |
+
{
|
| 223 |
+
"epoch": 0.19,
|
| 224 |
+
"grad_norm": 0.6118924617767334,
|
| 225 |
+
"learning_rate": 9.992369391101895e-06,
|
| 226 |
+
"loss": 0.8167,
|
| 227 |
+
"step": 300
|
| 228 |
+
},
|
| 229 |
+
{
|
| 230 |
+
"epoch": 0.19,
|
| 231 |
+
"grad_norm": 0.44934767484664917,
|
| 232 |
+
"learning_rate": 9.991814766103386e-06,
|
| 233 |
+
"loss": 0.7368,
|
| 234 |
+
"step": 310
|
| 235 |
+
},
|
| 236 |
+
{
|
| 237 |
+
"epoch": 0.2,
|
| 238 |
+
"grad_norm": 0.5106733441352844,
|
| 239 |
+
"learning_rate": 9.991240702193532e-06,
|
| 240 |
+
"loss": 0.7796,
|
| 241 |
+
"step": 320
|
| 242 |
+
},
|
| 243 |
+
{
|
| 244 |
+
"epoch": 0.21,
|
| 245 |
+
"grad_norm": 0.4405980706214905,
|
| 246 |
+
"learning_rate": 9.99064720160783e-06,
|
| 247 |
+
"loss": 0.7727,
|
| 248 |
+
"step": 330
|
| 249 |
+
},
|
| 250 |
+
{
|
| 251 |
+
"epoch": 0.21,
|
| 252 |
+
"grad_norm": 0.6010485887527466,
|
| 253 |
+
"learning_rate": 9.990034266657468e-06,
|
| 254 |
+
"loss": 0.7604,
|
| 255 |
+
"step": 340
|
| 256 |
+
},
|
| 257 |
+
{
|
| 258 |
+
"epoch": 0.22,
|
| 259 |
+
"grad_norm": 0.6098916530609131,
|
| 260 |
+
"learning_rate": 9.989401899729307e-06,
|
| 261 |
+
"loss": 0.7399,
|
| 262 |
+
"step": 350
|
| 263 |
+
},
|
| 264 |
+
{
|
| 265 |
+
"epoch": 0.23,
|
| 266 |
+
"grad_norm": 0.5837363004684448,
|
| 267 |
+
"learning_rate": 9.988750103285883e-06,
|
| 268 |
+
"loss": 0.7715,
|
| 269 |
+
"step": 360
|
| 270 |
+
},
|
| 271 |
+
{
|
| 272 |
+
"epoch": 0.23,
|
| 273 |
+
"grad_norm": 0.49089643359184265,
|
| 274 |
+
"learning_rate": 9.988078879865396e-06,
|
| 275 |
+
"loss": 0.738,
|
| 276 |
+
"step": 370
|
| 277 |
+
},
|
| 278 |
+
{
|
| 279 |
+
"epoch": 0.24,
|
| 280 |
+
"grad_norm": 0.508166491985321,
|
| 281 |
+
"learning_rate": 9.987388232081694e-06,
|
| 282 |
+
"loss": 0.8025,
|
| 283 |
+
"step": 380
|
| 284 |
+
},
|
| 285 |
+
{
|
| 286 |
+
"epoch": 0.24,
|
| 287 |
+
"grad_norm": 0.6415013074874878,
|
| 288 |
+
"learning_rate": 9.98667816262427e-06,
|
| 289 |
+
"loss": 0.7561,
|
| 290 |
+
"step": 390
|
| 291 |
+
},
|
| 292 |
+
{
|
| 293 |
+
"epoch": 0.25,
|
| 294 |
+
"grad_norm": 0.5850837230682373,
|
| 295 |
+
"learning_rate": 9.985948674258243e-06,
|
| 296 |
+
"loss": 0.7549,
|
| 297 |
+
"step": 400
|
| 298 |
+
},
|
| 299 |
+
{
|
| 300 |
+
"epoch": 0.25,
|
| 301 |
+
"eval_loss": 0.743410587310791,
|
| 302 |
+
"eval_runtime": 64.8376,
|
| 303 |
+
"eval_samples_per_second": 69.404,
|
| 304 |
+
"eval_steps_per_second": 4.349,
|
| 305 |
+
"step": 400
|
| 306 |
+
},
|
| 307 |
+
{
|
| 308 |
+
"epoch": 0.26,
|
| 309 |
+
"grad_norm": 0.627358615398407,
|
| 310 |
+
"learning_rate": 9.985199769824359e-06,
|
| 311 |
+
"loss": 0.7694,
|
| 312 |
+
"step": 410
|
| 313 |
+
},
|
| 314 |
+
{
|
| 315 |
+
"epoch": 0.26,
|
| 316 |
+
"grad_norm": 0.7586867213249207,
|
| 317 |
+
"learning_rate": 9.984431452238968e-06,
|
| 318 |
+
"loss": 0.7353,
|
| 319 |
+
"step": 420
|
| 320 |
+
},
|
| 321 |
+
{
|
| 322 |
+
"epoch": 0.27,
|
| 323 |
+
"grad_norm": 0.5713008642196655,
|
| 324 |
+
"learning_rate": 9.983643724494017e-06,
|
| 325 |
+
"loss": 0.7299,
|
| 326 |
+
"step": 430
|
| 327 |
+
},
|
| 328 |
+
{
|
| 329 |
+
"epoch": 0.28,
|
| 330 |
+
"grad_norm": 0.5664968490600586,
|
| 331 |
+
"learning_rate": 9.982836589657043e-06,
|
| 332 |
+
"loss": 0.754,
|
| 333 |
+
"step": 440
|
| 334 |
+
},
|
| 335 |
+
{
|
| 336 |
+
"epoch": 0.28,
|
| 337 |
+
"grad_norm": 0.4575900435447693,
|
| 338 |
+
"learning_rate": 9.98201005087116e-06,
|
| 339 |
+
"loss": 0.7355,
|
| 340 |
+
"step": 450
|
| 341 |
+
},
|
| 342 |
+
{
|
| 343 |
+
"epoch": 0.29,
|
| 344 |
+
"grad_norm": 0.6498897075653076,
|
| 345 |
+
"learning_rate": 9.981164111355036e-06,
|
| 346 |
+
"loss": 0.7543,
|
| 347 |
+
"step": 460
|
| 348 |
+
},
|
| 349 |
+
{
|
| 350 |
+
"epoch": 0.29,
|
| 351 |
+
"grad_norm": 0.6509144306182861,
|
| 352 |
+
"learning_rate": 9.98029877440289e-06,
|
| 353 |
+
"loss": 0.7568,
|
| 354 |
+
"step": 470
|
| 355 |
+
},
|
| 356 |
+
{
|
| 357 |
+
"epoch": 0.3,
|
| 358 |
+
"grad_norm": 0.44653260707855225,
|
| 359 |
+
"learning_rate": 9.979414043384485e-06,
|
| 360 |
+
"loss": 0.7313,
|
| 361 |
+
"step": 480
|
| 362 |
+
},
|
| 363 |
+
{
|
| 364 |
+
"epoch": 0.31,
|
| 365 |
+
"grad_norm": 0.7275229096412659,
|
| 366 |
+
"learning_rate": 9.978509921745101e-06,
|
| 367 |
+
"loss": 0.7456,
|
| 368 |
+
"step": 490
|
| 369 |
+
},
|
| 370 |
+
{
|
| 371 |
+
"epoch": 0.31,
|
| 372 |
+
"grad_norm": 0.4918762147426605,
|
| 373 |
+
"learning_rate": 9.97758641300553e-06,
|
| 374 |
+
"loss": 0.7585,
|
| 375 |
+
"step": 500
|
| 376 |
+
},
|
| 377 |
+
{
|
| 378 |
+
"epoch": 0.32,
|
| 379 |
+
"grad_norm": 0.5181304216384888,
|
| 380 |
+
"learning_rate": 9.97664352076206e-06,
|
| 381 |
+
"loss": 0.7311,
|
| 382 |
+
"step": 510
|
| 383 |
+
},
|
| 384 |
+
{
|
| 385 |
+
"epoch": 0.33,
|
| 386 |
+
"grad_norm": 0.5354281663894653,
|
| 387 |
+
"learning_rate": 9.97568124868646e-06,
|
| 388 |
+
"loss": 0.7173,
|
| 389 |
+
"step": 520
|
| 390 |
+
},
|
| 391 |
+
{
|
| 392 |
+
"epoch": 0.33,
|
| 393 |
+
"grad_norm": 0.47694316506385803,
|
| 394 |
+
"learning_rate": 9.974699600525972e-06,
|
| 395 |
+
"loss": 0.7408,
|
| 396 |
+
"step": 530
|
| 397 |
+
},
|
| 398 |
+
{
|
| 399 |
+
"epoch": 0.34,
|
| 400 |
+
"grad_norm": 0.5888867974281311,
|
| 401 |
+
"learning_rate": 9.973698580103286e-06,
|
| 402 |
+
"loss": 0.757,
|
| 403 |
+
"step": 540
|
| 404 |
+
},
|
| 405 |
+
{
|
| 406 |
+
"epoch": 0.35,
|
| 407 |
+
"grad_norm": 0.7656754851341248,
|
| 408 |
+
"learning_rate": 9.972678191316533e-06,
|
| 409 |
+
"loss": 0.7717,
|
| 410 |
+
"step": 550
|
| 411 |
+
},
|
| 412 |
+
{
|
| 413 |
+
"epoch": 0.35,
|
| 414 |
+
"grad_norm": 0.5808092355728149,
|
| 415 |
+
"learning_rate": 9.971638438139266e-06,
|
| 416 |
+
"loss": 0.7314,
|
| 417 |
+
"step": 560
|
| 418 |
+
},
|
| 419 |
+
{
|
| 420 |
+
"epoch": 0.36,
|
| 421 |
+
"grad_norm": 0.5002965331077576,
|
| 422 |
+
"learning_rate": 9.97057932462045e-06,
|
| 423 |
+
"loss": 0.7112,
|
| 424 |
+
"step": 570
|
| 425 |
+
},
|
| 426 |
+
{
|
| 427 |
+
"epoch": 0.36,
|
| 428 |
+
"grad_norm": 0.6044530272483826,
|
| 429 |
+
"learning_rate": 9.96950085488444e-06,
|
| 430 |
+
"loss": 0.7802,
|
| 431 |
+
"step": 580
|
| 432 |
+
},
|
| 433 |
+
{
|
| 434 |
+
"epoch": 0.37,
|
| 435 |
+
"grad_norm": 0.48741769790649414,
|
| 436 |
+
"learning_rate": 9.968403033130963e-06,
|
| 437 |
+
"loss": 0.7472,
|
| 438 |
+
"step": 590
|
| 439 |
+
},
|
| 440 |
+
{
|
| 441 |
+
"epoch": 0.38,
|
| 442 |
+
"grad_norm": 0.4956966042518616,
|
| 443 |
+
"learning_rate": 9.967285863635112e-06,
|
| 444 |
+
"loss": 0.7552,
|
| 445 |
+
"step": 600
|
| 446 |
+
},
|
| 447 |
+
{
|
| 448 |
+
"epoch": 0.38,
|
| 449 |
+
"eval_loss": 0.733000636100769,
|
| 450 |
+
"eval_runtime": 65.6052,
|
| 451 |
+
"eval_samples_per_second": 68.592,
|
| 452 |
+
"eval_steps_per_second": 4.298,
|
| 453 |
+
"step": 600
|
| 454 |
+
},
|
| 455 |
+
{
|
| 456 |
+
"epoch": 0.38,
|
| 457 |
+
"grad_norm": 0.528469979763031,
|
| 458 |
+
"learning_rate": 9.966149350747321e-06,
|
| 459 |
+
"loss": 0.7274,
|
| 460 |
+
"step": 610
|
| 461 |
+
},
|
| 462 |
+
{
|
| 463 |
+
"epoch": 0.39,
|
| 464 |
+
"grad_norm": 0.5717535614967346,
|
| 465 |
+
"learning_rate": 9.964993498893349e-06,
|
| 466 |
+
"loss": 0.7734,
|
| 467 |
+
"step": 620
|
| 468 |
+
},
|
| 469 |
+
{
|
| 470 |
+
"epoch": 0.4,
|
| 471 |
+
"grad_norm": 0.5049377083778381,
|
| 472 |
+
"learning_rate": 9.963818312574265e-06,
|
| 473 |
+
"loss": 0.7117,
|
| 474 |
+
"step": 630
|
| 475 |
+
},
|
| 476 |
+
{
|
| 477 |
+
"epoch": 0.4,
|
| 478 |
+
"grad_norm": 0.7002434134483337,
|
| 479 |
+
"learning_rate": 9.962623796366428e-06,
|
| 480 |
+
"loss": 0.7256,
|
| 481 |
+
"step": 640
|
| 482 |
+
},
|
| 483 |
+
{
|
| 484 |
+
"epoch": 0.41,
|
| 485 |
+
"grad_norm": 0.6600221991539001,
|
| 486 |
+
"learning_rate": 9.961409954921472e-06,
|
| 487 |
+
"loss": 0.764,
|
| 488 |
+
"step": 650
|
| 489 |
+
},
|
| 490 |
+
{
|
| 491 |
+
"epoch": 0.41,
|
| 492 |
+
"grad_norm": 0.5288920402526855,
|
| 493 |
+
"learning_rate": 9.96017679296629e-06,
|
| 494 |
+
"loss": 0.7385,
|
| 495 |
+
"step": 660
|
| 496 |
+
},
|
| 497 |
+
{
|
| 498 |
+
"epoch": 0.42,
|
| 499 |
+
"grad_norm": 0.6407844424247742,
|
| 500 |
+
"learning_rate": 9.958924315303005e-06,
|
| 501 |
+
"loss": 0.7386,
|
| 502 |
+
"step": 670
|
| 503 |
+
},
|
| 504 |
+
{
|
| 505 |
+
"epoch": 0.43,
|
| 506 |
+
"grad_norm": 0.6425316333770752,
|
| 507 |
+
"learning_rate": 9.95765252680896e-06,
|
| 508 |
+
"loss": 0.7013,
|
| 509 |
+
"step": 680
|
| 510 |
+
},
|
| 511 |
+
{
|
| 512 |
+
"epoch": 0.43,
|
| 513 |
+
"grad_norm": 0.6219075918197632,
|
| 514 |
+
"learning_rate": 9.956361432436705e-06,
|
| 515 |
+
"loss": 0.7104,
|
| 516 |
+
"step": 690
|
| 517 |
+
},
|
| 518 |
+
{
|
| 519 |
+
"epoch": 0.44,
|
| 520 |
+
"grad_norm": 0.5872789621353149,
|
| 521 |
+
"learning_rate": 9.95505103721396e-06,
|
| 522 |
+
"loss": 0.6988,
|
| 523 |
+
"step": 700
|
| 524 |
+
},
|
| 525 |
+
{
|
| 526 |
+
"epoch": 0.45,
|
| 527 |
+
"grad_norm": 0.8937903642654419,
|
| 528 |
+
"learning_rate": 9.953721346243613e-06,
|
| 529 |
+
"loss": 0.7177,
|
| 530 |
+
"step": 710
|
| 531 |
+
},
|
| 532 |
+
{
|
| 533 |
+
"epoch": 0.45,
|
| 534 |
+
"grad_norm": 0.5471718311309814,
|
| 535 |
+
"learning_rate": 9.952372364703688e-06,
|
| 536 |
+
"loss": 0.6804,
|
| 537 |
+
"step": 720
|
| 538 |
+
},
|
| 539 |
+
{
|
| 540 |
+
"epoch": 0.46,
|
| 541 |
+
"grad_norm": 0.7264242172241211,
|
| 542 |
+
"learning_rate": 9.95100409784733e-06,
|
| 543 |
+
"loss": 0.7432,
|
| 544 |
+
"step": 730
|
| 545 |
+
},
|
| 546 |
+
{
|
| 547 |
+
"epoch": 0.46,
|
| 548 |
+
"grad_norm": 0.7826697826385498,
|
| 549 |
+
"learning_rate": 9.949616551002787e-06,
|
| 550 |
+
"loss": 0.7521,
|
| 551 |
+
"step": 740
|
| 552 |
+
},
|
| 553 |
+
{
|
| 554 |
+
"epoch": 0.47,
|
| 555 |
+
"grad_norm": 0.6297461986541748,
|
| 556 |
+
"learning_rate": 9.948209729573384e-06,
|
| 557 |
+
"loss": 0.7624,
|
| 558 |
+
"step": 750
|
| 559 |
+
},
|
| 560 |
+
{
|
| 561 |
+
"epoch": 0.48,
|
| 562 |
+
"grad_norm": 0.7424671053886414,
|
| 563 |
+
"learning_rate": 9.946783639037503e-06,
|
| 564 |
+
"loss": 0.7619,
|
| 565 |
+
"step": 760
|
| 566 |
+
},
|
| 567 |
+
{
|
| 568 |
+
"epoch": 0.48,
|
| 569 |
+
"grad_norm": 0.5803889632225037,
|
| 570 |
+
"learning_rate": 9.945338284948568e-06,
|
| 571 |
+
"loss": 0.7159,
|
| 572 |
+
"step": 770
|
| 573 |
+
},
|
| 574 |
+
{
|
| 575 |
+
"epoch": 0.49,
|
| 576 |
+
"grad_norm": 0.895115077495575,
|
| 577 |
+
"learning_rate": 9.943873672935014e-06,
|
| 578 |
+
"loss": 0.7621,
|
| 579 |
+
"step": 780
|
| 580 |
+
},
|
| 581 |
+
{
|
| 582 |
+
"epoch": 0.5,
|
| 583 |
+
"grad_norm": 0.5325012803077698,
|
| 584 |
+
"learning_rate": 9.94238980870027e-06,
|
| 585 |
+
"loss": 0.6923,
|
| 586 |
+
"step": 790
|
| 587 |
+
},
|
| 588 |
+
{
|
| 589 |
+
"epoch": 0.5,
|
| 590 |
+
"grad_norm": 0.7378474473953247,
|
| 591 |
+
"learning_rate": 9.940886698022733e-06,
|
| 592 |
+
"loss": 0.7265,
|
| 593 |
+
"step": 800
|
| 594 |
+
},
|
| 595 |
+
{
|
| 596 |
+
"epoch": 0.5,
|
| 597 |
+
"eval_loss": 0.7256230711936951,
|
| 598 |
+
"eval_runtime": 70.6462,
|
| 599 |
+
"eval_samples_per_second": 63.698,
|
| 600 |
+
"eval_steps_per_second": 3.992,
|
| 601 |
+
"step": 800
|
| 602 |
+
},
|
| 603 |
+
{
|
| 604 |
+
"epoch": 0.51,
|
| 605 |
+
"grad_norm": 0.6454309225082397,
|
| 606 |
+
"learning_rate": 9.93936434675576e-06,
|
| 607 |
+
"loss": 0.6976,
|
| 608 |
+
"step": 810
|
| 609 |
+
},
|
| 610 |
+
{
|
| 611 |
+
"epoch": 0.51,
|
| 612 |
+
"grad_norm": 0.7399590015411377,
|
| 613 |
+
"learning_rate": 9.93782276082762e-06,
|
| 614 |
+
"loss": 0.7028,
|
| 615 |
+
"step": 820
|
| 616 |
+
},
|
| 617 |
+
{
|
| 618 |
+
"epoch": 0.52,
|
| 619 |
+
"grad_norm": 0.6661127209663391,
|
| 620 |
+
"learning_rate": 9.936261946241492e-06,
|
| 621 |
+
"loss": 0.7253,
|
| 622 |
+
"step": 830
|
| 623 |
+
},
|
| 624 |
+
{
|
| 625 |
+
"epoch": 0.53,
|
| 626 |
+
"grad_norm": 0.5480040907859802,
|
| 627 |
+
"learning_rate": 9.934681909075434e-06,
|
| 628 |
+
"loss": 0.7096,
|
| 629 |
+
"step": 840
|
| 630 |
+
},
|
| 631 |
+
{
|
| 632 |
+
"epoch": 0.53,
|
| 633 |
+
"grad_norm": 0.6889688968658447,
|
| 634 |
+
"learning_rate": 9.93308265548236e-06,
|
| 635 |
+
"loss": 0.745,
|
| 636 |
+
"step": 850
|
| 637 |
+
},
|
| 638 |
+
{
|
| 639 |
+
"epoch": 0.54,
|
| 640 |
+
"grad_norm": 0.6629202961921692,
|
| 641 |
+
"learning_rate": 9.931464191690015e-06,
|
| 642 |
+
"loss": 0.7111,
|
| 643 |
+
"step": 860
|
| 644 |
+
},
|
| 645 |
+
{
|
| 646 |
+
"epoch": 0.55,
|
| 647 |
+
"grad_norm": 0.5166647434234619,
|
| 648 |
+
"learning_rate": 9.929826524000948e-06,
|
| 649 |
+
"loss": 0.7296,
|
| 650 |
+
"step": 870
|
| 651 |
+
},
|
| 652 |
+
{
|
| 653 |
+
"epoch": 0.55,
|
| 654 |
+
"grad_norm": 0.6730151772499084,
|
| 655 |
+
"learning_rate": 9.928169658792498e-06,
|
| 656 |
+
"loss": 0.7387,
|
| 657 |
+
"step": 880
|
| 658 |
+
},
|
| 659 |
+
{
|
| 660 |
+
"epoch": 0.56,
|
| 661 |
+
"grad_norm": 0.6847391724586487,
|
| 662 |
+
"learning_rate": 9.926493602516758e-06,
|
| 663 |
+
"loss": 0.7156,
|
| 664 |
+
"step": 890
|
| 665 |
+
},
|
| 666 |
+
{
|
| 667 |
+
"epoch": 0.56,
|
| 668 |
+
"grad_norm": 0.7915560007095337,
|
| 669 |
+
"learning_rate": 9.924798361700554e-06,
|
| 670 |
+
"loss": 0.7956,
|
| 671 |
+
"step": 900
|
| 672 |
+
},
|
| 673 |
+
{
|
| 674 |
+
"epoch": 0.57,
|
| 675 |
+
"grad_norm": 0.5927907824516296,
|
| 676 |
+
"learning_rate": 9.923083942945419e-06,
|
| 677 |
+
"loss": 0.7361,
|
| 678 |
+
"step": 910
|
| 679 |
+
},
|
| 680 |
+
{
|
| 681 |
+
"epoch": 0.58,
|
| 682 |
+
"grad_norm": 0.7477264404296875,
|
| 683 |
+
"learning_rate": 9.92135035292757e-06,
|
| 684 |
+
"loss": 0.7091,
|
| 685 |
+
"step": 920
|
| 686 |
+
},
|
| 687 |
+
{
|
| 688 |
+
"epoch": 0.58,
|
| 689 |
+
"grad_norm": 0.7492902278900146,
|
| 690 |
+
"learning_rate": 9.919597598397882e-06,
|
| 691 |
+
"loss": 0.6967,
|
| 692 |
+
"step": 930
|
| 693 |
+
},
|
| 694 |
+
{
|
| 695 |
+
"epoch": 0.59,
|
| 696 |
+
"grad_norm": 0.7357175350189209,
|
| 697 |
+
"learning_rate": 9.91782568618185e-06,
|
| 698 |
+
"loss": 0.7509,
|
| 699 |
+
"step": 940
|
| 700 |
+
},
|
| 701 |
+
{
|
| 702 |
+
"epoch": 0.6,
|
| 703 |
+
"grad_norm": 0.6629440188407898,
|
| 704 |
+
"learning_rate": 9.916034623179584e-06,
|
| 705 |
+
"loss": 0.6999,
|
| 706 |
+
"step": 950
|
| 707 |
+
},
|
| 708 |
+
{
|
| 709 |
+
"epoch": 0.6,
|
| 710 |
+
"grad_norm": 0.5954321026802063,
|
| 711 |
+
"learning_rate": 9.914224416365765e-06,
|
| 712 |
+
"loss": 0.7194,
|
| 713 |
+
"step": 960
|
| 714 |
+
},
|
| 715 |
+
{
|
| 716 |
+
"epoch": 0.61,
|
| 717 |
+
"grad_norm": 0.9139691591262817,
|
| 718 |
+
"learning_rate": 9.91239507278962e-06,
|
| 719 |
+
"loss": 0.705,
|
| 720 |
+
"step": 970
|
| 721 |
+
},
|
| 722 |
+
{
|
| 723 |
+
"epoch": 0.61,
|
| 724 |
+
"grad_norm": 0.9337642788887024,
|
| 725 |
+
"learning_rate": 9.910546599574903e-06,
|
| 726 |
+
"loss": 0.7314,
|
| 727 |
+
"step": 980
|
| 728 |
+
},
|
| 729 |
+
{
|
| 730 |
+
"epoch": 0.62,
|
| 731 |
+
"grad_norm": 0.6616548299789429,
|
| 732 |
+
"learning_rate": 9.908679003919856e-06,
|
| 733 |
+
"loss": 0.7549,
|
| 734 |
+
"step": 990
|
| 735 |
+
},
|
| 736 |
+
{
|
| 737 |
+
"epoch": 0.63,
|
| 738 |
+
"grad_norm": 0.6958469152450562,
|
| 739 |
+
"learning_rate": 9.906792293097194e-06,
|
| 740 |
+
"loss": 0.7524,
|
| 741 |
+
"step": 1000
|
| 742 |
+
},
|
| 743 |
+
{
|
| 744 |
+
"epoch": 0.63,
|
| 745 |
+
"eval_loss": 0.7200015187263489,
|
| 746 |
+
"eval_runtime": 68.2079,
|
| 747 |
+
"eval_samples_per_second": 65.975,
|
| 748 |
+
"eval_steps_per_second": 4.134,
|
| 749 |
+
"step": 1000
|
| 750 |
+
},
|
| 751 |
+
{
|
| 752 |
+
"epoch": 0.63,
|
| 753 |
+
"grad_norm": 0.5352278351783752,
|
| 754 |
+
"learning_rate": 9.904886474454063e-06,
|
| 755 |
+
"loss": 0.7218,
|
| 756 |
+
"step": 1010
|
| 757 |
+
},
|
| 758 |
+
{
|
| 759 |
+
"epoch": 0.64,
|
| 760 |
+
"grad_norm": 0.6772333979606628,
|
| 761 |
+
"learning_rate": 9.90296155541202e-06,
|
| 762 |
+
"loss": 0.7171,
|
| 763 |
+
"step": 1020
|
| 764 |
+
},
|
| 765 |
+
{
|
| 766 |
+
"epoch": 0.65,
|
| 767 |
+
"grad_norm": 0.7102545499801636,
|
| 768 |
+
"learning_rate": 9.901017543467005e-06,
|
| 769 |
+
"loss": 0.758,
|
| 770 |
+
"step": 1030
|
| 771 |
+
},
|
| 772 |
+
{
|
| 773 |
+
"epoch": 0.65,
|
| 774 |
+
"grad_norm": 0.5379916429519653,
|
| 775 |
+
"learning_rate": 9.899054446189305e-06,
|
| 776 |
+
"loss": 0.7121,
|
| 777 |
+
"step": 1040
|
| 778 |
+
},
|
| 779 |
+
{
|
| 780 |
+
"epoch": 0.66,
|
| 781 |
+
"grad_norm": 0.6267510056495667,
|
| 782 |
+
"learning_rate": 9.897072271223526e-06,
|
| 783 |
+
"loss": 0.7088,
|
| 784 |
+
"step": 1050
|
| 785 |
+
},
|
| 786 |
+
{
|
| 787 |
+
"epoch": 0.66,
|
| 788 |
+
"grad_norm": 0.9392660856246948,
|
| 789 |
+
"learning_rate": 9.895071026288574e-06,
|
| 790 |
+
"loss": 0.7804,
|
| 791 |
+
"step": 1060
|
| 792 |
+
},
|
| 793 |
+
{
|
| 794 |
+
"epoch": 0.67,
|
| 795 |
+
"grad_norm": 0.9270221590995789,
|
| 796 |
+
"learning_rate": 9.893050719177608e-06,
|
| 797 |
+
"loss": 0.6935,
|
| 798 |
+
"step": 1070
|
| 799 |
+
},
|
| 800 |
+
{
|
| 801 |
+
"epoch": 0.68,
|
| 802 |
+
"grad_norm": 0.722115159034729,
|
| 803 |
+
"learning_rate": 9.891011357758022e-06,
|
| 804 |
+
"loss": 0.6894,
|
| 805 |
+
"step": 1080
|
| 806 |
+
},
|
| 807 |
+
{
|
| 808 |
+
"epoch": 0.68,
|
| 809 |
+
"grad_norm": 0.7055147886276245,
|
| 810 |
+
"learning_rate": 9.888952949971411e-06,
|
| 811 |
+
"loss": 0.7244,
|
| 812 |
+
"step": 1090
|
| 813 |
+
},
|
| 814 |
+
{
|
| 815 |
+
"epoch": 0.69,
|
| 816 |
+
"grad_norm": 0.7774051427841187,
|
| 817 |
+
"learning_rate": 9.886875503833537e-06,
|
| 818 |
+
"loss": 0.8156,
|
| 819 |
+
"step": 1100
|
| 820 |
+
},
|
| 821 |
+
{
|
| 822 |
+
"epoch": 0.7,
|
| 823 |
+
"grad_norm": 0.7965037226676941,
|
| 824 |
+
"learning_rate": 9.884779027434304e-06,
|
| 825 |
+
"loss": 0.7478,
|
| 826 |
+
"step": 1110
|
| 827 |
+
},
|
| 828 |
+
{
|
| 829 |
+
"epoch": 0.7,
|
| 830 |
+
"grad_norm": 0.8204682469367981,
|
| 831 |
+
"learning_rate": 9.882663528937716e-06,
|
| 832 |
+
"loss": 0.7187,
|
| 833 |
+
"step": 1120
|
| 834 |
+
},
|
| 835 |
+
{
|
| 836 |
+
"epoch": 0.71,
|
| 837 |
+
"grad_norm": 0.63904869556427,
|
| 838 |
+
"learning_rate": 9.880529016581863e-06,
|
| 839 |
+
"loss": 0.7145,
|
| 840 |
+
"step": 1130
|
| 841 |
+
},
|
| 842 |
+
{
|
| 843 |
+
"epoch": 0.72,
|
| 844 |
+
"grad_norm": 0.6523028612136841,
|
| 845 |
+
"learning_rate": 9.878375498678869e-06,
|
| 846 |
+
"loss": 0.731,
|
| 847 |
+
"step": 1140
|
| 848 |
+
},
|
| 849 |
+
{
|
| 850 |
+
"epoch": 0.72,
|
| 851 |
+
"grad_norm": 0.7148768901824951,
|
| 852 |
+
"learning_rate": 9.876202983614868e-06,
|
| 853 |
+
"loss": 0.7323,
|
| 854 |
+
"step": 1150
|
| 855 |
+
},
|
| 856 |
+
{
|
| 857 |
+
"epoch": 0.73,
|
| 858 |
+
"grad_norm": 0.6108402013778687,
|
| 859 |
+
"learning_rate": 9.874011479849981e-06,
|
| 860 |
+
"loss": 0.6757,
|
| 861 |
+
"step": 1160
|
| 862 |
+
},
|
| 863 |
+
{
|
| 864 |
+
"epoch": 0.73,
|
| 865 |
+
"grad_norm": 0.6056957244873047,
|
| 866 |
+
"learning_rate": 9.871800995918264e-06,
|
| 867 |
+
"loss": 0.7258,
|
| 868 |
+
"step": 1170
|
| 869 |
+
},
|
| 870 |
+
{
|
| 871 |
+
"epoch": 0.74,
|
| 872 |
+
"grad_norm": 0.7671077847480774,
|
| 873 |
+
"learning_rate": 9.86957154042769e-06,
|
| 874 |
+
"loss": 0.7334,
|
| 875 |
+
"step": 1180
|
| 876 |
+
},
|
| 877 |
+
{
|
| 878 |
+
"epoch": 0.75,
|
| 879 |
+
"grad_norm": 0.8327913880348206,
|
| 880 |
+
"learning_rate": 9.867323122060108e-06,
|
| 881 |
+
"loss": 0.7358,
|
| 882 |
+
"step": 1190
|
| 883 |
+
},
|
| 884 |
+
{
|
| 885 |
+
"epoch": 0.75,
|
| 886 |
+
"grad_norm": 0.7025701999664307,
|
| 887 |
+
"learning_rate": 9.865055749571215e-06,
|
| 888 |
+
"loss": 0.6976,
|
| 889 |
+
"step": 1200
|
| 890 |
+
},
|
| 891 |
+
{
|
| 892 |
+
"epoch": 0.75,
|
| 893 |
+
"eval_loss": 0.7151169180870056,
|
| 894 |
+
"eval_runtime": 64.9708,
|
| 895 |
+
"eval_samples_per_second": 69.262,
|
| 896 |
+
"eval_steps_per_second": 4.34,
|
| 897 |
+
"step": 1200
|
| 898 |
+
},
|
| 899 |
+
{
|
| 900 |
+
"epoch": 0.76,
|
| 901 |
+
"grad_norm": 0.8391425609588623,
|
| 902 |
+
"learning_rate": 9.862769431790513e-06,
|
| 903 |
+
"loss": 0.6983,
|
| 904 |
+
"step": 1210
|
| 905 |
+
},
|
| 906 |
+
{
|
| 907 |
+
"epoch": 0.77,
|
| 908 |
+
"grad_norm": 0.7243052124977112,
|
| 909 |
+
"learning_rate": 9.860464177621286e-06,
|
| 910 |
+
"loss": 0.7171,
|
| 911 |
+
"step": 1220
|
| 912 |
+
},
|
| 913 |
+
{
|
| 914 |
+
"epoch": 0.77,
|
| 915 |
+
"grad_norm": 0.6501705050468445,
|
| 916 |
+
"learning_rate": 9.858139996040554e-06,
|
| 917 |
+
"loss": 0.7206,
|
| 918 |
+
"step": 1230
|
| 919 |
+
},
|
| 920 |
+
{
|
| 921 |
+
"epoch": 0.78,
|
| 922 |
+
"grad_norm": 0.8618900179862976,
|
| 923 |
+
"learning_rate": 9.855796896099044e-06,
|
| 924 |
+
"loss": 0.7368,
|
| 925 |
+
"step": 1240
|
| 926 |
+
},
|
| 927 |
+
{
|
| 928 |
+
"epoch": 0.78,
|
| 929 |
+
"grad_norm": 0.826347291469574,
|
| 930 |
+
"learning_rate": 9.85343488692116e-06,
|
| 931 |
+
"loss": 0.7372,
|
| 932 |
+
"step": 1250
|
| 933 |
+
},
|
| 934 |
+
{
|
| 935 |
+
"epoch": 0.79,
|
| 936 |
+
"grad_norm": 0.8590556383132935,
|
| 937 |
+
"learning_rate": 9.851053977704931e-06,
|
| 938 |
+
"loss": 0.7373,
|
| 939 |
+
"step": 1260
|
| 940 |
+
},
|
| 941 |
+
{
|
| 942 |
+
"epoch": 0.8,
|
| 943 |
+
"grad_norm": 0.8719233274459839,
|
| 944 |
+
"learning_rate": 9.848654177721999e-06,
|
| 945 |
+
"loss": 0.7608,
|
| 946 |
+
"step": 1270
|
| 947 |
+
},
|
| 948 |
+
{
|
| 949 |
+
"epoch": 0.8,
|
| 950 |
+
"grad_norm": 0.7729814052581787,
|
| 951 |
+
"learning_rate": 9.846235496317556e-06,
|
| 952 |
+
"loss": 0.7227,
|
| 953 |
+
"step": 1280
|
| 954 |
+
},
|
| 955 |
+
{
|
| 956 |
+
"epoch": 0.81,
|
| 957 |
+
"grad_norm": 0.801908016204834,
|
| 958 |
+
"learning_rate": 9.843797942910328e-06,
|
| 959 |
+
"loss": 0.7415,
|
| 960 |
+
"step": 1290
|
| 961 |
+
},
|
| 962 |
+
{
|
| 963 |
+
"epoch": 0.82,
|
| 964 |
+
"grad_norm": 0.9884589910507202,
|
| 965 |
+
"learning_rate": 9.841341526992536e-06,
|
| 966 |
+
"loss": 0.7206,
|
| 967 |
+
"step": 1300
|
| 968 |
+
},
|
| 969 |
+
{
|
| 970 |
+
"epoch": 0.82,
|
| 971 |
+
"grad_norm": 0.7067356705665588,
|
| 972 |
+
"learning_rate": 9.838866258129847e-06,
|
| 973 |
+
"loss": 0.6704,
|
| 974 |
+
"step": 1310
|
| 975 |
+
},
|
| 976 |
+
{
|
| 977 |
+
"epoch": 0.83,
|
| 978 |
+
"grad_norm": 0.7258339524269104,
|
| 979 |
+
"learning_rate": 9.836372145961346e-06,
|
| 980 |
+
"loss": 0.7159,
|
| 981 |
+
"step": 1320
|
| 982 |
+
},
|
| 983 |
+
{
|
| 984 |
+
"epoch": 0.83,
|
| 985 |
+
"grad_norm": 0.8512592315673828,
|
| 986 |
+
"learning_rate": 9.833859200199498e-06,
|
| 987 |
+
"loss": 0.6916,
|
| 988 |
+
"step": 1330
|
| 989 |
+
},
|
| 990 |
+
{
|
| 991 |
+
"epoch": 0.84,
|
| 992 |
+
"grad_norm": 0.7856159210205078,
|
| 993 |
+
"learning_rate": 9.83132743063011e-06,
|
| 994 |
+
"loss": 0.7568,
|
| 995 |
+
"step": 1340
|
| 996 |
+
},
|
| 997 |
+
{
|
| 998 |
+
"epoch": 0.85,
|
| 999 |
+
"grad_norm": 0.7149519324302673,
|
| 1000 |
+
"learning_rate": 9.82877684711229e-06,
|
| 1001 |
+
"loss": 0.7017,
|
| 1002 |
+
"step": 1350
|
| 1003 |
+
},
|
| 1004 |
+
{
|
| 1005 |
+
"epoch": 0.85,
|
| 1006 |
+
"grad_norm": 1.0214589834213257,
|
| 1007 |
+
"learning_rate": 9.826207459578412e-06,
|
| 1008 |
+
"loss": 0.7127,
|
| 1009 |
+
"step": 1360
|
| 1010 |
+
},
|
| 1011 |
+
{
|
| 1012 |
+
"epoch": 0.86,
|
| 1013 |
+
"grad_norm": 1.0295792818069458,
|
| 1014 |
+
"learning_rate": 9.823619278034073e-06,
|
| 1015 |
+
"loss": 0.7013,
|
| 1016 |
+
"step": 1370
|
| 1017 |
+
},
|
| 1018 |
+
{
|
| 1019 |
+
"epoch": 0.87,
|
| 1020 |
+
"grad_norm": 0.8674212694168091,
|
| 1021 |
+
"learning_rate": 9.821012312558059e-06,
|
| 1022 |
+
"loss": 0.6942,
|
| 1023 |
+
"step": 1380
|
| 1024 |
+
},
|
| 1025 |
+
{
|
| 1026 |
+
"epoch": 0.87,
|
| 1027 |
+
"grad_norm": 0.7604880332946777,
|
| 1028 |
+
"learning_rate": 9.818386573302305e-06,
|
| 1029 |
+
"loss": 0.7013,
|
| 1030 |
+
"step": 1390
|
| 1031 |
+
},
|
| 1032 |
+
{
|
| 1033 |
+
"epoch": 0.88,
|
| 1034 |
+
"grad_norm": 0.7863268852233887,
|
| 1035 |
+
"learning_rate": 9.815742070491852e-06,
|
| 1036 |
+
"loss": 0.7408,
|
| 1037 |
+
"step": 1400
|
| 1038 |
+
},
|
| 1039 |
+
{
|
| 1040 |
+
"epoch": 0.88,
|
| 1041 |
+
"eval_loss": 0.7116020917892456,
|
| 1042 |
+
"eval_runtime": 64.7496,
|
| 1043 |
+
"eval_samples_per_second": 69.498,
|
| 1044 |
+
"eval_steps_per_second": 4.355,
|
| 1045 |
+
"step": 1400
|
| 1046 |
+
},
|
| 1047 |
+
{
|
| 1048 |
+
"epoch": 0.88,
|
| 1049 |
+
"grad_norm": 0.7451047301292419,
|
| 1050 |
+
"learning_rate": 9.81307881442481e-06,
|
| 1051 |
+
"loss": 0.7105,
|
| 1052 |
+
"step": 1410
|
| 1053 |
+
},
|
| 1054 |
+
{
|
| 1055 |
+
"epoch": 0.89,
|
| 1056 |
+
"grad_norm": 0.8191768527030945,
|
| 1057 |
+
"learning_rate": 9.810396815472316e-06,
|
| 1058 |
+
"loss": 0.6994,
|
| 1059 |
+
"step": 1420
|
| 1060 |
+
},
|
| 1061 |
+
{
|
| 1062 |
+
"epoch": 0.9,
|
| 1063 |
+
"grad_norm": 0.5049307942390442,
|
| 1064 |
+
"learning_rate": 9.807696084078494e-06,
|
| 1065 |
+
"loss": 0.7459,
|
| 1066 |
+
"step": 1430
|
| 1067 |
+
},
|
| 1068 |
+
{
|
| 1069 |
+
"epoch": 0.9,
|
| 1070 |
+
"grad_norm": 0.762649416923523,
|
| 1071 |
+
"learning_rate": 9.804976630760419e-06,
|
| 1072 |
+
"loss": 0.7048,
|
| 1073 |
+
"step": 1440
|
| 1074 |
+
},
|
| 1075 |
+
{
|
| 1076 |
+
"epoch": 0.91,
|
| 1077 |
+
"grad_norm": 0.8065420985221863,
|
| 1078 |
+
"learning_rate": 9.802238466108068e-06,
|
| 1079 |
+
"loss": 0.6975,
|
| 1080 |
+
"step": 1450
|
| 1081 |
+
},
|
| 1082 |
+
{
|
| 1083 |
+
"epoch": 0.92,
|
| 1084 |
+
"grad_norm": 0.899728000164032,
|
| 1085 |
+
"learning_rate": 9.799481600784286e-06,
|
| 1086 |
+
"loss": 0.737,
|
| 1087 |
+
"step": 1460
|
| 1088 |
+
},
|
| 1089 |
+
{
|
| 1090 |
+
"epoch": 0.92,
|
| 1091 |
+
"grad_norm": 0.7029632925987244,
|
| 1092 |
+
"learning_rate": 9.796706045524738e-06,
|
| 1093 |
+
"loss": 0.7236,
|
| 1094 |
+
"step": 1470
|
| 1095 |
+
},
|
| 1096 |
+
{
|
| 1097 |
+
"epoch": 0.93,
|
| 1098 |
+
"grad_norm": 0.7470441460609436,
|
| 1099 |
+
"learning_rate": 9.793911811137874e-06,
|
| 1100 |
+
"loss": 0.6984,
|
| 1101 |
+
"step": 1480
|
| 1102 |
+
},
|
| 1103 |
+
{
|
| 1104 |
+
"epoch": 0.93,
|
| 1105 |
+
"grad_norm": 0.8542289137840271,
|
| 1106 |
+
"learning_rate": 9.791098908504884e-06,
|
| 1107 |
+
"loss": 0.8019,
|
| 1108 |
+
"step": 1490
|
| 1109 |
+
},
|
| 1110 |
+
{
|
| 1111 |
+
"epoch": 0.94,
|
| 1112 |
+
"grad_norm": 0.749045193195343,
|
| 1113 |
+
"learning_rate": 9.788267348579649e-06,
|
| 1114 |
+
"loss": 0.7114,
|
| 1115 |
+
"step": 1500
|
| 1116 |
+
},
|
| 1117 |
+
{
|
| 1118 |
+
"epoch": 0.95,
|
| 1119 |
+
"grad_norm": 0.7834633588790894,
|
| 1120 |
+
"learning_rate": 9.78541714238871e-06,
|
| 1121 |
+
"loss": 0.7222,
|
| 1122 |
+
"step": 1510
|
| 1123 |
+
},
|
| 1124 |
+
{
|
| 1125 |
+
"epoch": 0.95,
|
| 1126 |
+
"grad_norm": 0.8488750457763672,
|
| 1127 |
+
"learning_rate": 9.782548301031218e-06,
|
| 1128 |
+
"loss": 0.7434,
|
| 1129 |
+
"step": 1520
|
| 1130 |
+
},
|
| 1131 |
+
{
|
| 1132 |
+
"epoch": 0.96,
|
| 1133 |
+
"grad_norm": 0.7018651962280273,
|
| 1134 |
+
"learning_rate": 9.77966083567889e-06,
|
| 1135 |
+
"loss": 0.7193,
|
| 1136 |
+
"step": 1530
|
| 1137 |
+
},
|
| 1138 |
+
{
|
| 1139 |
+
"epoch": 0.97,
|
| 1140 |
+
"grad_norm": 0.8260754346847534,
|
| 1141 |
+
"learning_rate": 9.776754757575975e-06,
|
| 1142 |
+
"loss": 0.7763,
|
| 1143 |
+
"step": 1540
|
| 1144 |
+
},
|
| 1145 |
+
{
|
| 1146 |
+
"epoch": 0.97,
|
| 1147 |
+
"grad_norm": 0.8732118010520935,
|
| 1148 |
+
"learning_rate": 9.773830078039193e-06,
|
| 1149 |
+
"loss": 0.7494,
|
| 1150 |
+
"step": 1550
|
| 1151 |
+
},
|
| 1152 |
+
{
|
| 1153 |
+
"epoch": 0.98,
|
| 1154 |
+
"grad_norm": 0.9026480317115784,
|
| 1155 |
+
"learning_rate": 9.77088680845771e-06,
|
| 1156 |
+
"loss": 0.7078,
|
| 1157 |
+
"step": 1560
|
| 1158 |
+
},
|
| 1159 |
+
{
|
| 1160 |
+
"epoch": 0.98,
|
| 1161 |
+
"grad_norm": 0.7559505105018616,
|
| 1162 |
+
"learning_rate": 9.767924960293076e-06,
|
| 1163 |
+
"loss": 0.7468,
|
| 1164 |
+
"step": 1570
|
| 1165 |
+
},
|
| 1166 |
+
{
|
| 1167 |
+
"epoch": 0.99,
|
| 1168 |
+
"grad_norm": 0.8832489848136902,
|
| 1169 |
+
"learning_rate": 9.764944545079197e-06,
|
| 1170 |
+
"loss": 0.7502,
|
| 1171 |
+
"step": 1580
|
| 1172 |
+
},
|
| 1173 |
+
{
|
| 1174 |
+
"epoch": 1.0,
|
| 1175 |
+
"grad_norm": 0.8065813183784485,
|
| 1176 |
+
"learning_rate": 9.761945574422276e-06,
|
| 1177 |
+
"loss": 0.7337,
|
| 1178 |
+
"step": 1590
|
| 1179 |
+
},
|
| 1180 |
+
{
|
| 1181 |
+
"epoch": 1.0,
|
| 1182 |
+
"grad_norm": 0.6966451406478882,
|
| 1183 |
+
"learning_rate": 9.758928060000779e-06,
|
| 1184 |
+
"loss": 0.701,
|
| 1185 |
+
"step": 1600
|
| 1186 |
+
},
|
| 1187 |
+
{
|
| 1188 |
+
"epoch": 1.0,
|
| 1189 |
+
"eval_loss": 0.7084596157073975,
|
| 1190 |
+
"eval_runtime": 64.9175,
|
| 1191 |
+
"eval_samples_per_second": 69.319,
|
| 1192 |
+
"eval_steps_per_second": 4.344,
|
| 1193 |
+
"step": 1600
|
| 1194 |
+
},
|
| 1195 |
+
{
|
| 1196 |
+
"epoch": 1.01,
|
| 1197 |
+
"grad_norm": 0.8769924640655518,
|
| 1198 |
+
"learning_rate": 9.755892013565377e-06,
|
| 1199 |
+
"loss": 0.7014,
|
| 1200 |
+
"step": 1610
|
| 1201 |
+
},
|
| 1202 |
+
{
|
| 1203 |
+
"epoch": 1.02,
|
| 1204 |
+
"grad_norm": 0.8940397500991821,
|
| 1205 |
+
"learning_rate": 9.752837446938915e-06,
|
| 1206 |
+
"loss": 0.7256,
|
| 1207 |
+
"step": 1620
|
| 1208 |
+
},
|
| 1209 |
+
{
|
| 1210 |
+
"epoch": 1.02,
|
| 1211 |
+
"grad_norm": 0.7818279266357422,
|
| 1212 |
+
"learning_rate": 9.749764372016355e-06,
|
| 1213 |
+
"loss": 0.7268,
|
| 1214 |
+
"step": 1630
|
| 1215 |
+
},
|
| 1216 |
+
{
|
| 1217 |
+
"epoch": 1.03,
|
| 1218 |
+
"grad_norm": 0.7369450330734253,
|
| 1219 |
+
"learning_rate": 9.746672800764734e-06,
|
| 1220 |
+
"loss": 0.6968,
|
| 1221 |
+
"step": 1640
|
| 1222 |
+
},
|
| 1223 |
+
{
|
| 1224 |
+
"epoch": 1.04,
|
| 1225 |
+
"grad_norm": 0.8924703001976013,
|
| 1226 |
+
"learning_rate": 9.743562745223118e-06,
|
| 1227 |
+
"loss": 0.7087,
|
| 1228 |
+
"step": 1650
|
| 1229 |
+
},
|
| 1230 |
+
{
|
| 1231 |
+
"epoch": 1.04,
|
| 1232 |
+
"grad_norm": 1.0398907661437988,
|
| 1233 |
+
"learning_rate": 9.740434217502549e-06,
|
| 1234 |
+
"loss": 0.7199,
|
| 1235 |
+
"step": 1660
|
| 1236 |
+
},
|
| 1237 |
+
{
|
| 1238 |
+
"epoch": 1.05,
|
| 1239 |
+
"grad_norm": 0.7427188754081726,
|
| 1240 |
+
"learning_rate": 9.737287229786007e-06,
|
| 1241 |
+
"loss": 0.687,
|
| 1242 |
+
"step": 1670
|
| 1243 |
+
},
|
| 1244 |
+
{
|
| 1245 |
+
"epoch": 1.05,
|
| 1246 |
+
"grad_norm": 0.9230946898460388,
|
| 1247 |
+
"learning_rate": 9.734121794328358e-06,
|
| 1248 |
+
"loss": 0.7003,
|
| 1249 |
+
"step": 1680
|
| 1250 |
+
},
|
| 1251 |
+
{
|
| 1252 |
+
"epoch": 1.06,
|
| 1253 |
+
"grad_norm": 0.8461260795593262,
|
| 1254 |
+
"learning_rate": 9.730937923456303e-06,
|
| 1255 |
+
"loss": 0.7329,
|
| 1256 |
+
"step": 1690
|
| 1257 |
+
},
|
| 1258 |
+
{
|
| 1259 |
+
"epoch": 1.07,
|
| 1260 |
+
"grad_norm": 0.783156156539917,
|
| 1261 |
+
"learning_rate": 9.727735629568335e-06,
|
| 1262 |
+
"loss": 0.6924,
|
| 1263 |
+
"step": 1700
|
| 1264 |
+
},
|
| 1265 |
+
{
|
| 1266 |
+
"epoch": 1.07,
|
| 1267 |
+
"grad_norm": 0.8659111261367798,
|
| 1268 |
+
"learning_rate": 9.724514925134696e-06,
|
| 1269 |
+
"loss": 0.7219,
|
| 1270 |
+
"step": 1710
|
| 1271 |
+
},
|
| 1272 |
+
{
|
| 1273 |
+
"epoch": 1.08,
|
| 1274 |
+
"grad_norm": 0.8218225240707397,
|
| 1275 |
+
"learning_rate": 9.721275822697307e-06,
|
| 1276 |
+
"loss": 0.6741,
|
| 1277 |
+
"step": 1720
|
| 1278 |
+
},
|
| 1279 |
+
{
|
| 1280 |
+
"epoch": 1.09,
|
| 1281 |
+
"grad_norm": 0.8807560205459595,
|
| 1282 |
+
"learning_rate": 9.718018334869748e-06,
|
| 1283 |
+
"loss": 0.7047,
|
| 1284 |
+
"step": 1730
|
| 1285 |
+
},
|
| 1286 |
+
{
|
| 1287 |
+
"epoch": 1.09,
|
| 1288 |
+
"grad_norm": 0.9925751090049744,
|
| 1289 |
+
"learning_rate": 9.714742474337187e-06,
|
| 1290 |
+
"loss": 0.7156,
|
| 1291 |
+
"step": 1740
|
| 1292 |
+
},
|
| 1293 |
+
{
|
| 1294 |
+
"epoch": 1.1,
|
| 1295 |
+
"grad_norm": 0.7675251364707947,
|
| 1296 |
+
"learning_rate": 9.711448253856336e-06,
|
| 1297 |
+
"loss": 0.6887,
|
| 1298 |
+
"step": 1750
|
| 1299 |
+
},
|
| 1300 |
+
{
|
| 1301 |
+
"epoch": 1.1,
|
| 1302 |
+
"grad_norm": 0.783015251159668,
|
| 1303 |
+
"learning_rate": 9.708135686255415e-06,
|
| 1304 |
+
"loss": 0.7373,
|
| 1305 |
+
"step": 1760
|
| 1306 |
+
},
|
| 1307 |
+
{
|
| 1308 |
+
"epoch": 1.11,
|
| 1309 |
+
"grad_norm": 0.8704028129577637,
|
| 1310 |
+
"learning_rate": 9.704804784434077e-06,
|
| 1311 |
+
"loss": 0.6652,
|
| 1312 |
+
"step": 1770
|
| 1313 |
+
},
|
| 1314 |
+
{
|
| 1315 |
+
"epoch": 1.12,
|
| 1316 |
+
"grad_norm": 0.9532449245452881,
|
| 1317 |
+
"learning_rate": 9.701455561363378e-06,
|
| 1318 |
+
"loss": 0.682,
|
| 1319 |
+
"step": 1780
|
| 1320 |
+
},
|
| 1321 |
+
{
|
| 1322 |
+
"epoch": 1.12,
|
| 1323 |
+
"grad_norm": 0.9703534245491028,
|
| 1324 |
+
"learning_rate": 9.698088030085721e-06,
|
| 1325 |
+
"loss": 0.6844,
|
| 1326 |
+
"step": 1790
|
| 1327 |
+
},
|
| 1328 |
+
{
|
| 1329 |
+
"epoch": 1.13,
|
| 1330 |
+
"grad_norm": 1.031153678894043,
|
| 1331 |
+
"learning_rate": 9.694702203714801e-06,
|
| 1332 |
+
"loss": 0.7084,
|
| 1333 |
+
"step": 1800
|
| 1334 |
+
},
|
| 1335 |
+
{
|
| 1336 |
+
"epoch": 1.13,
|
| 1337 |
+
"eval_loss": 0.705936074256897,
|
| 1338 |
+
"eval_runtime": 64.9167,
|
| 1339 |
+
"eval_samples_per_second": 69.32,
|
| 1340 |
+
"eval_steps_per_second": 4.344,
|
| 1341 |
+
"step": 1800
|
| 1342 |
+
},
|
| 1343 |
+
{
|
| 1344 |
+
"epoch": 1.14,
|
| 1345 |
+
"grad_norm": 0.8839524388313293,
|
| 1346 |
+
"learning_rate": 9.691298095435559e-06,
|
| 1347 |
+
"loss": 0.6897,
|
| 1348 |
+
"step": 1810
|
| 1349 |
+
},
|
| 1350 |
+
{
|
| 1351 |
+
"epoch": 1.14,
|
| 1352 |
+
"grad_norm": 1.0173550844192505,
|
| 1353 |
+
"learning_rate": 9.687875718504126e-06,
|
| 1354 |
+
"loss": 0.6851,
|
| 1355 |
+
"step": 1820
|
| 1356 |
+
},
|
| 1357 |
+
{
|
| 1358 |
+
"epoch": 1.15,
|
| 1359 |
+
"grad_norm": 1.0902131795883179,
|
| 1360 |
+
"learning_rate": 9.684435086247777e-06,
|
| 1361 |
+
"loss": 0.7132,
|
| 1362 |
+
"step": 1830
|
| 1363 |
+
},
|
| 1364 |
+
{
|
| 1365 |
+
"epoch": 1.15,
|
| 1366 |
+
"grad_norm": 0.8699798583984375,
|
| 1367 |
+
"learning_rate": 9.680976212064875e-06,
|
| 1368 |
+
"loss": 0.7129,
|
| 1369 |
+
"step": 1840
|
| 1370 |
+
},
|
| 1371 |
+
{
|
| 1372 |
+
"epoch": 1.16,
|
| 1373 |
+
"grad_norm": 0.879970133304596,
|
| 1374 |
+
"learning_rate": 9.677499109424818e-06,
|
| 1375 |
+
"loss": 0.6907,
|
| 1376 |
+
"step": 1850
|
| 1377 |
+
},
|
| 1378 |
+
{
|
| 1379 |
+
"epoch": 1.17,
|
| 1380 |
+
"grad_norm": 0.9659926295280457,
|
| 1381 |
+
"learning_rate": 9.674003791867993e-06,
|
| 1382 |
+
"loss": 0.7327,
|
| 1383 |
+
"step": 1860
|
| 1384 |
+
},
|
| 1385 |
+
{
|
| 1386 |
+
"epoch": 1.17,
|
| 1387 |
+
"grad_norm": 1.0900288820266724,
|
| 1388 |
+
"learning_rate": 9.670490273005713e-06,
|
| 1389 |
+
"loss": 0.7304,
|
| 1390 |
+
"step": 1870
|
| 1391 |
+
},
|
| 1392 |
+
{
|
| 1393 |
+
"epoch": 1.18,
|
| 1394 |
+
"grad_norm": 0.995785117149353,
|
| 1395 |
+
"learning_rate": 9.666958566520175e-06,
|
| 1396 |
+
"loss": 0.7076,
|
| 1397 |
+
"step": 1880
|
| 1398 |
+
},
|
| 1399 |
+
{
|
| 1400 |
+
"epoch": 1.19,
|
| 1401 |
+
"grad_norm": 1.0170907974243164,
|
| 1402 |
+
"learning_rate": 9.663408686164399e-06,
|
| 1403 |
+
"loss": 0.691,
|
| 1404 |
+
"step": 1890
|
| 1405 |
+
},
|
| 1406 |
+
{
|
| 1407 |
+
"epoch": 1.19,
|
| 1408 |
+
"grad_norm": 1.1418849229812622,
|
| 1409 |
+
"learning_rate": 9.659840645762176e-06,
|
| 1410 |
+
"loss": 0.74,
|
| 1411 |
+
"step": 1900
|
| 1412 |
+
},
|
| 1413 |
+
{
|
| 1414 |
+
"epoch": 1.2,
|
| 1415 |
+
"grad_norm": 0.7200061082839966,
|
| 1416 |
+
"learning_rate": 9.656254459208015e-06,
|
| 1417 |
+
"loss": 0.7295,
|
| 1418 |
+
"step": 1910
|
| 1419 |
+
},
|
| 1420 |
+
{
|
| 1421 |
+
"epoch": 1.2,
|
| 1422 |
+
"grad_norm": 0.9135183095932007,
|
| 1423 |
+
"learning_rate": 9.652650140467094e-06,
|
| 1424 |
+
"loss": 0.651,
|
| 1425 |
+
"step": 1920
|
| 1426 |
+
},
|
| 1427 |
+
{
|
| 1428 |
+
"epoch": 1.21,
|
| 1429 |
+
"grad_norm": 0.9724289774894714,
|
| 1430 |
+
"learning_rate": 9.649027703575193e-06,
|
| 1431 |
+
"loss": 0.7028,
|
| 1432 |
+
"step": 1930
|
| 1433 |
+
},
|
| 1434 |
+
{
|
| 1435 |
+
"epoch": 1.22,
|
| 1436 |
+
"grad_norm": 0.8180338740348816,
|
| 1437 |
+
"learning_rate": 9.645387162638652e-06,
|
| 1438 |
+
"loss": 0.7179,
|
| 1439 |
+
"step": 1940
|
| 1440 |
+
},
|
| 1441 |
+
{
|
| 1442 |
+
"epoch": 1.22,
|
| 1443 |
+
"grad_norm": 1.089158296585083,
|
| 1444 |
+
"learning_rate": 9.641728531834313e-06,
|
| 1445 |
+
"loss": 0.6872,
|
| 1446 |
+
"step": 1950
|
| 1447 |
+
},
|
| 1448 |
+
{
|
| 1449 |
+
"epoch": 1.23,
|
| 1450 |
+
"grad_norm": 1.0048317909240723,
|
| 1451 |
+
"learning_rate": 9.638051825409454e-06,
|
| 1452 |
+
"loss": 0.6991,
|
| 1453 |
+
"step": 1960
|
| 1454 |
+
},
|
| 1455 |
+
{
|
| 1456 |
+
"epoch": 1.24,
|
| 1457 |
+
"grad_norm": 1.1580454111099243,
|
| 1458 |
+
"learning_rate": 9.634357057681749e-06,
|
| 1459 |
+
"loss": 0.7183,
|
| 1460 |
+
"step": 1970
|
| 1461 |
+
},
|
| 1462 |
+
{
|
| 1463 |
+
"epoch": 1.24,
|
| 1464 |
+
"grad_norm": 1.0045746564865112,
|
| 1465 |
+
"learning_rate": 9.630644243039207e-06,
|
| 1466 |
+
"loss": 0.6795,
|
| 1467 |
+
"step": 1980
|
| 1468 |
+
},
|
| 1469 |
+
{
|
| 1470 |
+
"epoch": 1.25,
|
| 1471 |
+
"grad_norm": 0.9629393815994263,
|
| 1472 |
+
"learning_rate": 9.62691339594011e-06,
|
| 1473 |
+
"loss": 0.7075,
|
| 1474 |
+
"step": 1990
|
| 1475 |
+
},
|
| 1476 |
+
{
|
| 1477 |
+
"epoch": 1.25,
|
| 1478 |
+
"grad_norm": 0.946081280708313,
|
| 1479 |
+
"learning_rate": 9.623164530912963e-06,
|
| 1480 |
+
"loss": 0.6999,
|
| 1481 |
+
"step": 2000
|
| 1482 |
+
},
|
| 1483 |
+
{
|
| 1484 |
+
"epoch": 1.25,
|
| 1485 |
+
"eval_loss": 0.7040402293205261,
|
| 1486 |
+
"eval_runtime": 64.9289,
|
| 1487 |
+
"eval_samples_per_second": 69.307,
|
| 1488 |
+
"eval_steps_per_second": 4.343,
|
| 1489 |
+
"step": 2000
|
| 1490 |
+
},
|
| 1491 |
+
{
|
| 1492 |
+
"epoch": 1.26,
|
| 1493 |
+
"grad_norm": 1.0208806991577148,
|
| 1494 |
+
"learning_rate": 9.619397662556434e-06,
|
| 1495 |
+
"loss": 0.6947,
|
| 1496 |
+
"step": 2010
|
| 1497 |
+
},
|
| 1498 |
+
{
|
| 1499 |
+
"epoch": 1.27,
|
| 1500 |
+
"grad_norm": 1.3248392343521118,
|
| 1501 |
+
"learning_rate": 9.615612805539305e-06,
|
| 1502 |
+
"loss": 0.7102,
|
| 1503 |
+
"step": 2020
|
| 1504 |
+
},
|
| 1505 |
+
{
|
| 1506 |
+
"epoch": 1.27,
|
| 1507 |
+
"grad_norm": 0.9521629810333252,
|
| 1508 |
+
"learning_rate": 9.6118099746004e-06,
|
| 1509 |
+
"loss": 0.7068,
|
| 1510 |
+
"step": 2030
|
| 1511 |
+
},
|
| 1512 |
+
{
|
| 1513 |
+
"epoch": 1.28,
|
| 1514 |
+
"grad_norm": 1.129441738128662,
|
| 1515 |
+
"learning_rate": 9.607989184548544e-06,
|
| 1516 |
+
"loss": 0.6528,
|
| 1517 |
+
"step": 2040
|
| 1518 |
+
},
|
| 1519 |
+
{
|
| 1520 |
+
"epoch": 1.29,
|
| 1521 |
+
"grad_norm": 1.2303441762924194,
|
| 1522 |
+
"learning_rate": 9.604150450262488e-06,
|
| 1523 |
+
"loss": 0.6838,
|
| 1524 |
+
"step": 2050
|
| 1525 |
+
},
|
| 1526 |
+
{
|
| 1527 |
+
"epoch": 1.29,
|
| 1528 |
+
"grad_norm": 1.433111310005188,
|
| 1529 |
+
"learning_rate": 9.600293786690873e-06,
|
| 1530 |
+
"loss": 0.6908,
|
| 1531 |
+
"step": 2060
|
| 1532 |
+
},
|
| 1533 |
+
{
|
| 1534 |
+
"epoch": 1.3,
|
| 1535 |
+
"grad_norm": 1.11778724193573,
|
| 1536 |
+
"learning_rate": 9.596419208852152e-06,
|
| 1537 |
+
"loss": 0.7153,
|
| 1538 |
+
"step": 2070
|
| 1539 |
+
},
|
| 1540 |
+
{
|
| 1541 |
+
"epoch": 1.3,
|
| 1542 |
+
"grad_norm": 1.3464716672897339,
|
| 1543 |
+
"learning_rate": 9.592526731834536e-06,
|
| 1544 |
+
"loss": 0.67,
|
| 1545 |
+
"step": 2080
|
| 1546 |
+
},
|
| 1547 |
+
{
|
| 1548 |
+
"epoch": 1.31,
|
| 1549 |
+
"grad_norm": 1.0811423063278198,
|
| 1550 |
+
"learning_rate": 9.588616370795947e-06,
|
| 1551 |
+
"loss": 0.705,
|
| 1552 |
+
"step": 2090
|
| 1553 |
+
},
|
| 1554 |
+
{
|
| 1555 |
+
"epoch": 1.32,
|
| 1556 |
+
"grad_norm": 1.2497215270996094,
|
| 1557 |
+
"learning_rate": 9.584688140963945e-06,
|
| 1558 |
+
"loss": 0.7037,
|
| 1559 |
+
"step": 2100
|
| 1560 |
+
},
|
| 1561 |
+
{
|
| 1562 |
+
"epoch": 1.32,
|
| 1563 |
+
"grad_norm": 1.0369244813919067,
|
| 1564 |
+
"learning_rate": 9.580742057635672e-06,
|
| 1565 |
+
"loss": 0.7199,
|
| 1566 |
+
"step": 2110
|
| 1567 |
+
},
|
| 1568 |
+
{
|
| 1569 |
+
"epoch": 1.33,
|
| 1570 |
+
"grad_norm": 0.8846107125282288,
|
| 1571 |
+
"learning_rate": 9.576778136177798e-06,
|
| 1572 |
+
"loss": 0.7098,
|
| 1573 |
+
"step": 2120
|
| 1574 |
+
},
|
| 1575 |
+
{
|
| 1576 |
+
"epoch": 1.34,
|
| 1577 |
+
"grad_norm": 1.2424838542938232,
|
| 1578 |
+
"learning_rate": 9.572796392026455e-06,
|
| 1579 |
+
"loss": 0.7109,
|
| 1580 |
+
"step": 2130
|
| 1581 |
+
},
|
| 1582 |
+
{
|
| 1583 |
+
"epoch": 1.34,
|
| 1584 |
+
"grad_norm": 1.1349953413009644,
|
| 1585 |
+
"learning_rate": 9.568796840687184e-06,
|
| 1586 |
+
"loss": 0.693,
|
| 1587 |
+
"step": 2140
|
| 1588 |
+
},
|
| 1589 |
+
{
|
| 1590 |
+
"epoch": 1.35,
|
| 1591 |
+
"grad_norm": 0.9105272889137268,
|
| 1592 |
+
"learning_rate": 9.564779497734864e-06,
|
| 1593 |
+
"loss": 0.6679,
|
| 1594 |
+
"step": 2150
|
| 1595 |
+
},
|
| 1596 |
+
{
|
| 1597 |
+
"epoch": 1.36,
|
| 1598 |
+
"grad_norm": 1.021628737449646,
|
| 1599 |
+
"learning_rate": 9.56074437881366e-06,
|
| 1600 |
+
"loss": 0.6573,
|
| 1601 |
+
"step": 2160
|
| 1602 |
+
},
|
| 1603 |
+
{
|
| 1604 |
+
"epoch": 1.36,
|
| 1605 |
+
"grad_norm": 1.1030464172363281,
|
| 1606 |
+
"learning_rate": 9.55669149963696e-06,
|
| 1607 |
+
"loss": 0.705,
|
| 1608 |
+
"step": 2170
|
| 1609 |
+
},
|
| 1610 |
+
{
|
| 1611 |
+
"epoch": 1.37,
|
| 1612 |
+
"grad_norm": 1.1582733392715454,
|
| 1613 |
+
"learning_rate": 9.552620875987312e-06,
|
| 1614 |
+
"loss": 0.6932,
|
| 1615 |
+
"step": 2180
|
| 1616 |
+
},
|
| 1617 |
+
{
|
| 1618 |
+
"epoch": 1.37,
|
| 1619 |
+
"grad_norm": 1.2710620164871216,
|
| 1620 |
+
"learning_rate": 9.548532523716366e-06,
|
| 1621 |
+
"loss": 0.6616,
|
| 1622 |
+
"step": 2190
|
| 1623 |
+
},
|
| 1624 |
+
{
|
| 1625 |
+
"epoch": 1.38,
|
| 1626 |
+
"grad_norm": 1.1528280973434448,
|
| 1627 |
+
"learning_rate": 9.544426458744805e-06,
|
| 1628 |
+
"loss": 0.7182,
|
| 1629 |
+
"step": 2200
|
| 1630 |
+
},
|
| 1631 |
+
{
|
| 1632 |
+
"epoch": 1.38,
|
| 1633 |
+
"eval_loss": 0.7022137641906738,
|
| 1634 |
+
"eval_runtime": 64.965,
|
| 1635 |
+
"eval_samples_per_second": 69.268,
|
| 1636 |
+
"eval_steps_per_second": 4.341,
|
| 1637 |
+
"step": 2200
|
| 1638 |
+
},
|
| 1639 |
+
{
|
| 1640 |
+
"epoch": 1.39,
|
| 1641 |
+
"grad_norm": 1.1455330848693848,
|
| 1642 |
+
"learning_rate": 9.540302697062294e-06,
|
| 1643 |
+
"loss": 0.6878,
|
| 1644 |
+
"step": 2210
|
| 1645 |
+
},
|
| 1646 |
+
{
|
| 1647 |
+
"epoch": 1.39,
|
| 1648 |
+
"grad_norm": 1.4521374702453613,
|
| 1649 |
+
"learning_rate": 9.536161254727407e-06,
|
| 1650 |
+
"loss": 0.6979,
|
| 1651 |
+
"step": 2220
|
| 1652 |
+
},
|
| 1653 |
+
{
|
| 1654 |
+
"epoch": 1.4,
|
| 1655 |
+
"grad_norm": 1.4062340259552002,
|
| 1656 |
+
"learning_rate": 9.532002147867575e-06,
|
| 1657 |
+
"loss": 0.6749,
|
| 1658 |
+
"step": 2230
|
| 1659 |
+
},
|
| 1660 |
+
{
|
| 1661 |
+
"epoch": 1.41,
|
| 1662 |
+
"grad_norm": 1.0267623662948608,
|
| 1663 |
+
"learning_rate": 9.527825392679012e-06,
|
| 1664 |
+
"loss": 0.6987,
|
| 1665 |
+
"step": 2240
|
| 1666 |
+
},
|
| 1667 |
+
{
|
| 1668 |
+
"epoch": 1.41,
|
| 1669 |
+
"grad_norm": 1.0981144905090332,
|
| 1670 |
+
"learning_rate": 9.523631005426658e-06,
|
| 1671 |
+
"loss": 0.6888,
|
| 1672 |
+
"step": 2250
|
| 1673 |
+
},
|
| 1674 |
+
{
|
| 1675 |
+
"epoch": 1.42,
|
| 1676 |
+
"grad_norm": 1.0353021621704102,
|
| 1677 |
+
"learning_rate": 9.51941900244412e-06,
|
| 1678 |
+
"loss": 0.6471,
|
| 1679 |
+
"step": 2260
|
| 1680 |
+
},
|
| 1681 |
+
{
|
| 1682 |
+
"epoch": 1.42,
|
| 1683 |
+
"grad_norm": 1.1088558435440063,
|
| 1684 |
+
"learning_rate": 9.515189400133594e-06,
|
| 1685 |
+
"loss": 0.6689,
|
| 1686 |
+
"step": 2270
|
| 1687 |
+
},
|
| 1688 |
+
{
|
| 1689 |
+
"epoch": 1.43,
|
| 1690 |
+
"grad_norm": 1.1822565793991089,
|
| 1691 |
+
"learning_rate": 9.510942214965819e-06,
|
| 1692 |
+
"loss": 0.7001,
|
| 1693 |
+
"step": 2280
|
| 1694 |
+
},
|
| 1695 |
+
{
|
| 1696 |
+
"epoch": 1.44,
|
| 1697 |
+
"grad_norm": 1.2247307300567627,
|
| 1698 |
+
"learning_rate": 9.506677463480003e-06,
|
| 1699 |
+
"loss": 0.6999,
|
| 1700 |
+
"step": 2290
|
| 1701 |
+
},
|
| 1702 |
+
{
|
| 1703 |
+
"epoch": 1.44,
|
| 1704 |
+
"grad_norm": 1.163528323173523,
|
| 1705 |
+
"learning_rate": 9.50239516228376e-06,
|
| 1706 |
+
"loss": 0.7008,
|
| 1707 |
+
"step": 2300
|
| 1708 |
+
},
|
| 1709 |
+
{
|
| 1710 |
+
"epoch": 1.45,
|
| 1711 |
+
"grad_norm": 1.2677900791168213,
|
| 1712 |
+
"learning_rate": 9.49809532805304e-06,
|
| 1713 |
+
"loss": 0.7122,
|
| 1714 |
+
"step": 2310
|
| 1715 |
+
},
|
| 1716 |
+
{
|
| 1717 |
+
"epoch": 1.46,
|
| 1718 |
+
"grad_norm": 1.1475526094436646,
|
| 1719 |
+
"learning_rate": 9.493777977532072e-06,
|
| 1720 |
+
"loss": 0.7106,
|
| 1721 |
+
"step": 2320
|
| 1722 |
+
},
|
| 1723 |
+
{
|
| 1724 |
+
"epoch": 1.46,
|
| 1725 |
+
"grad_norm": 1.1459851264953613,
|
| 1726 |
+
"learning_rate": 9.489443127533304e-06,
|
| 1727 |
+
"loss": 0.6739,
|
| 1728 |
+
"step": 2330
|
| 1729 |
+
},
|
| 1730 |
+
{
|
| 1731 |
+
"epoch": 1.47,
|
| 1732 |
+
"grad_norm": 1.2973495721817017,
|
| 1733 |
+
"learning_rate": 9.485090794937319e-06,
|
| 1734 |
+
"loss": 0.6888,
|
| 1735 |
+
"step": 2340
|
| 1736 |
+
},
|
| 1737 |
+
{
|
| 1738 |
+
"epoch": 1.47,
|
| 1739 |
+
"grad_norm": 1.0322624444961548,
|
| 1740 |
+
"learning_rate": 9.480720996692783e-06,
|
| 1741 |
+
"loss": 0.6986,
|
| 1742 |
+
"step": 2350
|
| 1743 |
+
},
|
| 1744 |
+
{
|
| 1745 |
+
"epoch": 1.48,
|
| 1746 |
+
"grad_norm": 1.407605767250061,
|
| 1747 |
+
"learning_rate": 9.476333749816382e-06,
|
| 1748 |
+
"loss": 0.7314,
|
| 1749 |
+
"step": 2360
|
| 1750 |
+
},
|
| 1751 |
+
{
|
| 1752 |
+
"epoch": 1.49,
|
| 1753 |
+
"grad_norm": 1.1082048416137695,
|
| 1754 |
+
"learning_rate": 9.47192907139274e-06,
|
| 1755 |
+
"loss": 0.6602,
|
| 1756 |
+
"step": 2370
|
| 1757 |
+
},
|
| 1758 |
+
{
|
| 1759 |
+
"epoch": 1.49,
|
| 1760 |
+
"grad_norm": 1.02568519115448,
|
| 1761 |
+
"learning_rate": 9.46750697857437e-06,
|
| 1762 |
+
"loss": 0.6454,
|
| 1763 |
+
"step": 2380
|
| 1764 |
+
},
|
| 1765 |
+
{
|
| 1766 |
+
"epoch": 1.5,
|
| 1767 |
+
"grad_norm": 1.12267005443573,
|
| 1768 |
+
"learning_rate": 9.463067488581598e-06,
|
| 1769 |
+
"loss": 0.6499,
|
| 1770 |
+
"step": 2390
|
| 1771 |
+
},
|
| 1772 |
+
{
|
| 1773 |
+
"epoch": 1.51,
|
| 1774 |
+
"grad_norm": 1.0023943185806274,
|
| 1775 |
+
"learning_rate": 9.45861061870249e-06,
|
| 1776 |
+
"loss": 0.7267,
|
| 1777 |
+
"step": 2400
|
| 1778 |
+
},
|
| 1779 |
+
{
|
| 1780 |
+
"epoch": 1.51,
|
| 1781 |
+
"eval_loss": 0.6993948817253113,
|
| 1782 |
+
"eval_runtime": 64.9272,
|
| 1783 |
+
"eval_samples_per_second": 69.308,
|
| 1784 |
+
"eval_steps_per_second": 4.343,
|
| 1785 |
+
"step": 2400
|
| 1786 |
+
},
|
| 1787 |
+
{
|
| 1788 |
+
"epoch": 1.51,
|
| 1789 |
+
"grad_norm": 1.2597460746765137,
|
| 1790 |
+
"learning_rate": 9.454136386292804e-06,
|
| 1791 |
+
"loss": 0.6934,
|
| 1792 |
+
"step": 2410
|
| 1793 |
+
},
|
| 1794 |
+
{
|
| 1795 |
+
"epoch": 1.52,
|
| 1796 |
+
"grad_norm": 1.293137788772583,
|
| 1797 |
+
"learning_rate": 9.449644808775902e-06,
|
| 1798 |
+
"loss": 0.7095,
|
| 1799 |
+
"step": 2420
|
| 1800 |
+
},
|
| 1801 |
+
{
|
| 1802 |
+
"epoch": 1.52,
|
| 1803 |
+
"grad_norm": 1.0400352478027344,
|
| 1804 |
+
"learning_rate": 9.445135903642693e-06,
|
| 1805 |
+
"loss": 0.6626,
|
| 1806 |
+
"step": 2430
|
| 1807 |
+
},
|
| 1808 |
+
{
|
| 1809 |
+
"epoch": 1.53,
|
| 1810 |
+
"grad_norm": 1.0873581171035767,
|
| 1811 |
+
"learning_rate": 9.440609688451561e-06,
|
| 1812 |
+
"loss": 0.6513,
|
| 1813 |
+
"step": 2440
|
| 1814 |
+
},
|
| 1815 |
+
{
|
| 1816 |
+
"epoch": 1.54,
|
| 1817 |
+
"grad_norm": 1.0420424938201904,
|
| 1818 |
+
"learning_rate": 9.4360661808283e-06,
|
| 1819 |
+
"loss": 0.711,
|
| 1820 |
+
"step": 2450
|
| 1821 |
+
},
|
| 1822 |
+
{
|
| 1823 |
+
"epoch": 1.54,
|
| 1824 |
+
"grad_norm": 1.3502894639968872,
|
| 1825 |
+
"learning_rate": 9.431505398466045e-06,
|
| 1826 |
+
"loss": 0.6991,
|
| 1827 |
+
"step": 2460
|
| 1828 |
+
},
|
| 1829 |
+
{
|
| 1830 |
+
"epoch": 1.55,
|
| 1831 |
+
"grad_norm": 1.3502726554870605,
|
| 1832 |
+
"learning_rate": 9.426927359125195e-06,
|
| 1833 |
+
"loss": 0.7073,
|
| 1834 |
+
"step": 2470
|
| 1835 |
+
},
|
| 1836 |
+
{
|
| 1837 |
+
"epoch": 1.56,
|
| 1838 |
+
"grad_norm": 1.2768748998641968,
|
| 1839 |
+
"learning_rate": 9.422332080633361e-06,
|
| 1840 |
+
"loss": 0.6557,
|
| 1841 |
+
"step": 2480
|
| 1842 |
+
},
|
| 1843 |
+
{
|
| 1844 |
+
"epoch": 1.56,
|
| 1845 |
+
"grad_norm": 1.1925798654556274,
|
| 1846 |
+
"learning_rate": 9.417719580885275e-06,
|
| 1847 |
+
"loss": 0.6786,
|
| 1848 |
+
"step": 2490
|
| 1849 |
+
},
|
| 1850 |
+
{
|
| 1851 |
+
"epoch": 1.57,
|
| 1852 |
+
"grad_norm": 0.9290177822113037,
|
| 1853 |
+
"learning_rate": 9.413089877842735e-06,
|
| 1854 |
+
"loss": 0.6159,
|
| 1855 |
+
"step": 2500
|
| 1856 |
+
},
|
| 1857 |
+
{
|
| 1858 |
+
"epoch": 1.57,
|
| 1859 |
+
"grad_norm": 1.3553310632705688,
|
| 1860 |
+
"learning_rate": 9.408442989534536e-06,
|
| 1861 |
+
"loss": 0.7341,
|
| 1862 |
+
"step": 2510
|
| 1863 |
+
},
|
| 1864 |
+
{
|
| 1865 |
+
"epoch": 1.58,
|
| 1866 |
+
"grad_norm": 0.9777106642723083,
|
| 1867 |
+
"learning_rate": 9.403778934056392e-06,
|
| 1868 |
+
"loss": 0.6737,
|
| 1869 |
+
"step": 2520
|
| 1870 |
+
},
|
| 1871 |
+
{
|
| 1872 |
+
"epoch": 1.59,
|
| 1873 |
+
"grad_norm": 1.47153902053833,
|
| 1874 |
+
"learning_rate": 9.399097729570865e-06,
|
| 1875 |
+
"loss": 0.6832,
|
| 1876 |
+
"step": 2530
|
| 1877 |
+
},
|
| 1878 |
+
{
|
| 1879 |
+
"epoch": 1.59,
|
| 1880 |
+
"grad_norm": 1.2370259761810303,
|
| 1881 |
+
"learning_rate": 9.394399394307303e-06,
|
| 1882 |
+
"loss": 0.6691,
|
| 1883 |
+
"step": 2540
|
| 1884 |
+
},
|
| 1885 |
+
{
|
| 1886 |
+
"epoch": 1.6,
|
| 1887 |
+
"grad_norm": 1.2009457349777222,
|
| 1888 |
+
"learning_rate": 9.38968394656176e-06,
|
| 1889 |
+
"loss": 0.7072,
|
| 1890 |
+
"step": 2550
|
| 1891 |
+
},
|
| 1892 |
+
{
|
| 1893 |
+
"epoch": 1.61,
|
| 1894 |
+
"grad_norm": 1.095410704612732,
|
| 1895 |
+
"learning_rate": 9.384951404696933e-06,
|
| 1896 |
+
"loss": 0.7068,
|
| 1897 |
+
"step": 2560
|
| 1898 |
+
},
|
| 1899 |
+
{
|
| 1900 |
+
"epoch": 1.61,
|
| 1901 |
+
"grad_norm": 1.0805617570877075,
|
| 1902 |
+
"learning_rate": 9.380201787142085e-06,
|
| 1903 |
+
"loss": 0.6476,
|
| 1904 |
+
"step": 2570
|
| 1905 |
+
},
|
| 1906 |
+
{
|
| 1907 |
+
"epoch": 1.62,
|
| 1908 |
+
"grad_norm": 1.3433113098144531,
|
| 1909 |
+
"learning_rate": 9.37543511239297e-06,
|
| 1910 |
+
"loss": 0.6805,
|
| 1911 |
+
"step": 2580
|
| 1912 |
+
},
|
| 1913 |
+
{
|
| 1914 |
+
"epoch": 1.62,
|
| 1915 |
+
"grad_norm": 1.3151830434799194,
|
| 1916 |
+
"learning_rate": 9.370651399011769e-06,
|
| 1917 |
+
"loss": 0.6887,
|
| 1918 |
+
"step": 2590
|
| 1919 |
+
},
|
| 1920 |
+
{
|
| 1921 |
+
"epoch": 1.63,
|
| 1922 |
+
"grad_norm": 1.3367606401443481,
|
| 1923 |
+
"learning_rate": 9.365850665627016e-06,
|
| 1924 |
+
"loss": 0.6912,
|
| 1925 |
+
"step": 2600
|
| 1926 |
+
},
|
| 1927 |
+
{
|
| 1928 |
+
"epoch": 1.63,
|
| 1929 |
+
"eval_loss": 0.6971801519393921,
|
| 1930 |
+
"eval_runtime": 65.0021,
|
| 1931 |
+
"eval_samples_per_second": 69.229,
|
| 1932 |
+
"eval_steps_per_second": 4.338,
|
| 1933 |
+
"step": 2600
|
| 1934 |
+
},
|
| 1935 |
+
{
|
| 1936 |
+
"epoch": 1.64,
|
| 1937 |
+
"grad_norm": 1.3351305723190308,
|
| 1938 |
+
"learning_rate": 9.36103293093352e-06,
|
| 1939 |
+
"loss": 0.6479,
|
| 1940 |
+
"step": 2610
|
| 1941 |
+
},
|
| 1942 |
+
{
|
| 1943 |
+
"epoch": 1.64,
|
| 1944 |
+
"grad_norm": 1.3986787796020508,
|
| 1945 |
+
"learning_rate": 9.356198213692297e-06,
|
| 1946 |
+
"loss": 0.6788,
|
| 1947 |
+
"step": 2620
|
| 1948 |
+
},
|
| 1949 |
+
{
|
| 1950 |
+
"epoch": 1.65,
|
| 1951 |
+
"grad_norm": 1.0550477504730225,
|
| 1952 |
+
"learning_rate": 9.351346532730499e-06,
|
| 1953 |
+
"loss": 0.6481,
|
| 1954 |
+
"step": 2630
|
| 1955 |
+
},
|
| 1956 |
+
{
|
| 1957 |
+
"epoch": 1.66,
|
| 1958 |
+
"grad_norm": 1.275985836982727,
|
| 1959 |
+
"learning_rate": 9.346477906941331e-06,
|
| 1960 |
+
"loss": 0.6893,
|
| 1961 |
+
"step": 2640
|
| 1962 |
+
},
|
| 1963 |
+
{
|
| 1964 |
+
"epoch": 1.66,
|
| 1965 |
+
"grad_norm": 1.441588044166565,
|
| 1966 |
+
"learning_rate": 9.341592355283986e-06,
|
| 1967 |
+
"loss": 0.6784,
|
| 1968 |
+
"step": 2650
|
| 1969 |
+
},
|
| 1970 |
+
{
|
| 1971 |
+
"epoch": 1.67,
|
| 1972 |
+
"grad_norm": 1.2504793405532837,
|
| 1973 |
+
"learning_rate": 9.336689896783575e-06,
|
| 1974 |
+
"loss": 0.6834,
|
| 1975 |
+
"step": 2660
|
| 1976 |
+
},
|
| 1977 |
+
{
|
| 1978 |
+
"epoch": 1.68,
|
| 1979 |
+
"grad_norm": 1.2592806816101074,
|
| 1980 |
+
"learning_rate": 9.331770550531037e-06,
|
| 1981 |
+
"loss": 0.6701,
|
| 1982 |
+
"step": 2670
|
| 1983 |
+
},
|
| 1984 |
+
{
|
| 1985 |
+
"epoch": 1.68,
|
| 1986 |
+
"grad_norm": 1.494611382484436,
|
| 1987 |
+
"learning_rate": 9.32683433568308e-06,
|
| 1988 |
+
"loss": 0.6691,
|
| 1989 |
+
"step": 2680
|
| 1990 |
+
},
|
| 1991 |
+
{
|
| 1992 |
+
"epoch": 1.69,
|
| 1993 |
+
"grad_norm": 1.2938275337219238,
|
| 1994 |
+
"learning_rate": 9.321881271462104e-06,
|
| 1995 |
+
"loss": 0.6818,
|
| 1996 |
+
"step": 2690
|
| 1997 |
+
},
|
| 1998 |
+
{
|
| 1999 |
+
"epoch": 1.69,
|
| 2000 |
+
"grad_norm": 1.5548397302627563,
|
| 2001 |
+
"learning_rate": 9.316911377156116e-06,
|
| 2002 |
+
"loss": 0.6852,
|
| 2003 |
+
"step": 2700
|
| 2004 |
+
},
|
| 2005 |
+
{
|
| 2006 |
+
"epoch": 1.7,
|
| 2007 |
+
"grad_norm": 1.2488983869552612,
|
| 2008 |
+
"learning_rate": 9.31192467211867e-06,
|
| 2009 |
+
"loss": 0.6653,
|
| 2010 |
+
"step": 2710
|
| 2011 |
+
},
|
| 2012 |
+
{
|
| 2013 |
+
"epoch": 1.71,
|
| 2014 |
+
"grad_norm": 1.3493934869766235,
|
| 2015 |
+
"learning_rate": 9.306921175768776e-06,
|
| 2016 |
+
"loss": 0.6671,
|
| 2017 |
+
"step": 2720
|
| 2018 |
+
},
|
| 2019 |
+
{
|
| 2020 |
+
"epoch": 1.71,
|
| 2021 |
+
"grad_norm": 1.388487696647644,
|
| 2022 |
+
"learning_rate": 9.301900907590836e-06,
|
| 2023 |
+
"loss": 0.7066,
|
| 2024 |
+
"step": 2730
|
| 2025 |
+
},
|
| 2026 |
+
{
|
| 2027 |
+
"epoch": 1.72,
|
| 2028 |
+
"grad_norm": 1.2521592378616333,
|
| 2029 |
+
"learning_rate": 9.296863887134561e-06,
|
| 2030 |
+
"loss": 0.7326,
|
| 2031 |
+
"step": 2740
|
| 2032 |
+
},
|
| 2033 |
+
{
|
| 2034 |
+
"epoch": 1.73,
|
| 2035 |
+
"grad_norm": 1.0157365798950195,
|
| 2036 |
+
"learning_rate": 9.291810134014904e-06,
|
| 2037 |
+
"loss": 0.6758,
|
| 2038 |
+
"step": 2750
|
| 2039 |
+
},
|
| 2040 |
+
{
|
| 2041 |
+
"epoch": 1.73,
|
| 2042 |
+
"grad_norm": 1.0712261199951172,
|
| 2043 |
+
"learning_rate": 9.286739667911973e-06,
|
| 2044 |
+
"loss": 0.6645,
|
| 2045 |
+
"step": 2760
|
| 2046 |
+
},
|
| 2047 |
+
{
|
| 2048 |
+
"epoch": 1.74,
|
| 2049 |
+
"grad_norm": 1.5114517211914062,
|
| 2050 |
+
"learning_rate": 9.281652508570957e-06,
|
| 2051 |
+
"loss": 0.6968,
|
| 2052 |
+
"step": 2770
|
| 2053 |
+
},
|
| 2054 |
+
{
|
| 2055 |
+
"epoch": 1.74,
|
| 2056 |
+
"grad_norm": 1.3408139944076538,
|
| 2057 |
+
"learning_rate": 9.27654867580206e-06,
|
| 2058 |
+
"loss": 0.6718,
|
| 2059 |
+
"step": 2780
|
| 2060 |
+
},
|
| 2061 |
+
{
|
| 2062 |
+
"epoch": 1.75,
|
| 2063 |
+
"grad_norm": 1.3024680614471436,
|
| 2064 |
+
"learning_rate": 9.271428189480405e-06,
|
| 2065 |
+
"loss": 0.6915,
|
| 2066 |
+
"step": 2790
|
| 2067 |
+
},
|
| 2068 |
+
{
|
| 2069 |
+
"epoch": 1.76,
|
| 2070 |
+
"grad_norm": 1.3444178104400635,
|
| 2071 |
+
"learning_rate": 9.266291069545972e-06,
|
| 2072 |
+
"loss": 0.6821,
|
| 2073 |
+
"step": 2800
|
| 2074 |
+
},
|
| 2075 |
+
{
|
| 2076 |
+
"epoch": 1.76,
|
| 2077 |
+
"eval_loss": 0.6953641176223755,
|
| 2078 |
+
"eval_runtime": 65.0162,
|
| 2079 |
+
"eval_samples_per_second": 69.213,
|
| 2080 |
+
"eval_steps_per_second": 4.337,
|
| 2081 |
+
"step": 2800
|
| 2082 |
+
},
|
| 2083 |
+
{
|
| 2084 |
+
"epoch": 1.76,
|
| 2085 |
+
"grad_norm": 1.5429843664169312,
|
| 2086 |
+
"learning_rate": 9.261137336003511e-06,
|
| 2087 |
+
"loss": 0.666,
|
| 2088 |
+
"step": 2810
|
| 2089 |
+
},
|
| 2090 |
+
{
|
| 2091 |
+
"epoch": 1.77,
|
| 2092 |
+
"grad_norm": 1.143649697303772,
|
| 2093 |
+
"learning_rate": 9.255967008922475e-06,
|
| 2094 |
+
"loss": 0.6414,
|
| 2095 |
+
"step": 2820
|
| 2096 |
+
},
|
| 2097 |
+
{
|
| 2098 |
+
"epoch": 1.78,
|
| 2099 |
+
"grad_norm": 1.2989628314971924,
|
| 2100 |
+
"learning_rate": 9.250780108436926e-06,
|
| 2101 |
+
"loss": 0.7321,
|
| 2102 |
+
"step": 2830
|
| 2103 |
+
},
|
| 2104 |
+
{
|
| 2105 |
+
"epoch": 1.78,
|
| 2106 |
+
"grad_norm": 1.4191828966140747,
|
| 2107 |
+
"learning_rate": 9.245576654745471e-06,
|
| 2108 |
+
"loss": 0.735,
|
| 2109 |
+
"step": 2840
|
| 2110 |
+
},
|
| 2111 |
+
{
|
| 2112 |
+
"epoch": 1.79,
|
| 2113 |
+
"grad_norm": 1.5203850269317627,
|
| 2114 |
+
"learning_rate": 9.24035666811118e-06,
|
| 2115 |
+
"loss": 0.6809,
|
| 2116 |
+
"step": 2850
|
| 2117 |
+
},
|
| 2118 |
+
{
|
| 2119 |
+
"epoch": 1.79,
|
| 2120 |
+
"grad_norm": 1.3680098056793213,
|
| 2121 |
+
"learning_rate": 9.235120168861495e-06,
|
| 2122 |
+
"loss": 0.6378,
|
| 2123 |
+
"step": 2860
|
| 2124 |
+
},
|
| 2125 |
+
{
|
| 2126 |
+
"epoch": 1.8,
|
| 2127 |
+
"grad_norm": 1.346763014793396,
|
| 2128 |
+
"learning_rate": 9.229867177388172e-06,
|
| 2129 |
+
"loss": 0.6648,
|
| 2130 |
+
"step": 2870
|
| 2131 |
+
},
|
| 2132 |
+
{
|
| 2133 |
+
"epoch": 1.81,
|
| 2134 |
+
"grad_norm": 1.0098768472671509,
|
| 2135 |
+
"learning_rate": 9.224597714147186e-06,
|
| 2136 |
+
"loss": 0.6681,
|
| 2137 |
+
"step": 2880
|
| 2138 |
+
},
|
| 2139 |
+
{
|
| 2140 |
+
"epoch": 1.81,
|
| 2141 |
+
"grad_norm": 1.3174008131027222,
|
| 2142 |
+
"learning_rate": 9.219311799658652e-06,
|
| 2143 |
+
"loss": 0.6752,
|
| 2144 |
+
"step": 2890
|
| 2145 |
+
},
|
| 2146 |
+
{
|
| 2147 |
+
"epoch": 1.82,
|
| 2148 |
+
"grad_norm": 1.692084789276123,
|
| 2149 |
+
"learning_rate": 9.214009454506754e-06,
|
| 2150 |
+
"loss": 0.6427,
|
| 2151 |
+
"step": 2900
|
| 2152 |
+
},
|
| 2153 |
+
{
|
| 2154 |
+
"epoch": 1.83,
|
| 2155 |
+
"grad_norm": 1.3471505641937256,
|
| 2156 |
+
"learning_rate": 9.208690699339656e-06,
|
| 2157 |
+
"loss": 0.6763,
|
| 2158 |
+
"step": 2910
|
| 2159 |
+
},
|
| 2160 |
+
{
|
| 2161 |
+
"epoch": 1.83,
|
| 2162 |
+
"grad_norm": 1.202491283416748,
|
| 2163 |
+
"learning_rate": 9.203355554869428e-06,
|
| 2164 |
+
"loss": 0.6935,
|
| 2165 |
+
"step": 2920
|
| 2166 |
+
},
|
| 2167 |
+
{
|
| 2168 |
+
"epoch": 1.84,
|
| 2169 |
+
"grad_norm": 1.7211599349975586,
|
| 2170 |
+
"learning_rate": 9.198004041871962e-06,
|
| 2171 |
+
"loss": 0.7012,
|
| 2172 |
+
"step": 2930
|
| 2173 |
+
},
|
| 2174 |
+
{
|
| 2175 |
+
"epoch": 1.84,
|
| 2176 |
+
"grad_norm": 1.336504578590393,
|
| 2177 |
+
"learning_rate": 9.192636181186887e-06,
|
| 2178 |
+
"loss": 0.6713,
|
| 2179 |
+
"step": 2940
|
| 2180 |
+
},
|
| 2181 |
+
{
|
| 2182 |
+
"epoch": 1.85,
|
| 2183 |
+
"grad_norm": 1.2259244918823242,
|
| 2184 |
+
"learning_rate": 9.1872519937175e-06,
|
| 2185 |
+
"loss": 0.6344,
|
| 2186 |
+
"step": 2950
|
| 2187 |
+
},
|
| 2188 |
+
{
|
| 2189 |
+
"epoch": 1.86,
|
| 2190 |
+
"grad_norm": 1.3948123455047607,
|
| 2191 |
+
"learning_rate": 9.181851500430672e-06,
|
| 2192 |
+
"loss": 0.6699,
|
| 2193 |
+
"step": 2960
|
| 2194 |
+
},
|
| 2195 |
+
{
|
| 2196 |
+
"epoch": 1.86,
|
| 2197 |
+
"grad_norm": 1.2859784364700317,
|
| 2198 |
+
"learning_rate": 9.176434722356772e-06,
|
| 2199 |
+
"loss": 0.7029,
|
| 2200 |
+
"step": 2970
|
| 2201 |
+
},
|
| 2202 |
+
{
|
| 2203 |
+
"epoch": 1.87,
|
| 2204 |
+
"grad_norm": 1.1549146175384521,
|
| 2205 |
+
"learning_rate": 9.17100168058959e-06,
|
| 2206 |
+
"loss": 0.6491,
|
| 2207 |
+
"step": 2980
|
| 2208 |
+
},
|
| 2209 |
+
{
|
| 2210 |
+
"epoch": 1.88,
|
| 2211 |
+
"grad_norm": 1.575208306312561,
|
| 2212 |
+
"learning_rate": 9.165552396286236e-06,
|
| 2213 |
+
"loss": 0.6722,
|
| 2214 |
+
"step": 2990
|
| 2215 |
+
},
|
| 2216 |
+
{
|
| 2217 |
+
"epoch": 1.88,
|
| 2218 |
+
"grad_norm": 1.6159918308258057,
|
| 2219 |
+
"learning_rate": 9.160086890667086e-06,
|
| 2220 |
+
"loss": 0.7104,
|
| 2221 |
+
"step": 3000
|
| 2222 |
+
},
|
| 2223 |
+
{
|
| 2224 |
+
"epoch": 1.88,
|
| 2225 |
+
"eval_loss": 0.6943792104721069,
|
| 2226 |
+
"eval_runtime": 65.034,
|
| 2227 |
+
"eval_samples_per_second": 69.195,
|
| 2228 |
+
"eval_steps_per_second": 4.336,
|
| 2229 |
+
"step": 3000
|
| 2230 |
+
},
|
| 2231 |
+
{
|
| 2232 |
+
"epoch": 1.89,
|
| 2233 |
+
"grad_norm": 1.247308611869812,
|
| 2234 |
+
"learning_rate": 9.154605185015678e-06,
|
| 2235 |
+
"loss": 0.7042,
|
| 2236 |
+
"step": 3010
|
| 2237 |
+
},
|
| 2238 |
+
{
|
| 2239 |
+
"epoch": 1.89,
|
| 2240 |
+
"grad_norm": 1.540644884109497,
|
| 2241 |
+
"learning_rate": 9.14910730067863e-06,
|
| 2242 |
+
"loss": 0.6208,
|
| 2243 |
+
"step": 3020
|
| 2244 |
+
},
|
| 2245 |
+
{
|
| 2246 |
+
"epoch": 1.9,
|
| 2247 |
+
"grad_norm": 1.4479825496673584,
|
| 2248 |
+
"learning_rate": 9.143593259065573e-06,
|
| 2249 |
+
"loss": 0.6721,
|
| 2250 |
+
"step": 3030
|
| 2251 |
+
},
|
| 2252 |
+
{
|
| 2253 |
+
"epoch": 1.91,
|
| 2254 |
+
"grad_norm": 1.5486655235290527,
|
| 2255 |
+
"learning_rate": 9.138063081649052e-06,
|
| 2256 |
+
"loss": 0.6328,
|
| 2257 |
+
"step": 3040
|
| 2258 |
+
},
|
| 2259 |
+
{
|
| 2260 |
+
"epoch": 1.91,
|
| 2261 |
+
"grad_norm": 1.0703155994415283,
|
| 2262 |
+
"learning_rate": 9.132516789964443e-06,
|
| 2263 |
+
"loss": 0.6564,
|
| 2264 |
+
"step": 3050
|
| 2265 |
+
},
|
| 2266 |
+
{
|
| 2267 |
+
"epoch": 1.92,
|
| 2268 |
+
"grad_norm": 1.2725510597229004,
|
| 2269 |
+
"learning_rate": 9.126954405609882e-06,
|
| 2270 |
+
"loss": 0.6782,
|
| 2271 |
+
"step": 3060
|
| 2272 |
+
},
|
| 2273 |
+
{
|
| 2274 |
+
"epoch": 1.93,
|
| 2275 |
+
"grad_norm": 1.328399419784546,
|
| 2276 |
+
"learning_rate": 9.121375950246165e-06,
|
| 2277 |
+
"loss": 0.6686,
|
| 2278 |
+
"step": 3070
|
| 2279 |
+
},
|
| 2280 |
+
{
|
| 2281 |
+
"epoch": 1.93,
|
| 2282 |
+
"grad_norm": 1.2014747858047485,
|
| 2283 |
+
"learning_rate": 9.115781445596676e-06,
|
| 2284 |
+
"loss": 0.6445,
|
| 2285 |
+
"step": 3080
|
| 2286 |
+
},
|
| 2287 |
+
{
|
| 2288 |
+
"epoch": 1.94,
|
| 2289 |
+
"grad_norm": 1.3578124046325684,
|
| 2290 |
+
"learning_rate": 9.110170913447294e-06,
|
| 2291 |
+
"loss": 0.6306,
|
| 2292 |
+
"step": 3090
|
| 2293 |
+
},
|
| 2294 |
+
{
|
| 2295 |
+
"epoch": 1.94,
|
| 2296 |
+
"grad_norm": 1.3624286651611328,
|
| 2297 |
+
"learning_rate": 9.104544375646314e-06,
|
| 2298 |
+
"loss": 0.6465,
|
| 2299 |
+
"step": 3100
|
| 2300 |
+
},
|
| 2301 |
+
{
|
| 2302 |
+
"epoch": 1.95,
|
| 2303 |
+
"grad_norm": 1.709974765777588,
|
| 2304 |
+
"learning_rate": 9.098901854104359e-06,
|
| 2305 |
+
"loss": 0.6985,
|
| 2306 |
+
"step": 3110
|
| 2307 |
+
},
|
| 2308 |
+
{
|
| 2309 |
+
"epoch": 1.96,
|
| 2310 |
+
"grad_norm": 1.3302754163742065,
|
| 2311 |
+
"learning_rate": 9.09324337079429e-06,
|
| 2312 |
+
"loss": 0.7272,
|
| 2313 |
+
"step": 3120
|
| 2314 |
+
},
|
| 2315 |
+
{
|
| 2316 |
+
"epoch": 1.96,
|
| 2317 |
+
"grad_norm": 1.2946594953536987,
|
| 2318 |
+
"learning_rate": 9.08756894775114e-06,
|
| 2319 |
+
"loss": 0.6632,
|
| 2320 |
+
"step": 3130
|
| 2321 |
+
},
|
| 2322 |
+
{
|
| 2323 |
+
"epoch": 1.97,
|
| 2324 |
+
"grad_norm": 1.2699226140975952,
|
| 2325 |
+
"learning_rate": 9.081878607071996e-06,
|
| 2326 |
+
"loss": 0.6996,
|
| 2327 |
+
"step": 3140
|
| 2328 |
+
},
|
| 2329 |
+
{
|
| 2330 |
+
"epoch": 1.98,
|
| 2331 |
+
"grad_norm": 1.4561275243759155,
|
| 2332 |
+
"learning_rate": 9.076172370915944e-06,
|
| 2333 |
+
"loss": 0.734,
|
| 2334 |
+
"step": 3150
|
| 2335 |
+
},
|
| 2336 |
+
{
|
| 2337 |
+
"epoch": 1.98,
|
| 2338 |
+
"grad_norm": 1.4393534660339355,
|
| 2339 |
+
"learning_rate": 9.07045026150396e-06,
|
| 2340 |
+
"loss": 0.6578,
|
| 2341 |
+
"step": 3160
|
| 2342 |
+
},
|
| 2343 |
+
{
|
| 2344 |
+
"epoch": 1.99,
|
| 2345 |
+
"grad_norm": 1.4745630025863647,
|
| 2346 |
+
"learning_rate": 9.064712301118842e-06,
|
| 2347 |
+
"loss": 0.6527,
|
| 2348 |
+
"step": 3170
|
| 2349 |
+
},
|
| 2350 |
+
{
|
| 2351 |
+
"epoch": 1.99,
|
| 2352 |
+
"grad_norm": 1.1444178819656372,
|
| 2353 |
+
"learning_rate": 9.058958512105104e-06,
|
| 2354 |
+
"loss": 0.6487,
|
| 2355 |
+
"step": 3180
|
| 2356 |
+
},
|
| 2357 |
+
{
|
| 2358 |
+
"epoch": 2.0,
|
| 2359 |
+
"grad_norm": 1.433406114578247,
|
| 2360 |
+
"learning_rate": 9.053188916868912e-06,
|
| 2361 |
+
"loss": 0.7011,
|
| 2362 |
+
"step": 3190
|
| 2363 |
+
},
|
| 2364 |
+
{
|
| 2365 |
+
"epoch": 2.01,
|
| 2366 |
+
"grad_norm": 1.218345046043396,
|
| 2367 |
+
"learning_rate": 9.04740353787797e-06,
|
| 2368 |
+
"loss": 0.6222,
|
| 2369 |
+
"step": 3200
|
| 2370 |
+
},
|
| 2371 |
+
{
|
| 2372 |
+
"epoch": 2.01,
|
| 2373 |
+
"eval_loss": 0.693417489528656,
|
| 2374 |
+
"eval_runtime": 65.021,
|
| 2375 |
+
"eval_samples_per_second": 69.208,
|
| 2376 |
+
"eval_steps_per_second": 4.337,
|
| 2377 |
+
"step": 3200
|
| 2378 |
+
},
|
| 2379 |
+
{
|
| 2380 |
+
"epoch": 2.01,
|
| 2381 |
+
"grad_norm": 1.5473078489303589,
|
| 2382 |
+
"learning_rate": 9.041602397661459e-06,
|
| 2383 |
+
"loss": 0.6396,
|
| 2384 |
+
"step": 3210
|
| 2385 |
+
},
|
| 2386 |
+
{
|
| 2387 |
+
"epoch": 2.02,
|
| 2388 |
+
"grad_norm": 1.3116644620895386,
|
| 2389 |
+
"learning_rate": 9.035785518809928e-06,
|
| 2390 |
+
"loss": 0.6582,
|
| 2391 |
+
"step": 3220
|
| 2392 |
+
},
|
| 2393 |
+
{
|
| 2394 |
+
"epoch": 2.03,
|
| 2395 |
+
"grad_norm": 1.7744321823120117,
|
| 2396 |
+
"learning_rate": 9.029952923975217e-06,
|
| 2397 |
+
"loss": 0.6517,
|
| 2398 |
+
"step": 3230
|
| 2399 |
+
},
|
| 2400 |
+
{
|
| 2401 |
+
"epoch": 2.03,
|
| 2402 |
+
"grad_norm": 1.5516449213027954,
|
| 2403 |
+
"learning_rate": 9.024104635870368e-06,
|
| 2404 |
+
"loss": 0.6465,
|
| 2405 |
+
"step": 3240
|
| 2406 |
+
},
|
| 2407 |
+
{
|
| 2408 |
+
"epoch": 2.04,
|
| 2409 |
+
"grad_norm": 1.4612600803375244,
|
| 2410 |
+
"learning_rate": 9.018240677269532e-06,
|
| 2411 |
+
"loss": 0.6215,
|
| 2412 |
+
"step": 3250
|
| 2413 |
+
},
|
| 2414 |
+
{
|
| 2415 |
+
"epoch": 2.05,
|
| 2416 |
+
"grad_norm": 1.7983644008636475,
|
| 2417 |
+
"learning_rate": 9.012361071007892e-06,
|
| 2418 |
+
"loss": 0.6609,
|
| 2419 |
+
"step": 3260
|
| 2420 |
+
},
|
| 2421 |
+
{
|
| 2422 |
+
"epoch": 2.05,
|
| 2423 |
+
"grad_norm": 1.6382901668548584,
|
| 2424 |
+
"learning_rate": 9.00646583998155e-06,
|
| 2425 |
+
"loss": 0.6608,
|
| 2426 |
+
"step": 3270
|
| 2427 |
+
},
|
| 2428 |
+
{
|
| 2429 |
+
"epoch": 2.06,
|
| 2430 |
+
"grad_norm": 1.6763097047805786,
|
| 2431 |
+
"learning_rate": 9.000555007147469e-06,
|
| 2432 |
+
"loss": 0.6222,
|
| 2433 |
+
"step": 3280
|
| 2434 |
+
},
|
| 2435 |
+
{
|
| 2436 |
+
"epoch": 2.06,
|
| 2437 |
+
"grad_norm": 1.3221015930175781,
|
| 2438 |
+
"learning_rate": 8.994628595523358e-06,
|
| 2439 |
+
"loss": 0.6363,
|
| 2440 |
+
"step": 3290
|
| 2441 |
+
},
|
| 2442 |
+
{
|
| 2443 |
+
"epoch": 2.07,
|
| 2444 |
+
"grad_norm": 1.5837445259094238,
|
| 2445 |
+
"learning_rate": 8.988686628187597e-06,
|
| 2446 |
+
"loss": 0.6364,
|
| 2447 |
+
"step": 3300
|
| 2448 |
+
},
|
| 2449 |
+
{
|
| 2450 |
+
"epoch": 2.08,
|
| 2451 |
+
"grad_norm": 1.4271923303604126,
|
| 2452 |
+
"learning_rate": 8.98272912827914e-06,
|
| 2453 |
+
"loss": 0.6211,
|
| 2454 |
+
"step": 3310
|
| 2455 |
+
},
|
| 2456 |
+
{
|
| 2457 |
+
"epoch": 2.08,
|
| 2458 |
+
"grad_norm": 1.599827527999878,
|
| 2459 |
+
"learning_rate": 8.97675611899743e-06,
|
| 2460 |
+
"loss": 0.6326,
|
| 2461 |
+
"step": 3320
|
| 2462 |
+
},
|
| 2463 |
+
{
|
| 2464 |
+
"epoch": 2.09,
|
| 2465 |
+
"grad_norm": 1.6661384105682373,
|
| 2466 |
+
"learning_rate": 8.970767623602299e-06,
|
| 2467 |
+
"loss": 0.7006,
|
| 2468 |
+
"step": 3330
|
| 2469 |
+
},
|
| 2470 |
+
{
|
| 2471 |
+
"epoch": 2.1,
|
| 2472 |
+
"grad_norm": 1.9200857877731323,
|
| 2473 |
+
"learning_rate": 8.964763665413894e-06,
|
| 2474 |
+
"loss": 0.6316,
|
| 2475 |
+
"step": 3340
|
| 2476 |
+
},
|
| 2477 |
+
{
|
| 2478 |
+
"epoch": 2.1,
|
| 2479 |
+
"grad_norm": 1.8148436546325684,
|
| 2480 |
+
"learning_rate": 8.95874426781257e-06,
|
| 2481 |
+
"loss": 0.6318,
|
| 2482 |
+
"step": 3350
|
| 2483 |
+
},
|
| 2484 |
+
{
|
| 2485 |
+
"epoch": 2.11,
|
| 2486 |
+
"grad_norm": 1.3358807563781738,
|
| 2487 |
+
"learning_rate": 8.952709454238809e-06,
|
| 2488 |
+
"loss": 0.6067,
|
| 2489 |
+
"step": 3360
|
| 2490 |
+
},
|
| 2491 |
+
{
|
| 2492 |
+
"epoch": 2.11,
|
| 2493 |
+
"grad_norm": 1.8055490255355835,
|
| 2494 |
+
"learning_rate": 8.946659248193122e-06,
|
| 2495 |
+
"loss": 0.6289,
|
| 2496 |
+
"step": 3370
|
| 2497 |
+
},
|
| 2498 |
+
{
|
| 2499 |
+
"epoch": 2.12,
|
| 2500 |
+
"grad_norm": 1.4589310884475708,
|
| 2501 |
+
"learning_rate": 8.940593673235962e-06,
|
| 2502 |
+
"loss": 0.6537,
|
| 2503 |
+
"step": 3380
|
| 2504 |
+
},
|
| 2505 |
+
{
|
| 2506 |
+
"epoch": 2.13,
|
| 2507 |
+
"grad_norm": 1.903086543083191,
|
| 2508 |
+
"learning_rate": 8.934512752987635e-06,
|
| 2509 |
+
"loss": 0.6986,
|
| 2510 |
+
"step": 3390
|
| 2511 |
+
},
|
| 2512 |
+
{
|
| 2513 |
+
"epoch": 2.13,
|
| 2514 |
+
"grad_norm": 1.722476840019226,
|
| 2515 |
+
"learning_rate": 8.928416511128194e-06,
|
| 2516 |
+
"loss": 0.6383,
|
| 2517 |
+
"step": 3400
|
| 2518 |
+
},
|
| 2519 |
+
{
|
| 2520 |
+
"epoch": 2.13,
|
| 2521 |
+
"eval_loss": 0.6974382996559143,
|
| 2522 |
+
"eval_runtime": 65.0527,
|
| 2523 |
+
"eval_samples_per_second": 69.175,
|
| 2524 |
+
"eval_steps_per_second": 4.335,
|
| 2525 |
+
"step": 3400
|
| 2526 |
+
},
|
| 2527 |
+
{
|
| 2528 |
+
"epoch": 2.14,
|
| 2529 |
+
"grad_norm": 1.8206970691680908,
|
| 2530 |
+
"learning_rate": 8.922304971397369e-06,
|
| 2531 |
+
"loss": 0.6447,
|
| 2532 |
+
"step": 3410
|
| 2533 |
+
},
|
| 2534 |
+
{
|
| 2535 |
+
"epoch": 2.15,
|
| 2536 |
+
"grad_norm": 1.690631628036499,
|
| 2537 |
+
"learning_rate": 8.916178157594453e-06,
|
| 2538 |
+
"loss": 0.6441,
|
| 2539 |
+
"step": 3420
|
| 2540 |
+
},
|
| 2541 |
+
{
|
| 2542 |
+
"epoch": 2.15,
|
| 2543 |
+
"grad_norm": 2.108876943588257,
|
| 2544 |
+
"learning_rate": 8.910036093578223e-06,
|
| 2545 |
+
"loss": 0.6453,
|
| 2546 |
+
"step": 3430
|
| 2547 |
+
},
|
| 2548 |
+
{
|
| 2549 |
+
"epoch": 2.16,
|
| 2550 |
+
"grad_norm": 1.6356040239334106,
|
| 2551 |
+
"learning_rate": 8.90387880326684e-06,
|
| 2552 |
+
"loss": 0.6256,
|
| 2553 |
+
"step": 3440
|
| 2554 |
+
},
|
| 2555 |
+
{
|
| 2556 |
+
"epoch": 2.16,
|
| 2557 |
+
"grad_norm": 1.3783752918243408,
|
| 2558 |
+
"learning_rate": 8.897706310637766e-06,
|
| 2559 |
+
"loss": 0.6244,
|
| 2560 |
+
"step": 3450
|
| 2561 |
+
},
|
| 2562 |
+
{
|
| 2563 |
+
"epoch": 2.17,
|
| 2564 |
+
"grad_norm": 1.901208758354187,
|
| 2565 |
+
"learning_rate": 8.89151863972765e-06,
|
| 2566 |
+
"loss": 0.5975,
|
| 2567 |
+
"step": 3460
|
| 2568 |
+
},
|
| 2569 |
+
{
|
| 2570 |
+
"epoch": 2.18,
|
| 2571 |
+
"grad_norm": 1.595009446144104,
|
| 2572 |
+
"learning_rate": 8.88531581463226e-06,
|
| 2573 |
+
"loss": 0.6777,
|
| 2574 |
+
"step": 3470
|
| 2575 |
+
},
|
| 2576 |
+
{
|
| 2577 |
+
"epoch": 2.18,
|
| 2578 |
+
"grad_norm": 1.6791367530822754,
|
| 2579 |
+
"learning_rate": 8.879097859506371e-06,
|
| 2580 |
+
"loss": 0.6139,
|
| 2581 |
+
"step": 3480
|
| 2582 |
+
},
|
| 2583 |
+
{
|
| 2584 |
+
"epoch": 2.19,
|
| 2585 |
+
"grad_norm": 1.5569841861724854,
|
| 2586 |
+
"learning_rate": 8.872864798563676e-06,
|
| 2587 |
+
"loss": 0.5775,
|
| 2588 |
+
"step": 3490
|
| 2589 |
+
},
|
| 2590 |
+
{
|
| 2591 |
+
"epoch": 2.2,
|
| 2592 |
+
"grad_norm": 1.3804527521133423,
|
| 2593 |
+
"learning_rate": 8.866616656076696e-06,
|
| 2594 |
+
"loss": 0.624,
|
| 2595 |
+
"step": 3500
|
| 2596 |
+
},
|
| 2597 |
+
{
|
| 2598 |
+
"epoch": 2.2,
|
| 2599 |
+
"grad_norm": 1.7905445098876953,
|
| 2600 |
+
"learning_rate": 8.860353456376679e-06,
|
| 2601 |
+
"loss": 0.6647,
|
| 2602 |
+
"step": 3510
|
| 2603 |
+
},
|
| 2604 |
+
{
|
| 2605 |
+
"epoch": 2.21,
|
| 2606 |
+
"grad_norm": 1.802614688873291,
|
| 2607 |
+
"learning_rate": 8.854075223853509e-06,
|
| 2608 |
+
"loss": 0.7081,
|
| 2609 |
+
"step": 3520
|
| 2610 |
+
},
|
| 2611 |
+
{
|
| 2612 |
+
"epoch": 2.21,
|
| 2613 |
+
"grad_norm": 1.7135951519012451,
|
| 2614 |
+
"learning_rate": 8.847781982955613e-06,
|
| 2615 |
+
"loss": 0.6974,
|
| 2616 |
+
"step": 3530
|
| 2617 |
+
},
|
| 2618 |
+
{
|
| 2619 |
+
"epoch": 2.22,
|
| 2620 |
+
"grad_norm": 1.8468406200408936,
|
| 2621 |
+
"learning_rate": 8.841473758189853e-06,
|
| 2622 |
+
"loss": 0.6585,
|
| 2623 |
+
"step": 3540
|
| 2624 |
+
},
|
| 2625 |
+
{
|
| 2626 |
+
"epoch": 2.23,
|
| 2627 |
+
"grad_norm": 1.7520302534103394,
|
| 2628 |
+
"learning_rate": 8.835150574121455e-06,
|
| 2629 |
+
"loss": 0.6116,
|
| 2630 |
+
"step": 3550
|
| 2631 |
+
},
|
| 2632 |
+
{
|
| 2633 |
+
"epoch": 2.23,
|
| 2634 |
+
"grad_norm": 1.862479329109192,
|
| 2635 |
+
"learning_rate": 8.828812455373891e-06,
|
| 2636 |
+
"loss": 0.6333,
|
| 2637 |
+
"step": 3560
|
| 2638 |
+
},
|
| 2639 |
+
{
|
| 2640 |
+
"epoch": 2.24,
|
| 2641 |
+
"grad_norm": 1.7767084836959839,
|
| 2642 |
+
"learning_rate": 8.82245942662879e-06,
|
| 2643 |
+
"loss": 0.6015,
|
| 2644 |
+
"step": 3570
|
| 2645 |
+
},
|
| 2646 |
+
{
|
| 2647 |
+
"epoch": 2.25,
|
| 2648 |
+
"grad_norm": 1.6162598133087158,
|
| 2649 |
+
"learning_rate": 8.816091512625845e-06,
|
| 2650 |
+
"loss": 0.6719,
|
| 2651 |
+
"step": 3580
|
| 2652 |
+
},
|
| 2653 |
+
{
|
| 2654 |
+
"epoch": 2.25,
|
| 2655 |
+
"grad_norm": 1.8923571109771729,
|
| 2656 |
+
"learning_rate": 8.80970873816271e-06,
|
| 2657 |
+
"loss": 0.6562,
|
| 2658 |
+
"step": 3590
|
| 2659 |
+
},
|
| 2660 |
+
{
|
| 2661 |
+
"epoch": 2.26,
|
| 2662 |
+
"grad_norm": 1.7792338132858276,
|
| 2663 |
+
"learning_rate": 8.803311128094918e-06,
|
| 2664 |
+
"loss": 0.6436,
|
| 2665 |
+
"step": 3600
|
| 2666 |
+
},
|
| 2667 |
+
{
|
| 2668 |
+
"epoch": 2.26,
|
| 2669 |
+
"eval_loss": 0.6980520486831665,
|
| 2670 |
+
"eval_runtime": 65.0239,
|
| 2671 |
+
"eval_samples_per_second": 69.205,
|
| 2672 |
+
"eval_steps_per_second": 4.337,
|
| 2673 |
+
"step": 3600
|
| 2674 |
+
},
|
| 2675 |
+
{
|
| 2676 |
+
"epoch": 2.26,
|
| 2677 |
+
"grad_norm": 1.819449543952942,
|
| 2678 |
+
"learning_rate": 8.796898707335766e-06,
|
| 2679 |
+
"loss": 0.6022,
|
| 2680 |
+
"step": 3610
|
| 2681 |
+
},
|
| 2682 |
+
{
|
| 2683 |
+
"epoch": 2.27,
|
| 2684 |
+
"grad_norm": 1.923462986946106,
|
| 2685 |
+
"learning_rate": 8.790471500856229e-06,
|
| 2686 |
+
"loss": 0.6124,
|
| 2687 |
+
"step": 3620
|
| 2688 |
+
},
|
| 2689 |
+
{
|
| 2690 |
+
"epoch": 2.28,
|
| 2691 |
+
"grad_norm": 2.0969247817993164,
|
| 2692 |
+
"learning_rate": 8.784029533684857e-06,
|
| 2693 |
+
"loss": 0.6209,
|
| 2694 |
+
"step": 3630
|
| 2695 |
+
},
|
| 2696 |
+
{
|
| 2697 |
+
"epoch": 2.28,
|
| 2698 |
+
"grad_norm": 2.014631509780884,
|
| 2699 |
+
"learning_rate": 8.777572830907685e-06,
|
| 2700 |
+
"loss": 0.6179,
|
| 2701 |
+
"step": 3640
|
| 2702 |
+
},
|
| 2703 |
+
{
|
| 2704 |
+
"epoch": 2.29,
|
| 2705 |
+
"grad_norm": 1.7940195798873901,
|
| 2706 |
+
"learning_rate": 8.771101417668127e-06,
|
| 2707 |
+
"loss": 0.6815,
|
| 2708 |
+
"step": 3650
|
| 2709 |
+
},
|
| 2710 |
+
{
|
| 2711 |
+
"epoch": 2.3,
|
| 2712 |
+
"grad_norm": 1.7244881391525269,
|
| 2713 |
+
"learning_rate": 8.764615319166885e-06,
|
| 2714 |
+
"loss": 0.5767,
|
| 2715 |
+
"step": 3660
|
| 2716 |
+
},
|
| 2717 |
+
{
|
| 2718 |
+
"epoch": 2.3,
|
| 2719 |
+
"grad_norm": 2.157749652862549,
|
| 2720 |
+
"learning_rate": 8.758114560661846e-06,
|
| 2721 |
+
"loss": 0.6281,
|
| 2722 |
+
"step": 3670
|
| 2723 |
+
},
|
| 2724 |
+
{
|
| 2725 |
+
"epoch": 2.31,
|
| 2726 |
+
"grad_norm": 1.818303108215332,
|
| 2727 |
+
"learning_rate": 8.751599167467985e-06,
|
| 2728 |
+
"loss": 0.6368,
|
| 2729 |
+
"step": 3680
|
| 2730 |
+
},
|
| 2731 |
+
{
|
| 2732 |
+
"epoch": 2.31,
|
| 2733 |
+
"grad_norm": 1.8076434135437012,
|
| 2734 |
+
"learning_rate": 8.745069164957265e-06,
|
| 2735 |
+
"loss": 0.6503,
|
| 2736 |
+
"step": 3690
|
| 2737 |
+
},
|
| 2738 |
+
{
|
| 2739 |
+
"epoch": 2.32,
|
| 2740 |
+
"grad_norm": 1.7755082845687866,
|
| 2741 |
+
"learning_rate": 8.738524578558547e-06,
|
| 2742 |
+
"loss": 0.6503,
|
| 2743 |
+
"step": 3700
|
| 2744 |
+
},
|
| 2745 |
+
{
|
| 2746 |
+
"epoch": 2.33,
|
| 2747 |
+
"grad_norm": 2.0641837120056152,
|
| 2748 |
+
"learning_rate": 8.731965433757474e-06,
|
| 2749 |
+
"loss": 0.6412,
|
| 2750 |
+
"step": 3710
|
| 2751 |
+
},
|
| 2752 |
+
{
|
| 2753 |
+
"epoch": 2.33,
|
| 2754 |
+
"grad_norm": 2.174612045288086,
|
| 2755 |
+
"learning_rate": 8.72539175609639e-06,
|
| 2756 |
+
"loss": 0.6283,
|
| 2757 |
+
"step": 3720
|
| 2758 |
+
},
|
| 2759 |
+
{
|
| 2760 |
+
"epoch": 2.34,
|
| 2761 |
+
"grad_norm": 1.905965805053711,
|
| 2762 |
+
"learning_rate": 8.718803571174229e-06,
|
| 2763 |
+
"loss": 0.6316,
|
| 2764 |
+
"step": 3730
|
| 2765 |
+
},
|
| 2766 |
+
{
|
| 2767 |
+
"epoch": 2.35,
|
| 2768 |
+
"grad_norm": 2.290787935256958,
|
| 2769 |
+
"learning_rate": 8.712200904646417e-06,
|
| 2770 |
+
"loss": 0.6337,
|
| 2771 |
+
"step": 3740
|
| 2772 |
+
},
|
| 2773 |
+
{
|
| 2774 |
+
"epoch": 2.35,
|
| 2775 |
+
"grad_norm": 1.7773081064224243,
|
| 2776 |
+
"learning_rate": 8.705583782224776e-06,
|
| 2777 |
+
"loss": 0.6683,
|
| 2778 |
+
"step": 3750
|
| 2779 |
+
},
|
| 2780 |
+
{
|
| 2781 |
+
"epoch": 2.36,
|
| 2782 |
+
"grad_norm": 1.7513020038604736,
|
| 2783 |
+
"learning_rate": 8.698952229677422e-06,
|
| 2784 |
+
"loss": 0.6538,
|
| 2785 |
+
"step": 3760
|
| 2786 |
+
},
|
| 2787 |
+
{
|
| 2788 |
+
"epoch": 2.37,
|
| 2789 |
+
"grad_norm": 1.8641185760498047,
|
| 2790 |
+
"learning_rate": 8.692306272828661e-06,
|
| 2791 |
+
"loss": 0.6179,
|
| 2792 |
+
"step": 3770
|
| 2793 |
+
},
|
| 2794 |
+
{
|
| 2795 |
+
"epoch": 2.37,
|
| 2796 |
+
"grad_norm": 2.4094667434692383,
|
| 2797 |
+
"learning_rate": 8.685645937558896e-06,
|
| 2798 |
+
"loss": 0.6436,
|
| 2799 |
+
"step": 3780
|
| 2800 |
+
},
|
| 2801 |
+
{
|
| 2802 |
+
"epoch": 2.38,
|
| 2803 |
+
"grad_norm": 2.295719623565674,
|
| 2804 |
+
"learning_rate": 8.678971249804517e-06,
|
| 2805 |
+
"loss": 0.6242,
|
| 2806 |
+
"step": 3790
|
| 2807 |
+
},
|
| 2808 |
+
{
|
| 2809 |
+
"epoch": 2.38,
|
| 2810 |
+
"grad_norm": 2.3604509830474854,
|
| 2811 |
+
"learning_rate": 8.67228223555781e-06,
|
| 2812 |
+
"loss": 0.6444,
|
| 2813 |
+
"step": 3800
|
| 2814 |
+
},
|
| 2815 |
+
{
|
| 2816 |
+
"epoch": 2.38,
|
| 2817 |
+
"eval_loss": 0.6968220472335815,
|
| 2818 |
+
"eval_runtime": 65.0257,
|
| 2819 |
+
"eval_samples_per_second": 69.203,
|
| 2820 |
+
"eval_steps_per_second": 4.337,
|
| 2821 |
+
"step": 3800
|
| 2822 |
+
},
|
| 2823 |
+
{
|
| 2824 |
+
"epoch": 2.39,
|
| 2825 |
+
"grad_norm": 1.799545407295227,
|
| 2826 |
+
"learning_rate": 8.665578920866844e-06,
|
| 2827 |
+
"loss": 0.6562,
|
| 2828 |
+
"step": 3810
|
| 2829 |
+
},
|
| 2830 |
+
{
|
| 2831 |
+
"epoch": 2.4,
|
| 2832 |
+
"grad_norm": 1.87678062915802,
|
| 2833 |
+
"learning_rate": 8.658861331835384e-06,
|
| 2834 |
+
"loss": 0.6776,
|
| 2835 |
+
"step": 3820
|
| 2836 |
+
},
|
| 2837 |
+
{
|
| 2838 |
+
"epoch": 2.4,
|
| 2839 |
+
"grad_norm": 1.9466888904571533,
|
| 2840 |
+
"learning_rate": 8.652129494622776e-06,
|
| 2841 |
+
"loss": 0.6245,
|
| 2842 |
+
"step": 3830
|
| 2843 |
+
},
|
| 2844 |
+
{
|
| 2845 |
+
"epoch": 2.41,
|
| 2846 |
+
"grad_norm": 1.9451625347137451,
|
| 2847 |
+
"learning_rate": 8.645383435443853e-06,
|
| 2848 |
+
"loss": 0.6692,
|
| 2849 |
+
"step": 3840
|
| 2850 |
+
},
|
| 2851 |
+
{
|
| 2852 |
+
"epoch": 2.42,
|
| 2853 |
+
"grad_norm": 1.9275856018066406,
|
| 2854 |
+
"learning_rate": 8.638623180568829e-06,
|
| 2855 |
+
"loss": 0.6314,
|
| 2856 |
+
"step": 3850
|
| 2857 |
+
},
|
| 2858 |
+
{
|
| 2859 |
+
"epoch": 2.42,
|
| 2860 |
+
"grad_norm": 1.7316443920135498,
|
| 2861 |
+
"learning_rate": 8.631848756323198e-06,
|
| 2862 |
+
"loss": 0.6289,
|
| 2863 |
+
"step": 3860
|
| 2864 |
+
},
|
| 2865 |
+
{
|
| 2866 |
+
"epoch": 2.43,
|
| 2867 |
+
"grad_norm": 2.5170657634735107,
|
| 2868 |
+
"learning_rate": 8.625060189087636e-06,
|
| 2869 |
+
"loss": 0.6367,
|
| 2870 |
+
"step": 3870
|
| 2871 |
+
},
|
| 2872 |
+
{
|
| 2873 |
+
"epoch": 2.43,
|
| 2874 |
+
"grad_norm": 2.2198870182037354,
|
| 2875 |
+
"learning_rate": 8.618257505297887e-06,
|
| 2876 |
+
"loss": 0.6262,
|
| 2877 |
+
"step": 3880
|
| 2878 |
+
},
|
| 2879 |
+
{
|
| 2880 |
+
"epoch": 2.44,
|
| 2881 |
+
"grad_norm": 1.9207948446273804,
|
| 2882 |
+
"learning_rate": 8.611440731444673e-06,
|
| 2883 |
+
"loss": 0.6598,
|
| 2884 |
+
"step": 3890
|
| 2885 |
+
},
|
| 2886 |
+
{
|
| 2887 |
+
"epoch": 2.45,
|
| 2888 |
+
"grad_norm": 2.0086634159088135,
|
| 2889 |
+
"learning_rate": 8.604609894073583e-06,
|
| 2890 |
+
"loss": 0.6465,
|
| 2891 |
+
"step": 3900
|
| 2892 |
+
},
|
| 2893 |
+
{
|
| 2894 |
+
"epoch": 2.45,
|
| 2895 |
+
"grad_norm": 2.0597639083862305,
|
| 2896 |
+
"learning_rate": 8.597765019784972e-06,
|
| 2897 |
+
"loss": 0.665,
|
| 2898 |
+
"step": 3910
|
| 2899 |
+
},
|
| 2900 |
+
{
|
| 2901 |
+
"epoch": 2.46,
|
| 2902 |
+
"grad_norm": 1.8585723638534546,
|
| 2903 |
+
"learning_rate": 8.590906135233854e-06,
|
| 2904 |
+
"loss": 0.6207,
|
| 2905 |
+
"step": 3920
|
| 2906 |
+
},
|
| 2907 |
+
{
|
| 2908 |
+
"epoch": 2.47,
|
| 2909 |
+
"grad_norm": 2.1862194538116455,
|
| 2910 |
+
"learning_rate": 8.584033267129807e-06,
|
| 2911 |
+
"loss": 0.6626,
|
| 2912 |
+
"step": 3930
|
| 2913 |
+
},
|
| 2914 |
+
{
|
| 2915 |
+
"epoch": 2.47,
|
| 2916 |
+
"grad_norm": 2.048553228378296,
|
| 2917 |
+
"learning_rate": 8.577146442236856e-06,
|
| 2918 |
+
"loss": 0.6141,
|
| 2919 |
+
"step": 3940
|
| 2920 |
+
},
|
| 2921 |
+
{
|
| 2922 |
+
"epoch": 2.48,
|
| 2923 |
+
"grad_norm": 2.2547719478607178,
|
| 2924 |
+
"learning_rate": 8.570245687373384e-06,
|
| 2925 |
+
"loss": 0.6651,
|
| 2926 |
+
"step": 3950
|
| 2927 |
+
},
|
| 2928 |
+
{
|
| 2929 |
+
"epoch": 2.48,
|
| 2930 |
+
"grad_norm": 1.9522244930267334,
|
| 2931 |
+
"learning_rate": 8.563331029412013e-06,
|
| 2932 |
+
"loss": 0.6725,
|
| 2933 |
+
"step": 3960
|
| 2934 |
+
},
|
| 2935 |
+
{
|
| 2936 |
+
"epoch": 2.49,
|
| 2937 |
+
"grad_norm": 1.7376751899719238,
|
| 2938 |
+
"learning_rate": 8.556402495279506e-06,
|
| 2939 |
+
"loss": 0.6066,
|
| 2940 |
+
"step": 3970
|
| 2941 |
+
},
|
| 2942 |
+
{
|
| 2943 |
+
"epoch": 2.5,
|
| 2944 |
+
"grad_norm": 1.900639295578003,
|
| 2945 |
+
"learning_rate": 8.549460111956665e-06,
|
| 2946 |
+
"loss": 0.6752,
|
| 2947 |
+
"step": 3980
|
| 2948 |
+
},
|
| 2949 |
+
{
|
| 2950 |
+
"epoch": 2.5,
|
| 2951 |
+
"grad_norm": 2.1750218868255615,
|
| 2952 |
+
"learning_rate": 8.542503906478224e-06,
|
| 2953 |
+
"loss": 0.6554,
|
| 2954 |
+
"step": 3990
|
| 2955 |
+
},
|
| 2956 |
+
{
|
| 2957 |
+
"epoch": 2.51,
|
| 2958 |
+
"grad_norm": 1.6551765203475952,
|
| 2959 |
+
"learning_rate": 8.535533905932739e-06,
|
| 2960 |
+
"loss": 0.6368,
|
| 2961 |
+
"step": 4000
|
| 2962 |
+
},
|
| 2963 |
+
{
|
| 2964 |
+
"epoch": 2.51,
|
| 2965 |
+
"eval_loss": 0.6986888647079468,
|
| 2966 |
+
"eval_runtime": 65.007,
|
| 2967 |
+
"eval_samples_per_second": 69.223,
|
| 2968 |
+
"eval_steps_per_second": 4.338,
|
| 2969 |
+
"step": 4000
|
| 2970 |
+
},
|
| 2971 |
+
{
|
| 2972 |
+
"epoch": 2.51,
|
| 2973 |
+
"step": 4000,
|
| 2974 |
+
"total_flos": 9.03392259225944e+17,
|
| 2975 |
+
"train_loss": 0.7081527805328369,
|
| 2976 |
+
"train_runtime": 4312.5386,
|
| 2977 |
+
"train_samples_per_second": 59.13,
|
| 2978 |
+
"train_steps_per_second": 3.696
|
| 2979 |
+
}
|
| 2980 |
+
],
|
| 2981 |
+
"logging_steps": 10,
|
| 2982 |
+
"max_steps": 15940,
|
| 2983 |
+
"num_input_tokens_seen": 0,
|
| 2984 |
+
"num_train_epochs": 10,
|
| 2985 |
+
"save_steps": 1000,
|
| 2986 |
+
"total_flos": 9.03392259225944e+17,
|
| 2987 |
+
"train_batch_size": 8,
|
| 2988 |
+
"trial_name": null,
|
| 2989 |
+
"trial_params": null
|
| 2990 |
+
}
|
llama2_13b_peft/news_commentary_de/training_args.bin
ADDED
|
@@ -0,0 +1,3 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
version https://git-lfs.github.com/spec/v1
|
| 2 |
+
oid sha256:190c16756cb1d693a35940640a359e26ac9a5176a57e5fc3919e9b137d670ef2
|
| 3 |
+
size 5112
|
llama2_13b_peft/news_commentary_de/training_eval_loss.png
ADDED
|
llama2_13b_peft/news_commentary_de/training_loss.png
ADDED
|
llama2_13b_peft/news_commentary_it/README.md
ADDED
|
@@ -0,0 +1,67 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
---
|
| 2 |
+
license: other
|
| 3 |
+
library_name: peft
|
| 4 |
+
tags:
|
| 5 |
+
- llama-factory
|
| 6 |
+
- lora
|
| 7 |
+
- generated_from_trainer
|
| 8 |
+
base_model: /data1/model/llama2/meta-llama/Llama2-13b
|
| 9 |
+
model-index:
|
| 10 |
+
- name: news_commentary_it_no_sys
|
| 11 |
+
results: []
|
| 12 |
+
---
|
| 13 |
+
|
| 14 |
+
<!-- This model card has been generated automatically according to the information the Trainer had access to. You
|
| 15 |
+
should probably proofread and complete it, then remove this comment. -->
|
| 16 |
+
|
| 17 |
+
# news_commentary_it_no_sys
|
| 18 |
+
|
| 19 |
+
This model is a fine-tuned version of [/data1/model/llama2/meta-llama/Llama2-13b](https://huggingface.co//data1/model/llama2/meta-llama/Llama2-13b) on the news_commentary_it_no_sys dataset.
|
| 20 |
+
It achieves the following results on the evaluation set:
|
| 21 |
+
- Loss: 0.6415
|
| 22 |
+
|
| 23 |
+
## Model description
|
| 24 |
+
|
| 25 |
+
More information needed
|
| 26 |
+
|
| 27 |
+
## Intended uses & limitations
|
| 28 |
+
|
| 29 |
+
More information needed
|
| 30 |
+
|
| 31 |
+
## Training and evaluation data
|
| 32 |
+
|
| 33 |
+
More information needed
|
| 34 |
+
|
| 35 |
+
## Training procedure
|
| 36 |
+
|
| 37 |
+
### Training hyperparameters
|
| 38 |
+
|
| 39 |
+
The following hyperparameters were used during training:
|
| 40 |
+
- learning_rate: 5e-05
|
| 41 |
+
- train_batch_size: 4
|
| 42 |
+
- eval_batch_size: 4
|
| 43 |
+
- seed: 42
|
| 44 |
+
- optimizer: Adam with betas=(0.9,0.999) and epsilon=1e-08
|
| 45 |
+
- lr_scheduler_type: cosine
|
| 46 |
+
- lr_scheduler_warmup_steps: 20
|
| 47 |
+
- num_epochs: 5.0
|
| 48 |
+
|
| 49 |
+
### Training results
|
| 50 |
+
|
| 51 |
+
| Training Loss | Epoch | Step | Validation Loss |
|
| 52 |
+
|:-------------:|:------:|:-----:|:---------------:|
|
| 53 |
+
| 0.7236 | 0.2807 | 2000 | 0.6710 |
|
| 54 |
+
| 0.724 | 0.5614 | 4000 | 0.6521 |
|
| 55 |
+
| 0.6455 | 0.8421 | 6000 | 0.6415 |
|
| 56 |
+
| 0.5533 | 1.1228 | 8000 | 0.6548 |
|
| 57 |
+
| 0.5192 | 1.4035 | 10000 | 0.6501 |
|
| 58 |
+
| 0.4796 | 1.6842 | 12000 | 0.6500 |
|
| 59 |
+
|
| 60 |
+
|
| 61 |
+
### Framework versions
|
| 62 |
+
|
| 63 |
+
- PEFT 0.10.0
|
| 64 |
+
- Transformers 4.40.0
|
| 65 |
+
- Pytorch 2.2.1
|
| 66 |
+
- Datasets 2.18.0
|
| 67 |
+
- Tokenizers 0.19.1
|
llama2_13b_peft/news_commentary_it/adapter_config.json
ADDED
|
@@ -0,0 +1,34 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
{
|
| 2 |
+
"alpha_pattern": {},
|
| 3 |
+
"auto_mapping": null,
|
| 4 |
+
"base_model_name_or_path": "/data1/model/llama2/meta-llama/Llama2-13b",
|
| 5 |
+
"bias": "none",
|
| 6 |
+
"fan_in_fan_out": false,
|
| 7 |
+
"inference_mode": true,
|
| 8 |
+
"init_lora_weights": true,
|
| 9 |
+
"layer_replication": null,
|
| 10 |
+
"layers_pattern": null,
|
| 11 |
+
"layers_to_transform": null,
|
| 12 |
+
"loftq_config": {},
|
| 13 |
+
"lora_alpha": 16,
|
| 14 |
+
"lora_dropout": 0.0,
|
| 15 |
+
"megatron_config": null,
|
| 16 |
+
"megatron_core": "megatron.core",
|
| 17 |
+
"modules_to_save": null,
|
| 18 |
+
"peft_type": "LORA",
|
| 19 |
+
"r": 8,
|
| 20 |
+
"rank_pattern": {},
|
| 21 |
+
"revision": null,
|
| 22 |
+
"target_modules": [
|
| 23 |
+
"down_proj",
|
| 24 |
+
"up_proj",
|
| 25 |
+
"k_proj",
|
| 26 |
+
"gate_proj",
|
| 27 |
+
"q_proj",
|
| 28 |
+
"o_proj",
|
| 29 |
+
"v_proj"
|
| 30 |
+
],
|
| 31 |
+
"task_type": "CAUSAL_LM",
|
| 32 |
+
"use_dora": false,
|
| 33 |
+
"use_rslora": false
|
| 34 |
+
}
|
llama2_13b_peft/news_commentary_it/adapter_model.safetensors
ADDED
|
@@ -0,0 +1,3 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
version https://git-lfs.github.com/spec/v1
|
| 2 |
+
oid sha256:edef1dfcb7b9574ec3ab4820df092cc5ff382071e97048d16d7f69d7094885d3
|
| 3 |
+
size 125248064
|
llama2_13b_peft/news_commentary_it/all_results.json
ADDED
|
@@ -0,0 +1,12 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
{
|
| 2 |
+
"epoch": 1.6842105263157894,
|
| 3 |
+
"eval_loss": 0.6415141820907593,
|
| 4 |
+
"eval_runtime": 119.5773,
|
| 5 |
+
"eval_samples_per_second": 12.544,
|
| 6 |
+
"eval_steps_per_second": 3.136,
|
| 7 |
+
"total_flos": 6.933368738955264e+17,
|
| 8 |
+
"train_loss": 0.6038338423768679,
|
| 9 |
+
"train_runtime": 5861.7175,
|
| 10 |
+
"train_samples_per_second": 24.31,
|
| 11 |
+
"train_steps_per_second": 6.078
|
| 12 |
+
}
|
llama2_13b_peft/news_commentary_it/eval_results.json
ADDED
|
@@ -0,0 +1,7 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
{
|
| 2 |
+
"epoch": 1.6842105263157894,
|
| 3 |
+
"eval_loss": 0.6415141820907593,
|
| 4 |
+
"eval_runtime": 119.5773,
|
| 5 |
+
"eval_samples_per_second": 12.544,
|
| 6 |
+
"eval_steps_per_second": 3.136
|
| 7 |
+
}
|
llama2_13b_peft/news_commentary_it/special_tokens_map.json
ADDED
|
@@ -0,0 +1,24 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
{
|
| 2 |
+
"bos_token": {
|
| 3 |
+
"content": "<s>",
|
| 4 |
+
"lstrip": false,
|
| 5 |
+
"normalized": false,
|
| 6 |
+
"rstrip": false,
|
| 7 |
+
"single_word": false
|
| 8 |
+
},
|
| 9 |
+
"eos_token": {
|
| 10 |
+
"content": "</s>",
|
| 11 |
+
"lstrip": false,
|
| 12 |
+
"normalized": false,
|
| 13 |
+
"rstrip": false,
|
| 14 |
+
"single_word": false
|
| 15 |
+
},
|
| 16 |
+
"pad_token": "</s>",
|
| 17 |
+
"unk_token": {
|
| 18 |
+
"content": "<unk>",
|
| 19 |
+
"lstrip": false,
|
| 20 |
+
"normalized": false,
|
| 21 |
+
"rstrip": false,
|
| 22 |
+
"single_word": false
|
| 23 |
+
}
|
| 24 |
+
}
|
llama2_13b_peft/news_commentary_it/tokenizer.model
ADDED
|
@@ -0,0 +1,3 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
version https://git-lfs.github.com/spec/v1
|
| 2 |
+
oid sha256:9e556afd44213b6bd1be2b850ebbbd98f5481437a8021afaf58ee7fb1818d347
|
| 3 |
+
size 499723
|
llama2_13b_peft/news_commentary_it/tokenizer_config.json
ADDED
|
@@ -0,0 +1,45 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
{
|
| 2 |
+
"add_bos_token": true,
|
| 3 |
+
"add_eos_token": false,
|
| 4 |
+
"add_prefix_space": true,
|
| 5 |
+
"added_tokens_decoder": {
|
| 6 |
+
"0": {
|
| 7 |
+
"content": "<unk>",
|
| 8 |
+
"lstrip": false,
|
| 9 |
+
"normalized": false,
|
| 10 |
+
"rstrip": false,
|
| 11 |
+
"single_word": false,
|
| 12 |
+
"special": true
|
| 13 |
+
},
|
| 14 |
+
"1": {
|
| 15 |
+
"content": "<s>",
|
| 16 |
+
"lstrip": false,
|
| 17 |
+
"normalized": false,
|
| 18 |
+
"rstrip": false,
|
| 19 |
+
"single_word": false,
|
| 20 |
+
"special": true
|
| 21 |
+
},
|
| 22 |
+
"2": {
|
| 23 |
+
"content": "</s>",
|
| 24 |
+
"lstrip": false,
|
| 25 |
+
"normalized": false,
|
| 26 |
+
"rstrip": false,
|
| 27 |
+
"single_word": false,
|
| 28 |
+
"special": true
|
| 29 |
+
}
|
| 30 |
+
},
|
| 31 |
+
"bos_token": "<s>",
|
| 32 |
+
"chat_template": "{% if messages[0]['role'] == 'system' %}{% set system_message = messages[0]['content'] %}{% endif %}{% if system_message is defined %}{{ system_message }}{% endif %}{% for message in messages %}{% set content = message['content'] %}{% if message['role'] == 'user' %}{{ content }}{% elif message['role'] == 'assistant' %}{{ content + '\\n' }}{% endif %}{% endfor %}",
|
| 33 |
+
"clean_up_tokenization_spaces": false,
|
| 34 |
+
"eos_token": "</s>",
|
| 35 |
+
"legacy": true,
|
| 36 |
+
"model_max_length": 1000000000000000019884624838656,
|
| 37 |
+
"pad_token": "</s>",
|
| 38 |
+
"padding_side": "right",
|
| 39 |
+
"sp_model_kwargs": {},
|
| 40 |
+
"spaces_between_special_tokens": false,
|
| 41 |
+
"split_special_tokens": false,
|
| 42 |
+
"tokenizer_class": "LlamaTokenizer",
|
| 43 |
+
"unk_token": "<unk>",
|
| 44 |
+
"use_default_system_prompt": false
|
| 45 |
+
}
|
llama2_13b_peft/news_commentary_it/train_results.json
ADDED
|
@@ -0,0 +1,8 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
{
|
| 2 |
+
"epoch": 1.6842105263157894,
|
| 3 |
+
"total_flos": 6.933368738955264e+17,
|
| 4 |
+
"train_loss": 0.6038338423768679,
|
| 5 |
+
"train_runtime": 5861.7175,
|
| 6 |
+
"train_samples_per_second": 24.31,
|
| 7 |
+
"train_steps_per_second": 6.078
|
| 8 |
+
}
|
llama2_13b_peft/news_commentary_it/trainer_log.jsonl
ADDED
|
The diff for this file is too large to render.
See raw diff
|
|
|
llama2_13b_peft/news_commentary_it/trainer_state.json
ADDED
|
The diff for this file is too large to render.
See raw diff
|
|
|
llama2_13b_peft/news_commentary_it/training_args.bin
ADDED
|
@@ -0,0 +1,3 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
version https://git-lfs.github.com/spec/v1
|
| 2 |
+
oid sha256:c760325c4e915ed6add2ca1fa0f2456628f1a65e9e53ca6ae7e92088e8ec81d2
|
| 3 |
+
size 5176
|
llama2_13b_peft/news_commentary_it/training_eval_loss.png
ADDED
|
llama2_13b_peft/news_commentary_it/training_loss.png
ADDED
|
llama2_13b_peft/topical_chat/README.md
ADDED
|
@@ -0,0 +1,89 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
---
|
| 2 |
+
license: other
|
| 3 |
+
library_name: peft
|
| 4 |
+
tags:
|
| 5 |
+
- llama-factory
|
| 6 |
+
- lora
|
| 7 |
+
- generated_from_trainer
|
| 8 |
+
base_model: /data1/model/llama2/meta-llama/Llama2-13b
|
| 9 |
+
model-index:
|
| 10 |
+
- name: topical_chat_no_sys
|
| 11 |
+
results: []
|
| 12 |
+
---
|
| 13 |
+
|
| 14 |
+
<!-- This model card has been generated automatically according to the information the Trainer had access to. You
|
| 15 |
+
should probably proofread and complete it, then remove this comment. -->
|
| 16 |
+
|
| 17 |
+
# topical_chat_no_sys
|
| 18 |
+
|
| 19 |
+
This model is a fine-tuned version of [/data1/model/llama2/meta-llama/Llama2-13b](https://huggingface.co//data1/model/llama2/meta-llama/Llama2-13b) on the topical_chat_no_sys dataset.
|
| 20 |
+
It achieves the following results on the evaluation set:
|
| 21 |
+
- Loss: 1.8941
|
| 22 |
+
|
| 23 |
+
## Model description
|
| 24 |
+
|
| 25 |
+
More information needed
|
| 26 |
+
|
| 27 |
+
## Intended uses & limitations
|
| 28 |
+
|
| 29 |
+
More information needed
|
| 30 |
+
|
| 31 |
+
## Training and evaluation data
|
| 32 |
+
|
| 33 |
+
More information needed
|
| 34 |
+
|
| 35 |
+
## Training procedure
|
| 36 |
+
|
| 37 |
+
### Training hyperparameters
|
| 38 |
+
|
| 39 |
+
The following hyperparameters were used during training:
|
| 40 |
+
- learning_rate: 5e-05
|
| 41 |
+
- train_batch_size: 4
|
| 42 |
+
- eval_batch_size: 4
|
| 43 |
+
- seed: 42
|
| 44 |
+
- distributed_type: multi-GPU
|
| 45 |
+
- num_devices: 2
|
| 46 |
+
- total_train_batch_size: 8
|
| 47 |
+
- total_eval_batch_size: 8
|
| 48 |
+
- optimizer: Adam with betas=(0.9,0.999) and epsilon=1e-08
|
| 49 |
+
- lr_scheduler_type: cosine
|
| 50 |
+
- lr_scheduler_warmup_steps: 20
|
| 51 |
+
- num_epochs: 5.0
|
| 52 |
+
|
| 53 |
+
### Training results
|
| 54 |
+
|
| 55 |
+
| Training Loss | Epoch | Step | Validation Loss |
|
| 56 |
+
|:-------------:|:------:|:----:|:---------------:|
|
| 57 |
+
| 2.1904 | 0.0472 | 100 | 2.1137 |
|
| 58 |
+
| 1.9627 | 0.0944 | 200 | 2.0589 |
|
| 59 |
+
| 2.0172 | 0.1416 | 300 | 2.0221 |
|
| 60 |
+
| 1.8965 | 0.1889 | 400 | 1.9968 |
|
| 61 |
+
| 1.9534 | 0.2361 | 500 | 1.9823 |
|
| 62 |
+
| 1.8621 | 0.2833 | 600 | 1.9679 |
|
| 63 |
+
| 1.9777 | 0.3305 | 700 | 1.9611 |
|
| 64 |
+
| 2.0865 | 0.3777 | 800 | 1.9544 |
|
| 65 |
+
| 1.9662 | 0.4249 | 900 | 1.9461 |
|
| 66 |
+
| 1.8352 | 0.4721 | 1000 | 1.9376 |
|
| 67 |
+
| 1.8973 | 0.5194 | 1100 | 1.9329 |
|
| 68 |
+
| 1.9688 | 0.5666 | 1200 | 1.9264 |
|
| 69 |
+
| 1.8383 | 0.6138 | 1300 | 1.9192 |
|
| 70 |
+
| 1.9032 | 0.6610 | 1400 | 1.9146 |
|
| 71 |
+
| 1.9295 | 0.7082 | 1500 | 1.9109 |
|
| 72 |
+
| 1.8207 | 0.7554 | 1600 | 1.9061 |
|
| 73 |
+
| 1.9119 | 0.8026 | 1700 | 1.9032 |
|
| 74 |
+
| 1.8392 | 0.8499 | 1800 | 1.9019 |
|
| 75 |
+
| 1.961 | 0.8971 | 1900 | 1.8994 |
|
| 76 |
+
| 1.8913 | 0.9443 | 2000 | 1.8945 |
|
| 77 |
+
| 1.8187 | 0.9915 | 2100 | 1.8941 |
|
| 78 |
+
| 1.7296 | 1.0387 | 2200 | 1.9006 |
|
| 79 |
+
| 1.6184 | 1.0859 | 2300 | 1.9040 |
|
| 80 |
+
| 1.6973 | 1.1331 | 2400 | 1.9056 |
|
| 81 |
+
|
| 82 |
+
|
| 83 |
+
### Framework versions
|
| 84 |
+
|
| 85 |
+
- PEFT 0.10.0
|
| 86 |
+
- Transformers 4.40.0
|
| 87 |
+
- Pytorch 2.2.1
|
| 88 |
+
- Datasets 2.18.0
|
| 89 |
+
- Tokenizers 0.19.1
|
llama2_13b_peft/topical_chat/adapter_config.json
ADDED
|
@@ -0,0 +1,34 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
{
|
| 2 |
+
"alpha_pattern": {},
|
| 3 |
+
"auto_mapping": null,
|
| 4 |
+
"base_model_name_or_path": "/data1/model/llama2/meta-llama/Llama2-13b",
|
| 5 |
+
"bias": "none",
|
| 6 |
+
"fan_in_fan_out": false,
|
| 7 |
+
"inference_mode": true,
|
| 8 |
+
"init_lora_weights": true,
|
| 9 |
+
"layer_replication": null,
|
| 10 |
+
"layers_pattern": null,
|
| 11 |
+
"layers_to_transform": null,
|
| 12 |
+
"loftq_config": {},
|
| 13 |
+
"lora_alpha": 16,
|
| 14 |
+
"lora_dropout": 0.0,
|
| 15 |
+
"megatron_config": null,
|
| 16 |
+
"megatron_core": "megatron.core",
|
| 17 |
+
"modules_to_save": null,
|
| 18 |
+
"peft_type": "LORA",
|
| 19 |
+
"r": 8,
|
| 20 |
+
"rank_pattern": {},
|
| 21 |
+
"revision": null,
|
| 22 |
+
"target_modules": [
|
| 23 |
+
"v_proj",
|
| 24 |
+
"down_proj",
|
| 25 |
+
"q_proj",
|
| 26 |
+
"gate_proj",
|
| 27 |
+
"up_proj",
|
| 28 |
+
"k_proj",
|
| 29 |
+
"o_proj"
|
| 30 |
+
],
|
| 31 |
+
"task_type": "CAUSAL_LM",
|
| 32 |
+
"use_dora": false,
|
| 33 |
+
"use_rslora": false
|
| 34 |
+
}
|
llama2_13b_peft/topical_chat/adapter_model.safetensors
ADDED
|
@@ -0,0 +1,3 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
version https://git-lfs.github.com/spec/v1
|
| 2 |
+
oid sha256:3e6e2ffd0565e16b426d000f5f8d1e22505be002104c4426830c2fe30127625f
|
| 3 |
+
size 125248064
|
llama2_13b_peft/topical_chat/all_results.json
ADDED
|
@@ -0,0 +1,12 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
{
|
| 2 |
+
"epoch": 1.13314447592068,
|
| 3 |
+
"eval_loss": 1.8941270112991333,
|
| 4 |
+
"eval_runtime": 40.8751,
|
| 5 |
+
"eval_samples_per_second": 21.823,
|
| 6 |
+
"eval_steps_per_second": 2.74,
|
| 7 |
+
"total_flos": 9.512959383227597e+17,
|
| 8 |
+
"train_loss": 1.9100826263427735,
|
| 9 |
+
"train_runtime": 3885.2685,
|
| 10 |
+
"train_samples_per_second": 21.805,
|
| 11 |
+
"train_steps_per_second": 2.726
|
| 12 |
+
}
|
llama2_13b_peft/topical_chat/eval_results.json
ADDED
|
@@ -0,0 +1,7 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
{
|
| 2 |
+
"epoch": 1.13314447592068,
|
| 3 |
+
"eval_loss": 1.8941270112991333,
|
| 4 |
+
"eval_runtime": 40.8751,
|
| 5 |
+
"eval_samples_per_second": 21.823,
|
| 6 |
+
"eval_steps_per_second": 2.74
|
| 7 |
+
}
|
llama2_13b_peft/topical_chat/special_tokens_map.json
ADDED
|
@@ -0,0 +1,24 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
{
|
| 2 |
+
"bos_token": {
|
| 3 |
+
"content": "<s>",
|
| 4 |
+
"lstrip": false,
|
| 5 |
+
"normalized": false,
|
| 6 |
+
"rstrip": false,
|
| 7 |
+
"single_word": false
|
| 8 |
+
},
|
| 9 |
+
"eos_token": {
|
| 10 |
+
"content": "</s>",
|
| 11 |
+
"lstrip": false,
|
| 12 |
+
"normalized": false,
|
| 13 |
+
"rstrip": false,
|
| 14 |
+
"single_word": false
|
| 15 |
+
},
|
| 16 |
+
"pad_token": "</s>",
|
| 17 |
+
"unk_token": {
|
| 18 |
+
"content": "<unk>",
|
| 19 |
+
"lstrip": false,
|
| 20 |
+
"normalized": false,
|
| 21 |
+
"rstrip": false,
|
| 22 |
+
"single_word": false
|
| 23 |
+
}
|
| 24 |
+
}
|
llama2_13b_peft/topical_chat/tokenizer.model
ADDED
|
@@ -0,0 +1,3 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
version https://git-lfs.github.com/spec/v1
|
| 2 |
+
oid sha256:9e556afd44213b6bd1be2b850ebbbd98f5481437a8021afaf58ee7fb1818d347
|
| 3 |
+
size 499723
|
llama2_13b_peft/topical_chat/tokenizer_config.json
ADDED
|
@@ -0,0 +1,45 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
{
|
| 2 |
+
"add_bos_token": true,
|
| 3 |
+
"add_eos_token": false,
|
| 4 |
+
"add_prefix_space": true,
|
| 5 |
+
"added_tokens_decoder": {
|
| 6 |
+
"0": {
|
| 7 |
+
"content": "<unk>",
|
| 8 |
+
"lstrip": false,
|
| 9 |
+
"normalized": false,
|
| 10 |
+
"rstrip": false,
|
| 11 |
+
"single_word": false,
|
| 12 |
+
"special": true
|
| 13 |
+
},
|
| 14 |
+
"1": {
|
| 15 |
+
"content": "<s>",
|
| 16 |
+
"lstrip": false,
|
| 17 |
+
"normalized": false,
|
| 18 |
+
"rstrip": false,
|
| 19 |
+
"single_word": false,
|
| 20 |
+
"special": true
|
| 21 |
+
},
|
| 22 |
+
"2": {
|
| 23 |
+
"content": "</s>",
|
| 24 |
+
"lstrip": false,
|
| 25 |
+
"normalized": false,
|
| 26 |
+
"rstrip": false,
|
| 27 |
+
"single_word": false,
|
| 28 |
+
"special": true
|
| 29 |
+
}
|
| 30 |
+
},
|
| 31 |
+
"bos_token": "<s>",
|
| 32 |
+
"chat_template": "{% if messages[0]['role'] == 'system' %}{% set system_message = messages[0]['content'] %}{% endif %}{% if system_message is defined %}{{ system_message }}{% endif %}{% for message in messages %}{% set content = message['content'] %}{% if message['role'] == 'user' %}{{ content }}{% elif message['role'] == 'assistant' %}{{ content + '\\n' }}{% endif %}{% endfor %}",
|
| 33 |
+
"clean_up_tokenization_spaces": false,
|
| 34 |
+
"eos_token": "</s>",
|
| 35 |
+
"legacy": true,
|
| 36 |
+
"model_max_length": 1000000000000000019884624838656,
|
| 37 |
+
"pad_token": "</s>",
|
| 38 |
+
"padding_side": "right",
|
| 39 |
+
"sp_model_kwargs": {},
|
| 40 |
+
"spaces_between_special_tokens": false,
|
| 41 |
+
"split_special_tokens": false,
|
| 42 |
+
"tokenizer_class": "LlamaTokenizer",
|
| 43 |
+
"unk_token": "<unk>",
|
| 44 |
+
"use_default_system_prompt": false
|
| 45 |
+
}
|