Model save
Browse files- README.md +73 -0
- all_results.json +9 -0
- generation_config.json +14 -0
- train_results.json +9 -0
- trainer_state.json +1209 -0
README.md
ADDED
|
@@ -0,0 +1,73 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
---
|
| 2 |
+
library_name: transformers
|
| 3 |
+
license: apache-2.0
|
| 4 |
+
base_model: Qwen/Qwen2.5-7B-Instruct-1M
|
| 5 |
+
tags:
|
| 6 |
+
- trl
|
| 7 |
+
- ndcg
|
| 8 |
+
- generated_from_trainer
|
| 9 |
+
model-index:
|
| 10 |
+
- name: qwen2.5-7b-lipo
|
| 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 |
+
# qwen2.5-7b-lipo
|
| 18 |
+
|
| 19 |
+
This model is a fine-tuned version of [Qwen/Qwen2.5-7B-Instruct-1M](https://huggingface.co/Qwen/Qwen2.5-7B-Instruct-1M) on an unknown dataset.
|
| 20 |
+
It achieves the following results on the evaluation set:
|
| 21 |
+
- Loss: 0.0017
|
| 22 |
+
- Logps: -662.0203
|
| 23 |
+
- Logits: -0.6116
|
| 24 |
+
- Rank Correct Batch: 16.5244
|
| 25 |
+
- Rank Pair Batch: 28.0
|
| 26 |
+
- Rank Accuracy Batch: 0.5902
|
| 27 |
+
|
| 28 |
+
## Model description
|
| 29 |
+
|
| 30 |
+
More information needed
|
| 31 |
+
|
| 32 |
+
## Intended uses & limitations
|
| 33 |
+
|
| 34 |
+
More information needed
|
| 35 |
+
|
| 36 |
+
## Training and evaluation data
|
| 37 |
+
|
| 38 |
+
More information needed
|
| 39 |
+
|
| 40 |
+
## Training procedure
|
| 41 |
+
|
| 42 |
+
### Training hyperparameters
|
| 43 |
+
|
| 44 |
+
The following hyperparameters were used during training:
|
| 45 |
+
- learning_rate: 5e-07
|
| 46 |
+
- train_batch_size: 1
|
| 47 |
+
- eval_batch_size: 1
|
| 48 |
+
- seed: 42
|
| 49 |
+
- distributed_type: multi-GPU
|
| 50 |
+
- num_devices: 8
|
| 51 |
+
- gradient_accumulation_steps: 16
|
| 52 |
+
- total_train_batch_size: 128
|
| 53 |
+
- total_eval_batch_size: 8
|
| 54 |
+
- optimizer: Adam with betas=(0.9,0.999) and epsilon=1e-08
|
| 55 |
+
- lr_scheduler_type: cosine
|
| 56 |
+
- lr_scheduler_warmup_ratio: 0.1
|
| 57 |
+
- num_epochs: 1
|
| 58 |
+
|
| 59 |
+
### Training results
|
| 60 |
+
|
| 61 |
+
| Training Loss | Epoch | Step | Validation Loss | Logps | Logits | Rank Correct Batch | Rank Pair Batch | Rank Accuracy Batch |
|
| 62 |
+
|:-------------:|:------:|:----:|:---------------:|:---------:|:-------:|:------------------:|:---------------:|:-------------------:|
|
| 63 |
+
| 0.0021 | 0.2672 | 125 | 0.0021 | -582.2852 | -0.6831 | 15.8902 | 28.0 | 0.5675 |
|
| 64 |
+
| 0.0019 | 0.5344 | 250 | 0.0018 | -642.7833 | -0.6304 | 16.4431 | 28.0 | 0.5873 |
|
| 65 |
+
| 0.0017 | 0.8016 | 375 | 0.0017 | -662.0203 | -0.6116 | 16.5244 | 28.0 | 0.5902 |
|
| 66 |
+
|
| 67 |
+
|
| 68 |
+
### Framework versions
|
| 69 |
+
|
| 70 |
+
- Transformers 4.45.1
|
| 71 |
+
- Pytorch 2.6.0+cu126
|
| 72 |
+
- Datasets 2.19.1
|
| 73 |
+
- Tokenizers 0.20.3
|
all_results.json
ADDED
|
@@ -0,0 +1,9 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
{
|
| 2 |
+
"epoch": 0.9982631930527722,
|
| 3 |
+
"total_flos": 0.0,
|
| 4 |
+
"train_loss": 0.0024812357072425285,
|
| 5 |
+
"train_runtime": 43392.086,
|
| 6 |
+
"train_samples": 59880,
|
| 7 |
+
"train_samples_per_second": 1.38,
|
| 8 |
+
"train_steps_per_second": 0.011
|
| 9 |
+
}
|
generation_config.json
ADDED
|
@@ -0,0 +1,14 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
{
|
| 2 |
+
"bos_token_id": 151643,
|
| 3 |
+
"do_sample": true,
|
| 4 |
+
"eos_token_id": [
|
| 5 |
+
151645,
|
| 6 |
+
151643
|
| 7 |
+
],
|
| 8 |
+
"pad_token_id": 151643,
|
| 9 |
+
"repetition_penalty": 1.05,
|
| 10 |
+
"temperature": 0.7,
|
| 11 |
+
"top_k": 20,
|
| 12 |
+
"top_p": 0.8,
|
| 13 |
+
"transformers_version": "4.45.1"
|
| 14 |
+
}
|
train_results.json
ADDED
|
@@ -0,0 +1,9 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
{
|
| 2 |
+
"epoch": 0.9982631930527722,
|
| 3 |
+
"total_flos": 0.0,
|
| 4 |
+
"train_loss": 0.0024812357072425285,
|
| 5 |
+
"train_runtime": 43392.086,
|
| 6 |
+
"train_samples": 59880,
|
| 7 |
+
"train_samples_per_second": 1.38,
|
| 8 |
+
"train_steps_per_second": 0.011
|
| 9 |
+
}
|
trainer_state.json
ADDED
|
@@ -0,0 +1,1209 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
{
|
| 2 |
+
"best_metric": null,
|
| 3 |
+
"best_model_checkpoint": null,
|
| 4 |
+
"epoch": 0.9982631930527722,
|
| 5 |
+
"eval_steps": 125,
|
| 6 |
+
"global_step": 467,
|
| 7 |
+
"is_hyper_param_search": false,
|
| 8 |
+
"is_local_process_zero": true,
|
| 9 |
+
"is_world_process_zero": true,
|
| 10 |
+
"log_history": [
|
| 11 |
+
{
|
| 12 |
+
"epoch": 0.0021376085504342017,
|
| 13 |
+
"grad_norm": 0.946260824126923,
|
| 14 |
+
"learning_rate": 1.0638297872340425e-08,
|
| 15 |
+
"logits": -0.8763603568077087,
|
| 16 |
+
"logps": -333.0801086425781,
|
| 17 |
+
"loss": 0.0135,
|
| 18 |
+
"rank_accuracy_batch": 0.5357142686843872,
|
| 19 |
+
"rank_correct_batch": 15.0,
|
| 20 |
+
"rank_pair_batch": 28.0,
|
| 21 |
+
"step": 1
|
| 22 |
+
},
|
| 23 |
+
{
|
| 24 |
+
"epoch": 0.01068804275217101,
|
| 25 |
+
"grad_norm": 0.5832860071737439,
|
| 26 |
+
"learning_rate": 5.3191489361702123e-08,
|
| 27 |
+
"logits": -0.6306054592132568,
|
| 28 |
+
"logps": -445.8094177246094,
|
| 29 |
+
"loss": 0.0102,
|
| 30 |
+
"rank_accuracy_batch": 0.5094866156578064,
|
| 31 |
+
"rank_correct_batch": 14.265625,
|
| 32 |
+
"rank_pair_batch": 28.0,
|
| 33 |
+
"step": 5
|
| 34 |
+
},
|
| 35 |
+
{
|
| 36 |
+
"epoch": 0.02137608550434202,
|
| 37 |
+
"grad_norm": 0.5256642937190692,
|
| 38 |
+
"learning_rate": 1.0638297872340425e-07,
|
| 39 |
+
"logits": -0.6946691274642944,
|
| 40 |
+
"logps": -410.3299865722656,
|
| 41 |
+
"loss": 0.009,
|
| 42 |
+
"rank_accuracy_batch": 0.5151785612106323,
|
| 43 |
+
"rank_correct_batch": 14.425000190734863,
|
| 44 |
+
"rank_pair_batch": 28.0,
|
| 45 |
+
"step": 10
|
| 46 |
+
},
|
| 47 |
+
{
|
| 48 |
+
"epoch": 0.03206412825651302,
|
| 49 |
+
"grad_norm": 0.5688879024259528,
|
| 50 |
+
"learning_rate": 1.5957446808510638e-07,
|
| 51 |
+
"logits": -0.7815518379211426,
|
| 52 |
+
"logps": -440.68719482421875,
|
| 53 |
+
"loss": 0.0088,
|
| 54 |
+
"rank_accuracy_batch": 0.4973214566707611,
|
| 55 |
+
"rank_correct_batch": 13.925000190734863,
|
| 56 |
+
"rank_pair_batch": 28.0,
|
| 57 |
+
"step": 15
|
| 58 |
+
},
|
| 59 |
+
{
|
| 60 |
+
"epoch": 0.04275217100868404,
|
| 61 |
+
"grad_norm": 0.5292008384625928,
|
| 62 |
+
"learning_rate": 2.127659574468085e-07,
|
| 63 |
+
"logits": -0.7508541941642761,
|
| 64 |
+
"logps": -403.905029296875,
|
| 65 |
+
"loss": 0.0082,
|
| 66 |
+
"rank_accuracy_batch": 0.5066964030265808,
|
| 67 |
+
"rank_correct_batch": 14.1875,
|
| 68 |
+
"rank_pair_batch": 28.0,
|
| 69 |
+
"step": 20
|
| 70 |
+
},
|
| 71 |
+
{
|
| 72 |
+
"epoch": 0.053440213760855046,
|
| 73 |
+
"grad_norm": 0.44055358919363613,
|
| 74 |
+
"learning_rate": 2.659574468085106e-07,
|
| 75 |
+
"logits": -0.5607888102531433,
|
| 76 |
+
"logps": -490.1800842285156,
|
| 77 |
+
"loss": 0.0072,
|
| 78 |
+
"rank_accuracy_batch": 0.5267857313156128,
|
| 79 |
+
"rank_correct_batch": 14.75,
|
| 80 |
+
"rank_pair_batch": 28.0,
|
| 81 |
+
"step": 25
|
| 82 |
+
},
|
| 83 |
+
{
|
| 84 |
+
"epoch": 0.06412825651302605,
|
| 85 |
+
"grad_norm": 0.3173930509724987,
|
| 86 |
+
"learning_rate": 3.1914893617021275e-07,
|
| 87 |
+
"logits": -0.6815467476844788,
|
| 88 |
+
"logps": -458.843505859375,
|
| 89 |
+
"loss": 0.0068,
|
| 90 |
+
"rank_accuracy_batch": 0.46741074323654175,
|
| 91 |
+
"rank_correct_batch": 13.087499618530273,
|
| 92 |
+
"rank_pair_batch": 28.0,
|
| 93 |
+
"step": 30
|
| 94 |
+
},
|
| 95 |
+
{
|
| 96 |
+
"epoch": 0.07481629926519706,
|
| 97 |
+
"grad_norm": 0.19937438735024765,
|
| 98 |
+
"learning_rate": 3.7234042553191484e-07,
|
| 99 |
+
"logits": -0.6512653827667236,
|
| 100 |
+
"logps": -475.3102111816406,
|
| 101 |
+
"loss": 0.0053,
|
| 102 |
+
"rank_accuracy_batch": 0.5147321224212646,
|
| 103 |
+
"rank_correct_batch": 14.412500381469727,
|
| 104 |
+
"rank_pair_batch": 28.0,
|
| 105 |
+
"step": 35
|
| 106 |
+
},
|
| 107 |
+
{
|
| 108 |
+
"epoch": 0.08550434201736808,
|
| 109 |
+
"grad_norm": 0.22400025328295686,
|
| 110 |
+
"learning_rate": 4.25531914893617e-07,
|
| 111 |
+
"logits": -0.6020452976226807,
|
| 112 |
+
"logps": -492.62725830078125,
|
| 113 |
+
"loss": 0.005,
|
| 114 |
+
"rank_accuracy_batch": 0.5165178179740906,
|
| 115 |
+
"rank_correct_batch": 14.462499618530273,
|
| 116 |
+
"rank_pair_batch": 28.0,
|
| 117 |
+
"step": 40
|
| 118 |
+
},
|
| 119 |
+
{
|
| 120 |
+
"epoch": 0.09619238476953908,
|
| 121 |
+
"grad_norm": 0.3466345827980356,
|
| 122 |
+
"learning_rate": 4.787234042553192e-07,
|
| 123 |
+
"logits": -0.6586597561836243,
|
| 124 |
+
"logps": -489.76739501953125,
|
| 125 |
+
"loss": 0.0042,
|
| 126 |
+
"rank_accuracy_batch": 0.5674107074737549,
|
| 127 |
+
"rank_correct_batch": 15.887499809265137,
|
| 128 |
+
"rank_pair_batch": 28.0,
|
| 129 |
+
"step": 45
|
| 130 |
+
},
|
| 131 |
+
{
|
| 132 |
+
"epoch": 0.10688042752171009,
|
| 133 |
+
"grad_norm": 0.19018104091113128,
|
| 134 |
+
"learning_rate": 4.999370587356267e-07,
|
| 135 |
+
"logits": -0.7162013053894043,
|
| 136 |
+
"logps": -518.6836547851562,
|
| 137 |
+
"loss": 0.0036,
|
| 138 |
+
"rank_accuracy_batch": 0.5191963911056519,
|
| 139 |
+
"rank_correct_batch": 14.537500381469727,
|
| 140 |
+
"rank_pair_batch": 28.0,
|
| 141 |
+
"step": 50
|
| 142 |
+
},
|
| 143 |
+
{
|
| 144 |
+
"epoch": 0.11756847027388109,
|
| 145 |
+
"grad_norm": 0.0891015053073235,
|
| 146 |
+
"learning_rate": 4.995525324419337e-07,
|
| 147 |
+
"logits": -0.7136077880859375,
|
| 148 |
+
"logps": -543.9591064453125,
|
| 149 |
+
"loss": 0.0029,
|
| 150 |
+
"rank_accuracy_batch": 0.5294643640518188,
|
| 151 |
+
"rank_correct_batch": 14.824999809265137,
|
| 152 |
+
"rank_pair_batch": 28.0,
|
| 153 |
+
"step": 55
|
| 154 |
+
},
|
| 155 |
+
{
|
| 156 |
+
"epoch": 0.1282565130260521,
|
| 157 |
+
"grad_norm": 0.10337269806484616,
|
| 158 |
+
"learning_rate": 4.988189843662815e-07,
|
| 159 |
+
"logits": -0.7633501291275024,
|
| 160 |
+
"logps": -507.84332275390625,
|
| 161 |
+
"loss": 0.0028,
|
| 162 |
+
"rank_accuracy_batch": 0.5477678179740906,
|
| 163 |
+
"rank_correct_batch": 15.337499618530273,
|
| 164 |
+
"rank_pair_batch": 28.0,
|
| 165 |
+
"step": 60
|
| 166 |
+
},
|
| 167 |
+
{
|
| 168 |
+
"epoch": 0.13894455577822312,
|
| 169 |
+
"grad_norm": 0.08934007009145527,
|
| 170 |
+
"learning_rate": 4.977374404419837e-07,
|
| 171 |
+
"logits": -0.7554048299789429,
|
| 172 |
+
"logps": -541.6370849609375,
|
| 173 |
+
"loss": 0.0026,
|
| 174 |
+
"rank_accuracy_batch": 0.5352678894996643,
|
| 175 |
+
"rank_correct_batch": 14.987500190734863,
|
| 176 |
+
"rank_pair_batch": 28.0,
|
| 177 |
+
"step": 65
|
| 178 |
+
},
|
| 179 |
+
{
|
| 180 |
+
"epoch": 0.14963259853039412,
|
| 181 |
+
"grad_norm": 0.07381765178572891,
|
| 182 |
+
"learning_rate": 4.963094133060148e-07,
|
| 183 |
+
"logits": -0.7326418161392212,
|
| 184 |
+
"logps": -548.10888671875,
|
| 185 |
+
"loss": 0.0027,
|
| 186 |
+
"rank_accuracy_batch": 0.5191963911056519,
|
| 187 |
+
"rank_correct_batch": 14.537500381469727,
|
| 188 |
+
"rank_pair_batch": 28.0,
|
| 189 |
+
"step": 70
|
| 190 |
+
},
|
| 191 |
+
{
|
| 192 |
+
"epoch": 0.16032064128256512,
|
| 193 |
+
"grad_norm": 0.07900638364182098,
|
| 194 |
+
"learning_rate": 4.945369001834514e-07,
|
| 195 |
+
"logits": -0.5735518932342529,
|
| 196 |
+
"logps": -545.7005004882812,
|
| 197 |
+
"loss": 0.0026,
|
| 198 |
+
"rank_accuracy_batch": 0.5834820866584778,
|
| 199 |
+
"rank_correct_batch": 16.337499618530273,
|
| 200 |
+
"rank_pair_batch": 28.0,
|
| 201 |
+
"step": 75
|
| 202 |
+
},
|
| 203 |
+
{
|
| 204 |
+
"epoch": 0.17100868403473615,
|
| 205 |
+
"grad_norm": 0.0754388382054278,
|
| 206 |
+
"learning_rate": 4.924223800941717e-07,
|
| 207 |
+
"logits": -0.6778115034103394,
|
| 208 |
+
"logps": -587.3670654296875,
|
| 209 |
+
"loss": 0.0028,
|
| 210 |
+
"rank_accuracy_batch": 0.5406249761581421,
|
| 211 |
+
"rank_correct_batch": 15.137499809265137,
|
| 212 |
+
"rank_pair_batch": 28.0,
|
| 213 |
+
"step": 80
|
| 214 |
+
},
|
| 215 |
+
{
|
| 216 |
+
"epoch": 0.18169672678690715,
|
| 217 |
+
"grad_norm": 0.09079705752428875,
|
| 218 |
+
"learning_rate": 4.899688103857222e-07,
|
| 219 |
+
"logits": -0.6451147794723511,
|
| 220 |
+
"logps": -577.9484252929688,
|
| 221 |
+
"loss": 0.0025,
|
| 222 |
+
"rank_accuracy_batch": 0.543749988079071,
|
| 223 |
+
"rank_correct_batch": 15.225000381469727,
|
| 224 |
+
"rank_pair_batch": 28.0,
|
| 225 |
+
"step": 85
|
| 226 |
+
},
|
| 227 |
+
{
|
| 228 |
+
"epoch": 0.19238476953907815,
|
| 229 |
+
"grad_norm": 0.08714142887549842,
|
| 230 |
+
"learning_rate": 4.871796225971999e-07,
|
| 231 |
+
"logits": -0.6205800771713257,
|
| 232 |
+
"logps": -557.2830810546875,
|
| 233 |
+
"loss": 0.0022,
|
| 234 |
+
"rank_accuracy_batch": 0.5968749523162842,
|
| 235 |
+
"rank_correct_batch": 16.712499618530273,
|
| 236 |
+
"rank_pair_batch": 28.0,
|
| 237 |
+
"step": 90
|
| 238 |
+
},
|
| 239 |
+
{
|
| 240 |
+
"epoch": 0.20307281229124916,
|
| 241 |
+
"grad_norm": 0.07499434317114564,
|
| 242 |
+
"learning_rate": 4.840587176599343e-07,
|
| 243 |
+
"logits": -0.5821539759635925,
|
| 244 |
+
"logps": -567.4512939453125,
|
| 245 |
+
"loss": 0.0023,
|
| 246 |
+
"rank_accuracy_batch": 0.5758928060531616,
|
| 247 |
+
"rank_correct_batch": 16.125,
|
| 248 |
+
"rank_pair_batch": 28.0,
|
| 249 |
+
"step": 95
|
| 250 |
+
},
|
| 251 |
+
{
|
| 252 |
+
"epoch": 0.21376085504342018,
|
| 253 |
+
"grad_norm": 0.06200004666951232,
|
| 254 |
+
"learning_rate": 4.806104604416823e-07,
|
| 255 |
+
"logits": -0.6915315389633179,
|
| 256 |
+
"logps": -548.736328125,
|
| 257 |
+
"loss": 0.0023,
|
| 258 |
+
"rank_accuracy_batch": 0.6272321939468384,
|
| 259 |
+
"rank_correct_batch": 17.5625,
|
| 260 |
+
"rank_pair_batch": 28.0,
|
| 261 |
+
"step": 100
|
| 262 |
+
},
|
| 263 |
+
{
|
| 264 |
+
"epoch": 0.22444889779559118,
|
| 265 |
+
"grad_norm": 0.06860524649416584,
|
| 266 |
+
"learning_rate": 4.768396736419662e-07,
|
| 267 |
+
"logits": -0.5968545079231262,
|
| 268 |
+
"logps": -583.2061157226562,
|
| 269 |
+
"loss": 0.0023,
|
| 270 |
+
"rank_accuracy_batch": 0.5397320985794067,
|
| 271 |
+
"rank_correct_batch": 15.112500190734863,
|
| 272 |
+
"rank_pair_batch": 28.0,
|
| 273 |
+
"step": 105
|
| 274 |
+
},
|
| 275 |
+
{
|
| 276 |
+
"epoch": 0.23513694054776219,
|
| 277 |
+
"grad_norm": 0.06818885545810188,
|
| 278 |
+
"learning_rate": 4.7275163104709194e-07,
|
| 279 |
+
"logits": -0.7068358063697815,
|
| 280 |
+
"logps": -567.2909545898438,
|
| 281 |
+
"loss": 0.0022,
|
| 282 |
+
"rank_accuracy_batch": 0.5625,
|
| 283 |
+
"rank_correct_batch": 15.75,
|
| 284 |
+
"rank_pair_batch": 28.0,
|
| 285 |
+
"step": 110
|
| 286 |
+
},
|
| 287 |
+
{
|
| 288 |
+
"epoch": 0.2458249832999332,
|
| 289 |
+
"grad_norm": 0.11288632323265162,
|
| 290 |
+
"learning_rate": 4.683520501542824e-07,
|
| 291 |
+
"logits": -0.6993803977966309,
|
| 292 |
+
"logps": -578.380615234375,
|
| 293 |
+
"loss": 0.0029,
|
| 294 |
+
"rank_accuracy_batch": 0.5696429014205933,
|
| 295 |
+
"rank_correct_batch": 15.949999809265137,
|
| 296 |
+
"rank_pair_batch": 28.0,
|
| 297 |
+
"step": 115
|
| 298 |
+
},
|
| 299 |
+
{
|
| 300 |
+
"epoch": 0.2565130260521042,
|
| 301 |
+
"grad_norm": 0.10115494872852501,
|
| 302 |
+
"learning_rate": 4.636470841752404e-07,
|
| 303 |
+
"logits": -0.6544754505157471,
|
| 304 |
+
"logps": -579.39794921875,
|
| 305 |
+
"loss": 0.0023,
|
| 306 |
+
"rank_accuracy_batch": 0.5513392090797424,
|
| 307 |
+
"rank_correct_batch": 15.4375,
|
| 308 |
+
"rank_pair_batch": 28.0,
|
| 309 |
+
"step": 120
|
| 310 |
+
},
|
| 311 |
+
{
|
| 312 |
+
"epoch": 0.26720106880427524,
|
| 313 |
+
"grad_norm": 0.05669944797220972,
|
| 314 |
+
"learning_rate": 4.5864331343032565e-07,
|
| 315 |
+
"logits": -0.6703131794929504,
|
| 316 |
+
"logps": -584.9163818359375,
|
| 317 |
+
"loss": 0.0021,
|
| 318 |
+
"rank_accuracy_batch": 0.5651785731315613,
|
| 319 |
+
"rank_correct_batch": 15.824999809265137,
|
| 320 |
+
"rank_pair_batch": 28.0,
|
| 321 |
+
"step": 125
|
| 322 |
+
},
|
| 323 |
+
{
|
| 324 |
+
"epoch": 0.26720106880427524,
|
| 325 |
+
"eval_logits": -0.6831080317497253,
|
| 326 |
+
"eval_logps": -582.2852172851562,
|
| 327 |
+
"eval_loss": 0.002090274356305599,
|
| 328 |
+
"eval_rank_accuracy_batch": 0.5675087571144104,
|
| 329 |
+
"eval_rank_correct_batch": 15.890243530273438,
|
| 330 |
+
"eval_rank_pair_batch": 28.0,
|
| 331 |
+
"eval_runtime": 733.3311,
|
| 332 |
+
"eval_samples_per_second": 2.684,
|
| 333 |
+
"eval_steps_per_second": 0.335,
|
| 334 |
+
"step": 125
|
| 335 |
+
},
|
| 336 |
+
{
|
| 337 |
+
"epoch": 0.27788911155644624,
|
| 338 |
+
"grad_norm": 0.05664990070569458,
|
| 339 |
+
"learning_rate": 4.533477361453819e-07,
|
| 340 |
+
"logits": -0.6876681447029114,
|
| 341 |
+
"logps": -563.9698486328125,
|
| 342 |
+
"loss": 0.0023,
|
| 343 |
+
"rank_accuracy_batch": 0.5874999761581421,
|
| 344 |
+
"rank_correct_batch": 16.450000762939453,
|
| 345 |
+
"rank_pair_batch": 28.0,
|
| 346 |
+
"step": 130
|
| 347 |
+
},
|
| 348 |
+
{
|
| 349 |
+
"epoch": 0.28857715430861725,
|
| 350 |
+
"grad_norm": 0.06709747272300577,
|
| 351 |
+
"learning_rate": 4.4776775866408533e-07,
|
| 352 |
+
"logits": -0.6728977560997009,
|
| 353 |
+
"logps": -567.8697509765625,
|
| 354 |
+
"loss": 0.0023,
|
| 355 |
+
"rank_accuracy_batch": 0.6303571462631226,
|
| 356 |
+
"rank_correct_batch": 17.649999618530273,
|
| 357 |
+
"rank_pair_batch": 28.0,
|
| 358 |
+
"step": 135
|
| 359 |
+
},
|
| 360 |
+
{
|
| 361 |
+
"epoch": 0.29926519706078825,
|
| 362 |
+
"grad_norm": 0.06283846802681914,
|
| 363 |
+
"learning_rate": 4.4191118508950277e-07,
|
| 364 |
+
"logits": -0.6071655750274658,
|
| 365 |
+
"logps": -605.4354858398438,
|
| 366 |
+
"loss": 0.0021,
|
| 367 |
+
"rank_accuracy_batch": 0.6120535731315613,
|
| 368 |
+
"rank_correct_batch": 17.137500762939453,
|
| 369 |
+
"rank_pair_batch": 28.0,
|
| 370 |
+
"step": 140
|
| 371 |
+
},
|
| 372 |
+
{
|
| 373 |
+
"epoch": 0.30995323981295925,
|
| 374 |
+
"grad_norm": 0.120942854394424,
|
| 375 |
+
"learning_rate": 4.357862063693485e-07,
|
| 376 |
+
"logits": -0.5925200581550598,
|
| 377 |
+
"logps": -602.9215087890625,
|
| 378 |
+
"loss": 0.0024,
|
| 379 |
+
"rank_accuracy_batch": 0.6044643521308899,
|
| 380 |
+
"rank_correct_batch": 16.924999237060547,
|
| 381 |
+
"rank_pair_batch": 28.0,
|
| 382 |
+
"step": 145
|
| 383 |
+
},
|
| 384 |
+
{
|
| 385 |
+
"epoch": 0.32064128256513025,
|
| 386 |
+
"grad_norm": 0.05765935553467444,
|
| 387 |
+
"learning_rate": 4.294013888402029e-07,
|
| 388 |
+
"logits": -0.582276463508606,
|
| 389 |
+
"logps": -627.7027587890625,
|
| 390 |
+
"loss": 0.0021,
|
| 391 |
+
"rank_accuracy_batch": 0.5727678537368774,
|
| 392 |
+
"rank_correct_batch": 16.037500381469727,
|
| 393 |
+
"rank_pair_batch": 28.0,
|
| 394 |
+
"step": 150
|
| 395 |
+
},
|
| 396 |
+
{
|
| 397 |
+
"epoch": 0.33132932531730125,
|
| 398 |
+
"grad_norm": 0.05747505683203417,
|
| 399 |
+
"learning_rate": 4.227656622467162e-07,
|
| 400 |
+
"logits": -0.6392695903778076,
|
| 401 |
+
"logps": -602.9111938476562,
|
| 402 |
+
"loss": 0.0021,
|
| 403 |
+
"rank_accuracy_batch": 0.6191964149475098,
|
| 404 |
+
"rank_correct_batch": 17.337499618530273,
|
| 405 |
+
"rank_pair_batch": 28.0,
|
| 406 |
+
"step": 155
|
| 407 |
+
},
|
| 408 |
+
{
|
| 409 |
+
"epoch": 0.3420173680694723,
|
| 410 |
+
"grad_norm": 0.053776462481106506,
|
| 411 |
+
"learning_rate": 4.158883072525528e-07,
|
| 412 |
+
"logits": -0.577269971370697,
|
| 413 |
+
"logps": -572.0589599609375,
|
| 414 |
+
"loss": 0.002,
|
| 415 |
+
"rank_accuracy_batch": 0.581250011920929,
|
| 416 |
+
"rank_correct_batch": 16.274999618530273,
|
| 417 |
+
"rank_pair_batch": 28.0,
|
| 418 |
+
"step": 160
|
| 419 |
+
},
|
| 420 |
+
{
|
| 421 |
+
"epoch": 0.3527054108216433,
|
| 422 |
+
"grad_norm": 0.068483282911774,
|
| 423 |
+
"learning_rate": 4.087789424605447e-07,
|
| 424 |
+
"logits": -0.628365159034729,
|
| 425 |
+
"logps": -625.1907958984375,
|
| 426 |
+
"loss": 0.002,
|
| 427 |
+
"rank_accuracy_batch": 0.6031249761581421,
|
| 428 |
+
"rank_correct_batch": 16.887500762939453,
|
| 429 |
+
"rank_pair_batch": 28.0,
|
| 430 |
+
"step": 165
|
| 431 |
+
},
|
| 432 |
+
{
|
| 433 |
+
"epoch": 0.3633934535738143,
|
| 434 |
+
"grad_norm": 0.05447449479619988,
|
| 435 |
+
"learning_rate": 4.0144751096020497e-07,
|
| 436 |
+
"logits": -0.6775780916213989,
|
| 437 |
+
"logps": -604.8009033203125,
|
| 438 |
+
"loss": 0.002,
|
| 439 |
+
"rank_accuracy_batch": 0.5937499403953552,
|
| 440 |
+
"rank_correct_batch": 16.625,
|
| 441 |
+
"rank_pair_batch": 28.0,
|
| 442 |
+
"step": 170
|
| 443 |
+
},
|
| 444 |
+
{
|
| 445 |
+
"epoch": 0.3740814963259853,
|
| 446 |
+
"grad_norm": 0.054725196545229296,
|
| 447 |
+
"learning_rate": 3.939042664214184e-07,
|
| 448 |
+
"logits": -0.6487151384353638,
|
| 449 |
+
"logps": -608.8302612304688,
|
| 450 |
+
"loss": 0.002,
|
| 451 |
+
"rank_accuracy_batch": 0.5821428894996643,
|
| 452 |
+
"rank_correct_batch": 16.299999237060547,
|
| 453 |
+
"rank_pair_batch": 28.0,
|
| 454 |
+
"step": 175
|
| 455 |
+
},
|
| 456 |
+
{
|
| 457 |
+
"epoch": 0.3847695390781563,
|
| 458 |
+
"grad_norm": 0.06150290350278254,
|
| 459 |
+
"learning_rate": 3.8615975875375676e-07,
|
| 460 |
+
"logits": -0.6963125467300415,
|
| 461 |
+
"logps": -614.14111328125,
|
| 462 |
+
"loss": 0.002,
|
| 463 |
+
"rank_accuracy_batch": 0.5928571820259094,
|
| 464 |
+
"rank_correct_batch": 16.600000381469727,
|
| 465 |
+
"rank_pair_batch": 28.0,
|
| 466 |
+
"step": 180
|
| 467 |
+
},
|
| 468 |
+
{
|
| 469 |
+
"epoch": 0.3954575818303273,
|
| 470 |
+
"grad_norm": 0.04659758683281854,
|
| 471 |
+
"learning_rate": 3.7822481935147655e-07,
|
| 472 |
+
"logits": -0.7263899445533752,
|
| 473 |
+
"logps": -577.3916015625,
|
| 474 |
+
"loss": 0.0022,
|
| 475 |
+
"rank_accuracy_batch": 0.5816964507102966,
|
| 476 |
+
"rank_correct_batch": 16.287500381469727,
|
| 477 |
+
"rank_pair_batch": 28.0,
|
| 478 |
+
"step": 185
|
| 479 |
+
},
|
| 480 |
+
{
|
| 481 |
+
"epoch": 0.4061456245824983,
|
| 482 |
+
"grad_norm": 0.050337368452175985,
|
| 483 |
+
"learning_rate": 3.7011054594483443e-07,
|
| 484 |
+
"logits": -0.624728262424469,
|
| 485 |
+
"logps": -642.1470947265625,
|
| 486 |
+
"loss": 0.002,
|
| 487 |
+
"rank_accuracy_batch": 0.5633928179740906,
|
| 488 |
+
"rank_correct_batch": 15.774999618530273,
|
| 489 |
+
"rank_pair_batch": 28.0,
|
| 490 |
+
"step": 190
|
| 491 |
+
},
|
| 492 |
+
{
|
| 493 |
+
"epoch": 0.4168336673346693,
|
| 494 |
+
"grad_norm": 0.04713598792833116,
|
| 495 |
+
"learning_rate": 3.618282870789081e-07,
|
| 496 |
+
"logits": -0.6701709032058716,
|
| 497 |
+
"logps": -639.9006958007812,
|
| 498 |
+
"loss": 0.0017,
|
| 499 |
+
"rank_accuracy_batch": 0.5825892686843872,
|
| 500 |
+
"rank_correct_batch": 16.3125,
|
| 501 |
+
"rank_pair_batch": 28.0,
|
| 502 |
+
"step": 195
|
| 503 |
+
},
|
| 504 |
+
{
|
| 505 |
+
"epoch": 0.42752171008684037,
|
| 506 |
+
"grad_norm": 0.07879592696619511,
|
| 507 |
+
"learning_rate": 3.5338962624163016e-07,
|
| 508 |
+
"logits": -0.6116264462471008,
|
| 509 |
+
"logps": -617.0225219726562,
|
| 510 |
+
"loss": 0.0018,
|
| 511 |
+
"rank_accuracy_batch": 0.6174107789993286,
|
| 512 |
+
"rank_correct_batch": 17.287500381469727,
|
| 513 |
+
"rank_pair_batch": 28.0,
|
| 514 |
+
"step": 200
|
| 515 |
+
},
|
| 516 |
+
{
|
| 517 |
+
"epoch": 0.43820975283901137,
|
| 518 |
+
"grad_norm": 0.041090647303877605,
|
| 519 |
+
"learning_rate": 3.448063656632321e-07,
|
| 520 |
+
"logits": -0.6368564367294312,
|
| 521 |
+
"logps": -597.2711181640625,
|
| 522 |
+
"loss": 0.0019,
|
| 523 |
+
"rank_accuracy_batch": 0.6026785969734192,
|
| 524 |
+
"rank_correct_batch": 16.875,
|
| 525 |
+
"rank_pair_batch": 28.0,
|
| 526 |
+
"step": 205
|
| 527 |
+
},
|
| 528 |
+
{
|
| 529 |
+
"epoch": 0.44889779559118237,
|
| 530 |
+
"grad_norm": 0.05082964812925144,
|
| 531 |
+
"learning_rate": 3.360905098097587e-07,
|
| 532 |
+
"logits": -0.6129381060600281,
|
| 533 |
+
"logps": -628.8250732421875,
|
| 534 |
+
"loss": 0.002,
|
| 535 |
+
"rank_accuracy_batch": 0.579464316368103,
|
| 536 |
+
"rank_correct_batch": 16.225000381469727,
|
| 537 |
+
"rank_pair_batch": 28.0,
|
| 538 |
+
"step": 210
|
| 539 |
+
},
|
| 540 |
+
{
|
| 541 |
+
"epoch": 0.45958583834335337,
|
| 542 |
+
"grad_norm": 0.07641485533445969,
|
| 543 |
+
"learning_rate": 3.272542485937368e-07,
|
| 544 |
+
"logits": -0.5754351019859314,
|
| 545 |
+
"logps": -617.5399169921875,
|
| 546 |
+
"loss": 0.0016,
|
| 547 |
+
"rank_accuracy_batch": 0.6040178537368774,
|
| 548 |
+
"rank_correct_batch": 16.912500381469727,
|
| 549 |
+
"rank_pair_batch": 28.0,
|
| 550 |
+
"step": 215
|
| 551 |
+
},
|
| 552 |
+
{
|
| 553 |
+
"epoch": 0.47027388109552437,
|
| 554 |
+
"grad_norm": 0.07330919105386577,
|
| 555 |
+
"learning_rate": 3.1830994032548e-07,
|
| 556 |
+
"logits": -0.6116534471511841,
|
| 557 |
+
"logps": -615.790771484375,
|
| 558 |
+
"loss": 0.0019,
|
| 559 |
+
"rank_accuracy_batch": 0.5776785612106323,
|
| 560 |
+
"rank_correct_batch": 16.174999237060547,
|
| 561 |
+
"rank_pair_batch": 28.0,
|
| 562 |
+
"step": 220
|
| 563 |
+
},
|
| 564 |
+
{
|
| 565 |
+
"epoch": 0.48096192384769537,
|
| 566 |
+
"grad_norm": 0.06469864723171964,
|
| 567 |
+
"learning_rate": 3.0927009442887437e-07,
|
| 568 |
+
"logits": -0.7047996520996094,
|
| 569 |
+
"logps": -664.7101440429688,
|
| 570 |
+
"loss": 0.0019,
|
| 571 |
+
"rank_accuracy_batch": 0.5843750238418579,
|
| 572 |
+
"rank_correct_batch": 16.362499237060547,
|
| 573 |
+
"rank_pair_batch": 28.0,
|
| 574 |
+
"step": 225
|
| 575 |
+
},
|
| 576 |
+
{
|
| 577 |
+
"epoch": 0.4916499665998664,
|
| 578 |
+
"grad_norm": 0.04289842312835369,
|
| 579 |
+
"learning_rate": 3.001473539458182e-07,
|
| 580 |
+
"logits": -0.6492472290992737,
|
| 581 |
+
"logps": -664.0413208007812,
|
| 582 |
+
"loss": 0.0019,
|
| 583 |
+
"rank_accuracy_batch": 0.5455357432365417,
|
| 584 |
+
"rank_correct_batch": 15.274999618530273,
|
| 585 |
+
"rank_pair_batch": 28.0,
|
| 586 |
+
"step": 230
|
| 587 |
+
},
|
| 588 |
+
{
|
| 589 |
+
"epoch": 0.5023380093520374,
|
| 590 |
+
"grad_norm": 0.04591086518117152,
|
| 591 |
+
"learning_rate": 2.909544778537844e-07,
|
| 592 |
+
"logits": -0.5703697800636292,
|
| 593 |
+
"logps": -643.8348388671875,
|
| 594 |
+
"loss": 0.0017,
|
| 595 |
+
"rank_accuracy_batch": 0.5745536088943481,
|
| 596 |
+
"rank_correct_batch": 16.087499618530273,
|
| 597 |
+
"rank_pair_batch": 28.0,
|
| 598 |
+
"step": 235
|
| 599 |
+
},
|
| 600 |
+
{
|
| 601 |
+
"epoch": 0.5130260521042084,
|
| 602 |
+
"grad_norm": 0.05802670095800759,
|
| 603 |
+
"learning_rate": 2.817043232212371e-07,
|
| 604 |
+
"logits": -0.6384927034378052,
|
| 605 |
+
"logps": -650.6683349609375,
|
| 606 |
+
"loss": 0.0018,
|
| 607 |
+
"rank_accuracy_batch": 0.5625,
|
| 608 |
+
"rank_correct_batch": 15.75,
|
| 609 |
+
"rank_pair_batch": 28.0,
|
| 610 |
+
"step": 240
|
| 611 |
+
},
|
| 612 |
+
{
|
| 613 |
+
"epoch": 0.5237140948563794,
|
| 614 |
+
"grad_norm": 0.05993926521438956,
|
| 615 |
+
"learning_rate": 2.7240982722585837e-07,
|
| 616 |
+
"logits": -0.6757279634475708,
|
| 617 |
+
"logps": -668.7239990234375,
|
| 618 |
+
"loss": 0.0017,
|
| 619 |
+
"rank_accuracy_batch": 0.6022320985794067,
|
| 620 |
+
"rank_correct_batch": 16.862499237060547,
|
| 621 |
+
"rank_pair_batch": 28.0,
|
| 622 |
+
"step": 245
|
| 623 |
+
},
|
| 624 |
+
{
|
| 625 |
+
"epoch": 0.5344021376085505,
|
| 626 |
+
"grad_norm": 0.06433940212096037,
|
| 627 |
+
"learning_rate": 2.63083989060736e-07,
|
| 628 |
+
"logits": -0.6638821959495544,
|
| 629 |
+
"logps": -635.4695434570312,
|
| 630 |
+
"loss": 0.0019,
|
| 631 |
+
"rank_accuracy_batch": 0.604910671710968,
|
| 632 |
+
"rank_correct_batch": 16.9375,
|
| 633 |
+
"rank_pair_batch": 28.0,
|
| 634 |
+
"step": 250
|
| 635 |
+
},
|
| 636 |
+
{
|
| 637 |
+
"epoch": 0.5344021376085505,
|
| 638 |
+
"eval_logits": -0.6304262280464172,
|
| 639 |
+
"eval_logps": -642.7833251953125,
|
| 640 |
+
"eval_loss": 0.0018058380810543895,
|
| 641 |
+
"eval_rank_accuracy_batch": 0.5872531533241272,
|
| 642 |
+
"eval_rank_correct_batch": 16.44308853149414,
|
| 643 |
+
"eval_rank_pair_batch": 28.0,
|
| 644 |
+
"eval_runtime": 631.7704,
|
| 645 |
+
"eval_samples_per_second": 3.115,
|
| 646 |
+
"eval_steps_per_second": 0.389,
|
| 647 |
+
"step": 250
|
| 648 |
+
},
|
| 649 |
+
{
|
| 650 |
+
"epoch": 0.5450901803607214,
|
| 651 |
+
"grad_norm": 0.07400871579554123,
|
| 652 |
+
"learning_rate": 2.537398517538159e-07,
|
| 653 |
+
"logits": -0.5997278690338135,
|
| 654 |
+
"logps": -636.6998901367188,
|
| 655 |
+
"loss": 0.0017,
|
| 656 |
+
"rank_accuracy_batch": 0.5732142925262451,
|
| 657 |
+
"rank_correct_batch": 16.049999237060547,
|
| 658 |
+
"rank_pair_batch": 28.0,
|
| 659 |
+
"step": 255
|
| 660 |
+
},
|
| 661 |
+
{
|
| 662 |
+
"epoch": 0.5557782231128925,
|
| 663 |
+
"grad_norm": 0.059296255308015175,
|
| 664 |
+
"learning_rate": 2.4439048392604877e-07,
|
| 665 |
+
"logits": -0.5596941709518433,
|
| 666 |
+
"logps": -648.1635131835938,
|
| 667 |
+
"loss": 0.002,
|
| 668 |
+
"rank_accuracy_batch": 0.5723214149475098,
|
| 669 |
+
"rank_correct_batch": 16.024999618530273,
|
| 670 |
+
"rank_pair_batch": 28.0,
|
| 671 |
+
"step": 260
|
| 672 |
+
},
|
| 673 |
+
{
|
| 674 |
+
"epoch": 0.5664662658650634,
|
| 675 |
+
"grad_norm": 0.056990364840059676,
|
| 676 |
+
"learning_rate": 2.3504896151374144e-07,
|
| 677 |
+
"logits": -0.5953022837638855,
|
| 678 |
+
"logps": -648.7028198242188,
|
| 679 |
+
"loss": 0.0019,
|
| 680 |
+
"rank_accuracy_batch": 0.6004465222358704,
|
| 681 |
+
"rank_correct_batch": 16.8125,
|
| 682 |
+
"rank_pair_batch": 28.0,
|
| 683 |
+
"step": 265
|
| 684 |
+
},
|
| 685 |
+
{
|
| 686 |
+
"epoch": 0.5771543086172345,
|
| 687 |
+
"grad_norm": 0.06983836548087012,
|
| 688 |
+
"learning_rate": 2.2572834948067795e-07,
|
| 689 |
+
"logits": -0.5165086984634399,
|
| 690 |
+
"logps": -629.17138671875,
|
| 691 |
+
"loss": 0.0017,
|
| 692 |
+
"rank_accuracy_batch": 0.6178570985794067,
|
| 693 |
+
"rank_correct_batch": 17.299999237060547,
|
| 694 |
+
"rank_pair_batch": 28.0,
|
| 695 |
+
"step": 270
|
| 696 |
+
},
|
| 697 |
+
{
|
| 698 |
+
"epoch": 0.5878423513694054,
|
| 699 |
+
"grad_norm": 0.0908282752152867,
|
| 700 |
+
"learning_rate": 2.164416835455862e-07,
|
| 701 |
+
"logits": -0.6052151918411255,
|
| 702 |
+
"logps": -636.0174560546875,
|
| 703 |
+
"loss": 0.0022,
|
| 704 |
+
"rank_accuracy_batch": 0.6089285612106323,
|
| 705 |
+
"rank_correct_batch": 17.049999237060547,
|
| 706 |
+
"rank_pair_batch": 28.0,
|
| 707 |
+
"step": 275
|
| 708 |
+
},
|
| 709 |
+
{
|
| 710 |
+
"epoch": 0.5985303941215765,
|
| 711 |
+
"grad_norm": 0.07810449359022072,
|
| 712 |
+
"learning_rate": 2.072019519505062e-07,
|
| 713 |
+
"logits": -0.6738878488540649,
|
| 714 |
+
"logps": -674.0641479492188,
|
| 715 |
+
"loss": 0.0018,
|
| 716 |
+
"rank_accuracy_batch": 0.6031249761581421,
|
| 717 |
+
"rank_correct_batch": 16.887500762939453,
|
| 718 |
+
"rank_pair_batch": 28.0,
|
| 719 |
+
"step": 280
|
| 720 |
+
},
|
| 721 |
+
{
|
| 722 |
+
"epoch": 0.6092184368737475,
|
| 723 |
+
"grad_norm": 0.04424869680826283,
|
| 724 |
+
"learning_rate": 1.980220772955602e-07,
|
| 725 |
+
"logits": -0.6369552612304688,
|
| 726 |
+
"logps": -636.3786010742188,
|
| 727 |
+
"loss": 0.0017,
|
| 728 |
+
"rank_accuracy_batch": 0.6178571581840515,
|
| 729 |
+
"rank_correct_batch": 17.299999237060547,
|
| 730 |
+
"rank_pair_batch": 28.0,
|
| 731 |
+
"step": 285
|
| 732 |
+
},
|
| 733 |
+
{
|
| 734 |
+
"epoch": 0.6199064796259185,
|
| 735 |
+
"grad_norm": 0.03976151746590099,
|
| 736 |
+
"learning_rate": 1.8891489846552644e-07,
|
| 737 |
+
"logits": -0.6333539485931396,
|
| 738 |
+
"logps": -659.5064086914062,
|
| 739 |
+
"loss": 0.0018,
|
| 740 |
+
"rank_accuracy_batch": 0.5874999761581421,
|
| 741 |
+
"rank_correct_batch": 16.450000762939453,
|
| 742 |
+
"rank_pair_batch": 28.0,
|
| 743 |
+
"step": 290
|
| 744 |
+
},
|
| 745 |
+
{
|
| 746 |
+
"epoch": 0.6305945223780896,
|
| 747 |
+
"grad_norm": 0.06288491191103909,
|
| 748 |
+
"learning_rate": 1.7989315267349933e-07,
|
| 749 |
+
"logits": -0.5686911344528198,
|
| 750 |
+
"logps": -644.9364013671875,
|
| 751 |
+
"loss": 0.0017,
|
| 752 |
+
"rank_accuracy_batch": 0.5888393521308899,
|
| 753 |
+
"rank_correct_batch": 16.487499237060547,
|
| 754 |
+
"rank_pair_batch": 28.0,
|
| 755 |
+
"step": 295
|
| 756 |
+
},
|
| 757 |
+
{
|
| 758 |
+
"epoch": 0.6412825651302605,
|
| 759 |
+
"grad_norm": 0.08488837449955576,
|
| 760 |
+
"learning_rate": 1.7096945764674398e-07,
|
| 761 |
+
"logits": -0.4138154089450836,
|
| 762 |
+
"logps": -657.07421875,
|
| 763 |
+
"loss": 0.0018,
|
| 764 |
+
"rank_accuracy_batch": 0.6241070628166199,
|
| 765 |
+
"rank_correct_batch": 17.475000381469727,
|
| 766 |
+
"rank_pair_batch": 28.0,
|
| 767 |
+
"step": 300
|
| 768 |
+
},
|
| 769 |
+
{
|
| 770 |
+
"epoch": 0.6519706078824316,
|
| 771 |
+
"grad_norm": 0.050504774406439176,
|
| 772 |
+
"learning_rate": 1.621562939796643e-07,
|
| 773 |
+
"logits": -0.613270103931427,
|
| 774 |
+
"logps": -667.7855834960938,
|
| 775 |
+
"loss": 0.0018,
|
| 776 |
+
"rank_accuracy_batch": 0.5459821224212646,
|
| 777 |
+
"rank_correct_batch": 15.287500381469727,
|
| 778 |
+
"rank_pair_batch": 28.0,
|
| 779 |
+
"step": 305
|
| 780 |
+
},
|
| 781 |
+
{
|
| 782 |
+
"epoch": 0.6626586506346025,
|
| 783 |
+
"grad_norm": 0.04829182476552713,
|
| 784 |
+
"learning_rate": 1.5346598767856345e-07,
|
| 785 |
+
"logits": -0.5592894554138184,
|
| 786 |
+
"logps": -657.7384033203125,
|
| 787 |
+
"loss": 0.0016,
|
| 788 |
+
"rank_accuracy_batch": 0.5892857313156128,
|
| 789 |
+
"rank_correct_batch": 16.5,
|
| 790 |
+
"rank_pair_batch": 28.0,
|
| 791 |
+
"step": 310
|
| 792 |
+
},
|
| 793 |
+
{
|
| 794 |
+
"epoch": 0.6733466933867736,
|
| 795 |
+
"grad_norm": 0.052101161713427706,
|
| 796 |
+
"learning_rate": 1.4491069292260866e-07,
|
| 797 |
+
"logits": -0.6785243153572083,
|
| 798 |
+
"logps": -684.28955078125,
|
| 799 |
+
"loss": 0.0017,
|
| 800 |
+
"rank_accuracy_batch": 0.563839316368103,
|
| 801 |
+
"rank_correct_batch": 15.787500381469727,
|
| 802 |
+
"rank_pair_batch": 28.0,
|
| 803 |
+
"step": 315
|
| 804 |
+
},
|
| 805 |
+
{
|
| 806 |
+
"epoch": 0.6840347361389446,
|
| 807 |
+
"grad_norm": 0.05260035842708517,
|
| 808 |
+
"learning_rate": 1.365023750651133e-07,
|
| 809 |
+
"logits": -0.559980034828186,
|
| 810 |
+
"logps": -647.8008422851562,
|
| 811 |
+
"loss": 0.0014,
|
| 812 |
+
"rank_accuracy_batch": 0.5607143044471741,
|
| 813 |
+
"rank_correct_batch": 15.699999809265137,
|
| 814 |
+
"rank_pair_batch": 28.0,
|
| 815 |
+
"step": 320
|
| 816 |
+
},
|
| 817 |
+
{
|
| 818 |
+
"epoch": 0.6947227788911156,
|
| 819 |
+
"grad_norm": 0.061113730390280376,
|
| 820 |
+
"learning_rate": 1.2825279389890818e-07,
|
| 821 |
+
"logits": -0.5641463398933411,
|
| 822 |
+
"logps": -687.2400512695312,
|
| 823 |
+
"loss": 0.0017,
|
| 824 |
+
"rank_accuracy_batch": 0.5861607193946838,
|
| 825 |
+
"rank_correct_batch": 16.412500381469727,
|
| 826 |
+
"rank_pair_batch": 28.0,
|
| 827 |
+
"step": 325
|
| 828 |
+
},
|
| 829 |
+
{
|
| 830 |
+
"epoch": 0.7054108216432866,
|
| 831 |
+
"grad_norm": 0.05120488122611655,
|
| 832 |
+
"learning_rate": 1.201734872092077e-07,
|
| 833 |
+
"logits": -0.5746558308601379,
|
| 834 |
+
"logps": -658.9490356445312,
|
| 835 |
+
"loss": 0.0018,
|
| 836 |
+
"rank_accuracy_batch": 0.5892857313156128,
|
| 837 |
+
"rank_correct_batch": 16.5,
|
| 838 |
+
"rank_pair_batch": 28.0,
|
| 839 |
+
"step": 330
|
| 840 |
+
},
|
| 841 |
+
{
|
| 842 |
+
"epoch": 0.7160988643954576,
|
| 843 |
+
"grad_norm": 0.049974882977960114,
|
| 844 |
+
"learning_rate": 1.1227575463697439e-07,
|
| 845 |
+
"logits": -0.5673651099205017,
|
| 846 |
+
"logps": -685.9613647460938,
|
| 847 |
+
"loss": 0.0016,
|
| 848 |
+
"rank_accuracy_batch": 0.5714285373687744,
|
| 849 |
+
"rank_correct_batch": 16.0,
|
| 850 |
+
"rank_pair_batch": 28.0,
|
| 851 |
+
"step": 335
|
| 852 |
+
},
|
| 853 |
+
{
|
| 854 |
+
"epoch": 0.7267869071476286,
|
| 855 |
+
"grad_norm": 0.04159244816573028,
|
| 856 |
+
"learning_rate": 1.0457064187534861e-07,
|
| 857 |
+
"logits": -0.5426948666572571,
|
| 858 |
+
"logps": -656.2877197265625,
|
| 859 |
+
"loss": 0.0016,
|
| 860 |
+
"rank_accuracy_batch": 0.6299106478691101,
|
| 861 |
+
"rank_correct_batch": 17.637500762939453,
|
| 862 |
+
"rank_pair_batch": 28.0,
|
| 863 |
+
"step": 340
|
| 864 |
+
},
|
| 865 |
+
{
|
| 866 |
+
"epoch": 0.7374749498997996,
|
| 867 |
+
"grad_norm": 0.0377970499675676,
|
| 868 |
+
"learning_rate": 9.706892522124838e-08,
|
| 869 |
+
"logits": -0.6124971508979797,
|
| 870 |
+
"logps": -677.3993530273438,
|
| 871 |
+
"loss": 0.0018,
|
| 872 |
+
"rank_accuracy_batch": 0.5669642686843872,
|
| 873 |
+
"rank_correct_batch": 15.875,
|
| 874 |
+
"rank_pair_batch": 28.0,
|
| 875 |
+
"step": 345
|
| 876 |
+
},
|
| 877 |
+
{
|
| 878 |
+
"epoch": 0.7481629926519706,
|
| 879 |
+
"grad_norm": 0.05057143601493517,
|
| 880 |
+
"learning_rate": 8.978109650374396e-08,
|
| 881 |
+
"logits": -0.594768226146698,
|
| 882 |
+
"logps": -667.6634521484375,
|
| 883 |
+
"loss": 0.0018,
|
| 884 |
+
"rank_accuracy_batch": 0.5986607074737549,
|
| 885 |
+
"rank_correct_batch": 16.762500762939453,
|
| 886 |
+
"rank_pair_batch": 28.0,
|
| 887 |
+
"step": 350
|
| 888 |
+
},
|
| 889 |
+
{
|
| 890 |
+
"epoch": 0.7588510354041417,
|
| 891 |
+
"grad_norm": 0.040921399914796694,
|
| 892 |
+
"learning_rate": 8.271734841028552e-08,
|
| 893 |
+
"logits": -0.5947057604789734,
|
| 894 |
+
"logps": -662.2523193359375,
|
| 895 |
+
"loss": 0.0016,
|
| 896 |
+
"rank_accuracy_batch": 0.5982142686843872,
|
| 897 |
+
"rank_correct_batch": 16.75,
|
| 898 |
+
"rank_pair_batch": 28.0,
|
| 899 |
+
"step": 355
|
| 900 |
+
},
|
| 901 |
+
{
|
| 902 |
+
"epoch": 0.7695390781563126,
|
| 903 |
+
"grad_norm": 0.06268413493880642,
|
| 904 |
+
"learning_rate": 7.588756023130833e-08,
|
| 905 |
+
"logits": -0.5514832139015198,
|
| 906 |
+
"logps": -698.9403686523438,
|
| 907 |
+
"loss": 0.0016,
|
| 908 |
+
"rank_accuracy_batch": 0.5700892210006714,
|
| 909 |
+
"rank_correct_batch": 15.962499618530273,
|
| 910 |
+
"rank_pair_batch": 28.0,
|
| 911 |
+
"step": 360
|
| 912 |
+
},
|
| 913 |
+
{
|
| 914 |
+
"epoch": 0.7802271209084837,
|
| 915 |
+
"grad_norm": 0.059066915713659185,
|
| 916 |
+
"learning_rate": 6.930128404315214e-08,
|
| 917 |
+
"logits": -0.5609195828437805,
|
| 918 |
+
"logps": -666.8607788085938,
|
| 919 |
+
"loss": 0.0018,
|
| 920 |
+
"rank_accuracy_batch": 0.5544642806053162,
|
| 921 |
+
"rank_correct_batch": 15.524999618530273,
|
| 922 |
+
"rank_pair_batch": 28.0,
|
| 923 |
+
"step": 365
|
| 924 |
+
},
|
| 925 |
+
{
|
| 926 |
+
"epoch": 0.7909151636606546,
|
| 927 |
+
"grad_norm": 0.05124461244763767,
|
| 928 |
+
"learning_rate": 6.296773134861824e-08,
|
| 929 |
+
"logits": -0.6028842926025391,
|
| 930 |
+
"logps": -660.15185546875,
|
| 931 |
+
"loss": 0.0017,
|
| 932 |
+
"rank_accuracy_batch": 0.5941964387893677,
|
| 933 |
+
"rank_correct_batch": 16.637500762939453,
|
| 934 |
+
"rank_pair_batch": 28.0,
|
| 935 |
+
"step": 370
|
| 936 |
+
},
|
| 937 |
+
{
|
| 938 |
+
"epoch": 0.8016032064128257,
|
| 939 |
+
"grad_norm": 0.04353661338852584,
|
| 940 |
+
"learning_rate": 5.6895760193850145e-08,
|
| 941 |
+
"logits": -0.5873713493347168,
|
| 942 |
+
"logps": -678.1385498046875,
|
| 943 |
+
"loss": 0.0017,
|
| 944 |
+
"rank_accuracy_batch": 0.5718749761581421,
|
| 945 |
+
"rank_correct_batch": 16.012500762939453,
|
| 946 |
+
"rank_pair_batch": 28.0,
|
| 947 |
+
"step": 375
|
| 948 |
+
},
|
| 949 |
+
{
|
| 950 |
+
"epoch": 0.8016032064128257,
|
| 951 |
+
"eval_logits": -0.6115520000457764,
|
| 952 |
+
"eval_logps": -662.020263671875,
|
| 953 |
+
"eval_loss": 0.0017035834025591612,
|
| 954 |
+
"eval_rank_accuracy_batch": 0.5901567935943604,
|
| 955 |
+
"eval_rank_correct_batch": 16.524391174316406,
|
| 956 |
+
"eval_rank_pair_batch": 28.0,
|
| 957 |
+
"eval_runtime": 600.3842,
|
| 958 |
+
"eval_samples_per_second": 3.278,
|
| 959 |
+
"eval_steps_per_second": 0.41,
|
| 960 |
+
"step": 375
|
| 961 |
+
},
|
| 962 |
+
{
|
| 963 |
+
"epoch": 0.8122912491649966,
|
| 964 |
+
"grad_norm": 0.05889122643030273,
|
| 965 |
+
"learning_rate": 5.109386277955477e-08,
|
| 966 |
+
"logits": -0.5327494740486145,
|
| 967 |
+
"logps": -670.4705810546875,
|
| 968 |
+
"loss": 0.0016,
|
| 969 |
+
"rank_accuracy_batch": 0.6031249761581421,
|
| 970 |
+
"rank_correct_batch": 16.887500762939453,
|
| 971 |
+
"rank_pair_batch": 28.0,
|
| 972 |
+
"step": 380
|
| 973 |
+
},
|
| 974 |
+
{
|
| 975 |
+
"epoch": 0.8229792919171677,
|
| 976 |
+
"grad_norm": 0.061541748541570206,
|
| 977 |
+
"learning_rate": 4.557015358389216e-08,
|
| 978 |
+
"logits": -0.5657153725624084,
|
| 979 |
+
"logps": -643.9216918945312,
|
| 980 |
+
"loss": 0.0017,
|
| 981 |
+
"rank_accuracy_batch": 0.5910714864730835,
|
| 982 |
+
"rank_correct_batch": 16.549999237060547,
|
| 983 |
+
"rank_pair_batch": 28.0,
|
| 984 |
+
"step": 385
|
| 985 |
+
},
|
| 986 |
+
{
|
| 987 |
+
"epoch": 0.8336673346693386,
|
| 988 |
+
"grad_norm": 0.044590523129162556,
|
| 989 |
+
"learning_rate": 4.0332358013644015e-08,
|
| 990 |
+
"logits": -0.611028254032135,
|
| 991 |
+
"logps": -663.544677734375,
|
| 992 |
+
"loss": 0.0015,
|
| 993 |
+
"rank_accuracy_batch": 0.5924107432365417,
|
| 994 |
+
"rank_correct_batch": 16.587499618530273,
|
| 995 |
+
"rank_pair_batch": 28.0,
|
| 996 |
+
"step": 390
|
| 997 |
+
},
|
| 998 |
+
{
|
| 999 |
+
"epoch": 0.8443553774215097,
|
| 1000 |
+
"grad_norm": 0.053630996492697786,
|
| 1001 |
+
"learning_rate": 3.538780159953347e-08,
|
| 1002 |
+
"logits": -0.5933468341827393,
|
| 1003 |
+
"logps": -651.5979614257812,
|
| 1004 |
+
"loss": 0.0017,
|
| 1005 |
+
"rank_accuracy_batch": 0.592410683631897,
|
| 1006 |
+
"rank_correct_batch": 16.587499618530273,
|
| 1007 |
+
"rank_pair_batch": 28.0,
|
| 1008 |
+
"step": 395
|
| 1009 |
+
},
|
| 1010 |
+
{
|
| 1011 |
+
"epoch": 0.8550434201736807,
|
| 1012 |
+
"grad_norm": 0.053367322510510094,
|
| 1013 |
+
"learning_rate": 3.074339975080836e-08,
|
| 1014 |
+
"logits": -0.5933715105056763,
|
| 1015 |
+
"logps": -664.7044677734375,
|
| 1016 |
+
"loss": 0.0017,
|
| 1017 |
+
"rank_accuracy_batch": 0.6183035373687744,
|
| 1018 |
+
"rank_correct_batch": 17.3125,
|
| 1019 |
+
"rank_pair_batch": 28.0,
|
| 1020 |
+
"step": 400
|
| 1021 |
+
},
|
| 1022 |
+
{
|
| 1023 |
+
"epoch": 0.8657314629258517,
|
| 1024 |
+
"grad_norm": 0.03849991688990423,
|
| 1025 |
+
"learning_rate": 2.6405648083415833e-08,
|
| 1026 |
+
"logits": -0.5992918610572815,
|
| 1027 |
+
"logps": -686.837158203125,
|
| 1028 |
+
"loss": 0.0017,
|
| 1029 |
+
"rank_accuracy_batch": 0.5808035731315613,
|
| 1030 |
+
"rank_correct_batch": 16.262500762939453,
|
| 1031 |
+
"rank_pair_batch": 28.0,
|
| 1032 |
+
"step": 405
|
| 1033 |
+
},
|
| 1034 |
+
{
|
| 1035 |
+
"epoch": 0.8764195056780227,
|
| 1036 |
+
"grad_norm": 0.03299753297821205,
|
| 1037 |
+
"learning_rate": 2.2380613335296033e-08,
|
| 1038 |
+
"logits": -0.5921497344970703,
|
| 1039 |
+
"logps": -658.4564819335938,
|
| 1040 |
+
"loss": 0.0019,
|
| 1041 |
+
"rank_accuracy_batch": 0.6089285612106323,
|
| 1042 |
+
"rank_correct_batch": 17.049999237060547,
|
| 1043 |
+
"rank_pair_batch": 28.0,
|
| 1044 |
+
"step": 410
|
| 1045 |
+
},
|
| 1046 |
+
{
|
| 1047 |
+
"epoch": 0.8871075484301937,
|
| 1048 |
+
"grad_norm": 0.05993595081853875,
|
| 1049 |
+
"learning_rate": 1.8673924881500823e-08,
|
| 1050 |
+
"logits": -0.6167685985565186,
|
| 1051 |
+
"logps": -657.6021118164062,
|
| 1052 |
+
"loss": 0.0017,
|
| 1053 |
+
"rank_accuracy_batch": 0.621874988079071,
|
| 1054 |
+
"rank_correct_batch": 17.412500381469727,
|
| 1055 |
+
"rank_pair_batch": 28.0,
|
| 1056 |
+
"step": 415
|
| 1057 |
+
},
|
| 1058 |
+
{
|
| 1059 |
+
"epoch": 0.8977955911823647,
|
| 1060 |
+
"grad_norm": 0.10580724292169638,
|
| 1061 |
+
"learning_rate": 1.5290766861003475e-08,
|
| 1062 |
+
"logits": -0.608423113822937,
|
| 1063 |
+
"logps": -666.0797729492188,
|
| 1064 |
+
"loss": 0.0021,
|
| 1065 |
+
"rank_accuracy_batch": 0.6267857551574707,
|
| 1066 |
+
"rank_correct_batch": 17.549999237060547,
|
| 1067 |
+
"rank_pair_batch": 28.0,
|
| 1068 |
+
"step": 420
|
| 1069 |
+
},
|
| 1070 |
+
{
|
| 1071 |
+
"epoch": 0.9084836339345357,
|
| 1072 |
+
"grad_norm": 0.05179962147512424,
|
| 1073 |
+
"learning_rate": 1.2235870926211616e-08,
|
| 1074 |
+
"logits": -0.5300483107566833,
|
| 1075 |
+
"logps": -674.7176513671875,
|
| 1076 |
+
"loss": 0.0016,
|
| 1077 |
+
"rank_accuracy_batch": 0.5964285731315613,
|
| 1078 |
+
"rank_correct_batch": 16.700000762939453,
|
| 1079 |
+
"rank_pair_batch": 28.0,
|
| 1080 |
+
"step": 425
|
| 1081 |
+
},
|
| 1082 |
+
{
|
| 1083 |
+
"epoch": 0.9191716766867067,
|
| 1084 |
+
"grad_norm": 0.05533039782642434,
|
| 1085 |
+
"learning_rate": 9.513509625323518e-09,
|
| 1086 |
+
"logits": -0.5916275978088379,
|
| 1087 |
+
"logps": -654.2515869140625,
|
| 1088 |
+
"loss": 0.0017,
|
| 1089 |
+
"rank_accuracy_batch": 0.6165178418159485,
|
| 1090 |
+
"rank_correct_batch": 17.262500762939453,
|
| 1091 |
+
"rank_pair_batch": 28.0,
|
| 1092 |
+
"step": 430
|
| 1093 |
+
},
|
| 1094 |
+
{
|
| 1095 |
+
"epoch": 0.9298597194388778,
|
| 1096 |
+
"grad_norm": 0.045085582087942604,
|
| 1097 |
+
"learning_rate": 7.127490426783123e-09,
|
| 1098 |
+
"logits": -0.6653744578361511,
|
| 1099 |
+
"logps": -658.5715942382812,
|
| 1100 |
+
"loss": 0.0017,
|
| 1101 |
+
"rank_accuracy_batch": 0.5919642448425293,
|
| 1102 |
+
"rank_correct_batch": 16.575000762939453,
|
| 1103 |
+
"rank_pair_batch": 28.0,
|
| 1104 |
+
"step": 435
|
| 1105 |
+
},
|
| 1106 |
+
{
|
| 1107 |
+
"epoch": 0.9405477621910487,
|
| 1108 |
+
"grad_norm": 0.07434339207451536,
|
| 1109 |
+
"learning_rate": 5.08115039419113e-09,
|
| 1110 |
+
"logits": -0.6592944264411926,
|
| 1111 |
+
"logps": -648.865966796875,
|
| 1112 |
+
"loss": 0.0016,
|
| 1113 |
+
"rank_accuracy_batch": 0.6058035492897034,
|
| 1114 |
+
"rank_correct_batch": 16.962499618530273,
|
| 1115 |
+
"rank_pair_batch": 28.0,
|
| 1116 |
+
"step": 440
|
| 1117 |
+
},
|
| 1118 |
+
{
|
| 1119 |
+
"epoch": 0.9512358049432198,
|
| 1120 |
+
"grad_norm": 0.057951436930576285,
|
| 1121 |
+
"learning_rate": 3.3773515191196646e-09,
|
| 1122 |
+
"logits": -0.4745956361293793,
|
| 1123 |
+
"logps": -659.5784912109375,
|
| 1124 |
+
"loss": 0.0018,
|
| 1125 |
+
"rank_accuracy_batch": 0.6017856597900391,
|
| 1126 |
+
"rank_correct_batch": 16.850000381469727,
|
| 1127 |
+
"rank_pair_batch": 28.0,
|
| 1128 |
+
"step": 445
|
| 1129 |
+
},
|
| 1130 |
+
{
|
| 1131 |
+
"epoch": 0.9619238476953907,
|
| 1132 |
+
"grad_norm": 0.10661448567775118,
|
| 1133 |
+
"learning_rate": 2.0184767183584474e-09,
|
| 1134 |
+
"logits": -0.5836338996887207,
|
| 1135 |
+
"logps": -637.9088134765625,
|
| 1136 |
+
"loss": 0.002,
|
| 1137 |
+
"rank_accuracy_batch": 0.628125011920929,
|
| 1138 |
+
"rank_correct_batch": 17.587499618530273,
|
| 1139 |
+
"rank_pair_batch": 28.0,
|
| 1140 |
+
"step": 450
|
| 1141 |
+
},
|
| 1142 |
+
{
|
| 1143 |
+
"epoch": 0.9726118904475618,
|
| 1144 |
+
"grad_norm": 0.043661597735003485,
|
| 1145 |
+
"learning_rate": 1.0064265011902328e-09,
|
| 1146 |
+
"logits": -0.5709097385406494,
|
| 1147 |
+
"logps": -651.0455322265625,
|
| 1148 |
+
"loss": 0.0018,
|
| 1149 |
+
"rank_accuracy_batch": 0.6285714507102966,
|
| 1150 |
+
"rank_correct_batch": 17.600000381469727,
|
| 1151 |
+
"rank_pair_batch": 28.0,
|
| 1152 |
+
"step": 455
|
| 1153 |
+
},
|
| 1154 |
+
{
|
| 1155 |
+
"epoch": 0.9832999331997327,
|
| 1156 |
+
"grad_norm": 0.04040440877903873,
|
| 1157 |
+
"learning_rate": 3.4261631135654167e-10,
|
| 1158 |
+
"logits": -0.5855480432510376,
|
| 1159 |
+
"logps": -654.5772705078125,
|
| 1160 |
+
"loss": 0.0017,
|
| 1161 |
+
"rank_accuracy_batch": 0.6035714745521545,
|
| 1162 |
+
"rank_correct_batch": 16.899999618530273,
|
| 1163 |
+
"rank_pair_batch": 28.0,
|
| 1164 |
+
"step": 460
|
| 1165 |
+
},
|
| 1166 |
+
{
|
| 1167 |
+
"epoch": 0.9939879759519038,
|
| 1168 |
+
"grad_norm": 0.0866710364530591,
|
| 1169 |
+
"learning_rate": 2.797454743164174e-11,
|
| 1170 |
+
"logits": -0.6281521916389465,
|
| 1171 |
+
"logps": -675.218505859375,
|
| 1172 |
+
"loss": 0.0019,
|
| 1173 |
+
"rank_accuracy_batch": 0.6071428060531616,
|
| 1174 |
+
"rank_correct_batch": 17.0,
|
| 1175 |
+
"rank_pair_batch": 28.0,
|
| 1176 |
+
"step": 465
|
| 1177 |
+
},
|
| 1178 |
+
{
|
| 1179 |
+
"epoch": 0.9982631930527722,
|
| 1180 |
+
"step": 467,
|
| 1181 |
+
"total_flos": 0.0,
|
| 1182 |
+
"train_loss": 0.0024812357072425285,
|
| 1183 |
+
"train_runtime": 43392.086,
|
| 1184 |
+
"train_samples_per_second": 1.38,
|
| 1185 |
+
"train_steps_per_second": 0.011
|
| 1186 |
+
}
|
| 1187 |
+
],
|
| 1188 |
+
"logging_steps": 5,
|
| 1189 |
+
"max_steps": 467,
|
| 1190 |
+
"num_input_tokens_seen": 0,
|
| 1191 |
+
"num_train_epochs": 1,
|
| 1192 |
+
"save_steps": 1000,
|
| 1193 |
+
"stateful_callbacks": {
|
| 1194 |
+
"TrainerControl": {
|
| 1195 |
+
"args": {
|
| 1196 |
+
"should_epoch_stop": false,
|
| 1197 |
+
"should_evaluate": false,
|
| 1198 |
+
"should_log": false,
|
| 1199 |
+
"should_save": true,
|
| 1200 |
+
"should_training_stop": true
|
| 1201 |
+
},
|
| 1202 |
+
"attributes": {}
|
| 1203 |
+
}
|
| 1204 |
+
},
|
| 1205 |
+
"total_flos": 0.0,
|
| 1206 |
+
"train_batch_size": 1,
|
| 1207 |
+
"trial_name": null,
|
| 1208 |
+
"trial_params": null
|
| 1209 |
+
}
|