Model save
Browse files- README.md +70 -0
- all_results.json +9 -0
- generation_config.json +6 -0
- train_results.json +9 -0
- trainer_state.json +122 -0
README.md
ADDED
@@ -0,0 +1,70 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
1 |
+
---
|
2 |
+
library_name: transformers
|
3 |
+
license: apache-2.0
|
4 |
+
base_model: mistralai/Mistral-Nemo-Base-2407
|
5 |
+
tags:
|
6 |
+
- trl
|
7 |
+
- sft
|
8 |
+
- generated_from_trainer
|
9 |
+
datasets:
|
10 |
+
- generator
|
11 |
+
model-index:
|
12 |
+
- name: Mistral-Nemo-Instruct-Baseline-SFT
|
13 |
+
results: []
|
14 |
+
---
|
15 |
+
|
16 |
+
<!-- This model card has been generated automatically according to the information the Trainer had access to. You
|
17 |
+
should probably proofread and complete it, then remove this comment. -->
|
18 |
+
|
19 |
+
# Mistral-Nemo-Instruct-Baseline-SFT
|
20 |
+
|
21 |
+
This model is a fine-tuned version of [mistralai/Mistral-Nemo-Base-2407](https://huggingface.co/mistralai/Mistral-Nemo-Base-2407) on the generator dataset.
|
22 |
+
It achieves the following results on the evaluation set:
|
23 |
+
- Loss: 1.1750
|
24 |
+
|
25 |
+
## Model description
|
26 |
+
|
27 |
+
More information needed
|
28 |
+
|
29 |
+
## Intended uses & limitations
|
30 |
+
|
31 |
+
More information needed
|
32 |
+
|
33 |
+
## Training and evaluation data
|
34 |
+
|
35 |
+
More information needed
|
36 |
+
|
37 |
+
## Training procedure
|
38 |
+
|
39 |
+
### Training hyperparameters
|
40 |
+
|
41 |
+
The following hyperparameters were used during training:
|
42 |
+
- learning_rate: 2e-05
|
43 |
+
- train_batch_size: 8
|
44 |
+
- eval_batch_size: 8
|
45 |
+
- seed: 42
|
46 |
+
- distributed_type: multi-GPU
|
47 |
+
- num_devices: 8
|
48 |
+
- gradient_accumulation_steps: 4
|
49 |
+
- total_train_batch_size: 256
|
50 |
+
- total_eval_batch_size: 64
|
51 |
+
- optimizer: Adam with betas=(0.9,0.999) and epsilon=1e-08
|
52 |
+
- lr_scheduler_type: cosine
|
53 |
+
- lr_scheduler_warmup_ratio: 0.1
|
54 |
+
- num_epochs: 3
|
55 |
+
|
56 |
+
### Training results
|
57 |
+
|
58 |
+
| Training Loss | Epoch | Step | Validation Loss |
|
59 |
+
|:-------------:|:------:|:----:|:---------------:|
|
60 |
+
| 1.6158 | 0.9455 | 13 | 1.3800 |
|
61 |
+
| 1.1061 | 1.9636 | 27 | 1.1854 |
|
62 |
+
| 0.9071 | 2.8364 | 39 | 1.1750 |
|
63 |
+
|
64 |
+
|
65 |
+
### Framework versions
|
66 |
+
|
67 |
+
- Transformers 4.45.1
|
68 |
+
- Pytorch 2.4.1+cu121
|
69 |
+
- Datasets 3.0.1
|
70 |
+
- Tokenizers 0.20.0
|
all_results.json
ADDED
@@ -0,0 +1,9 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
1 |
+
{
|
2 |
+
"epoch": 2.8363636363636364,
|
3 |
+
"total_flos": 50553107251200.0,
|
4 |
+
"train_loss": 1.237262059480716,
|
5 |
+
"train_runtime": 1937.0641,
|
6 |
+
"train_samples": 18999,
|
7 |
+
"train_samples_per_second": 5.391,
|
8 |
+
"train_steps_per_second": 0.02
|
9 |
+
}
|
generation_config.json
ADDED
@@ -0,0 +1,6 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
1 |
+
{
|
2 |
+
"_from_model_config": true,
|
3 |
+
"bos_token_id": 1,
|
4 |
+
"eos_token_id": 2,
|
5 |
+
"transformers_version": "4.45.1"
|
6 |
+
}
|
train_results.json
ADDED
@@ -0,0 +1,9 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
1 |
+
{
|
2 |
+
"epoch": 2.8363636363636364,
|
3 |
+
"total_flos": 50553107251200.0,
|
4 |
+
"train_loss": 1.237262059480716,
|
5 |
+
"train_runtime": 1937.0641,
|
6 |
+
"train_samples": 18999,
|
7 |
+
"train_samples_per_second": 5.391,
|
8 |
+
"train_steps_per_second": 0.02
|
9 |
+
}
|
trainer_state.json
ADDED
@@ -0,0 +1,122 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
1 |
+
{
|
2 |
+
"best_metric": null,
|
3 |
+
"best_model_checkpoint": null,
|
4 |
+
"epoch": 2.8363636363636364,
|
5 |
+
"eval_steps": 500,
|
6 |
+
"global_step": 39,
|
7 |
+
"is_hyper_param_search": false,
|
8 |
+
"is_local_process_zero": true,
|
9 |
+
"is_world_process_zero": true,
|
10 |
+
"log_history": [
|
11 |
+
{
|
12 |
+
"epoch": 0.07272727272727272,
|
13 |
+
"grad_norm": 40.167989617647805,
|
14 |
+
"learning_rate": 5e-06,
|
15 |
+
"loss": 1.8364,
|
16 |
+
"step": 1
|
17 |
+
},
|
18 |
+
{
|
19 |
+
"epoch": 0.36363636363636365,
|
20 |
+
"grad_norm": 24.721297347748504,
|
21 |
+
"learning_rate": 1.9959742939952393e-05,
|
22 |
+
"loss": 1.7021,
|
23 |
+
"step": 5
|
24 |
+
},
|
25 |
+
{
|
26 |
+
"epoch": 0.7272727272727273,
|
27 |
+
"grad_norm": 5.200831471943857,
|
28 |
+
"learning_rate": 1.8584487936018663e-05,
|
29 |
+
"loss": 1.6158,
|
30 |
+
"step": 10
|
31 |
+
},
|
32 |
+
{
|
33 |
+
"epoch": 0.9454545454545454,
|
34 |
+
"eval_loss": 1.380013108253479,
|
35 |
+
"eval_runtime": 17.9955,
|
36 |
+
"eval_samples_per_second": 19.838,
|
37 |
+
"eval_steps_per_second": 0.333,
|
38 |
+
"step": 13
|
39 |
+
},
|
40 |
+
{
|
41 |
+
"epoch": 1.0909090909090908,
|
42 |
+
"grad_norm": 8.417480656354106,
|
43 |
+
"learning_rate": 1.5508969814521026e-05,
|
44 |
+
"loss": 1.3717,
|
45 |
+
"step": 15
|
46 |
+
},
|
47 |
+
{
|
48 |
+
"epoch": 1.4545454545454546,
|
49 |
+
"grad_norm": 2.9015403857161433,
|
50 |
+
"learning_rate": 1.1342332658176556e-05,
|
51 |
+
"loss": 1.194,
|
52 |
+
"step": 20
|
53 |
+
},
|
54 |
+
{
|
55 |
+
"epoch": 1.8181818181818183,
|
56 |
+
"grad_norm": 1.4331768387247874,
|
57 |
+
"learning_rate": 6.909830056250527e-06,
|
58 |
+
"loss": 1.1061,
|
59 |
+
"step": 25
|
60 |
+
},
|
61 |
+
{
|
62 |
+
"epoch": 1.9636363636363636,
|
63 |
+
"eval_loss": 1.1853971481323242,
|
64 |
+
"eval_runtime": 17.707,
|
65 |
+
"eval_samples_per_second": 20.161,
|
66 |
+
"eval_steps_per_second": 0.339,
|
67 |
+
"step": 27
|
68 |
+
},
|
69 |
+
{
|
70 |
+
"epoch": 2.1818181818181817,
|
71 |
+
"grad_norm": 1.2443658592960614,
|
72 |
+
"learning_rate": 3.089373510131354e-06,
|
73 |
+
"loss": 1.0003,
|
74 |
+
"step": 30
|
75 |
+
},
|
76 |
+
{
|
77 |
+
"epoch": 2.5454545454545454,
|
78 |
+
"grad_norm": 0.8568687436721839,
|
79 |
+
"learning_rate": 6.37651293602628e-07,
|
80 |
+
"loss": 0.9071,
|
81 |
+
"step": 35
|
82 |
+
},
|
83 |
+
{
|
84 |
+
"epoch": 2.8363636363636364,
|
85 |
+
"eval_loss": 1.1750216484069824,
|
86 |
+
"eval_runtime": 17.732,
|
87 |
+
"eval_samples_per_second": 20.133,
|
88 |
+
"eval_steps_per_second": 0.338,
|
89 |
+
"step": 39
|
90 |
+
},
|
91 |
+
{
|
92 |
+
"epoch": 2.8363636363636364,
|
93 |
+
"step": 39,
|
94 |
+
"total_flos": 50553107251200.0,
|
95 |
+
"train_loss": 1.237262059480716,
|
96 |
+
"train_runtime": 1937.0641,
|
97 |
+
"train_samples_per_second": 5.391,
|
98 |
+
"train_steps_per_second": 0.02
|
99 |
+
}
|
100 |
+
],
|
101 |
+
"logging_steps": 5,
|
102 |
+
"max_steps": 39,
|
103 |
+
"num_input_tokens_seen": 0,
|
104 |
+
"num_train_epochs": 3,
|
105 |
+
"save_steps": 100,
|
106 |
+
"stateful_callbacks": {
|
107 |
+
"TrainerControl": {
|
108 |
+
"args": {
|
109 |
+
"should_epoch_stop": false,
|
110 |
+
"should_evaluate": false,
|
111 |
+
"should_log": false,
|
112 |
+
"should_save": true,
|
113 |
+
"should_training_stop": true
|
114 |
+
},
|
115 |
+
"attributes": {}
|
116 |
+
}
|
117 |
+
},
|
118 |
+
"total_flos": 50553107251200.0,
|
119 |
+
"train_batch_size": 8,
|
120 |
+
"trial_name": null,
|
121 |
+
"trial_params": null
|
122 |
+
}
|