End of training
Browse files- README.md +2 -1
- all_results.json +12 -0
- eval_results.json +7 -0
- train_results.json +8 -0
- trainer_state.json +115 -0
- training_eval_loss.png +0 -0
- training_loss.png +0 -0
README.md
CHANGED
@@ -4,6 +4,7 @@ license: llama3.1
|
|
4 |
base_model: meta-llama/Meta-Llama-3.1-8B
|
5 |
tags:
|
6 |
- llama-factory
|
|
|
7 |
- generated_from_trainer
|
8 |
model-index:
|
9 |
- name: stackexchange_engineering
|
@@ -15,7 +16,7 @@ should probably proofread and complete it, then remove this comment. -->
|
|
15 |
|
16 |
# stackexchange_engineering
|
17 |
|
18 |
-
This model is a fine-tuned version of [meta-llama/Meta-Llama-3.1-8B](https://huggingface.co/meta-llama/Meta-Llama-3.1-8B) on
|
19 |
It achieves the following results on the evaluation set:
|
20 |
- Loss: 0.9366
|
21 |
|
|
|
4 |
base_model: meta-llama/Meta-Llama-3.1-8B
|
5 |
tags:
|
6 |
- llama-factory
|
7 |
+
- full
|
8 |
- generated_from_trainer
|
9 |
model-index:
|
10 |
- name: stackexchange_engineering
|
|
|
16 |
|
17 |
# stackexchange_engineering
|
18 |
|
19 |
+
This model is a fine-tuned version of [meta-llama/Meta-Llama-3.1-8B](https://huggingface.co/meta-llama/Meta-Llama-3.1-8B) on the mlfoundations-dev/stackexchange_engineering dataset.
|
20 |
It achieves the following results on the evaluation set:
|
21 |
- Loss: 0.9366
|
22 |
|
all_results.json
ADDED
@@ -0,0 +1,12 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
1 |
+
{
|
2 |
+
"epoch": 2.926829268292683,
|
3 |
+
"eval_loss": 0.9366496801376343,
|
4 |
+
"eval_runtime": 17.6253,
|
5 |
+
"eval_samples_per_second": 39.205,
|
6 |
+
"eval_steps_per_second": 0.624,
|
7 |
+
"total_flos": 125418413752320.0,
|
8 |
+
"train_loss": 0.9610127830505371,
|
9 |
+
"train_runtime": 2801.3394,
|
10 |
+
"train_samples_per_second": 14.047,
|
11 |
+
"train_steps_per_second": 0.027
|
12 |
+
}
|
eval_results.json
ADDED
@@ -0,0 +1,7 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
1 |
+
{
|
2 |
+
"epoch": 2.926829268292683,
|
3 |
+
"eval_loss": 0.9366496801376343,
|
4 |
+
"eval_runtime": 17.6253,
|
5 |
+
"eval_samples_per_second": 39.205,
|
6 |
+
"eval_steps_per_second": 0.624
|
7 |
+
}
|
train_results.json
ADDED
@@ -0,0 +1,8 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
1 |
+
{
|
2 |
+
"epoch": 2.926829268292683,
|
3 |
+
"total_flos": 125418413752320.0,
|
4 |
+
"train_loss": 0.9610127830505371,
|
5 |
+
"train_runtime": 2801.3394,
|
6 |
+
"train_samples_per_second": 14.047,
|
7 |
+
"train_steps_per_second": 0.027
|
8 |
+
}
|
trainer_state.json
ADDED
@@ -0,0 +1,115 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
1 |
+
{
|
2 |
+
"best_metric": null,
|
3 |
+
"best_model_checkpoint": null,
|
4 |
+
"epoch": 2.926829268292683,
|
5 |
+
"eval_steps": 500,
|
6 |
+
"global_step": 75,
|
7 |
+
"is_hyper_param_search": false,
|
8 |
+
"is_local_process_zero": true,
|
9 |
+
"is_world_process_zero": true,
|
10 |
+
"log_history": [
|
11 |
+
{
|
12 |
+
"epoch": 0.3902439024390244,
|
13 |
+
"grad_norm": 6.465833920476205,
|
14 |
+
"learning_rate": 5e-06,
|
15 |
+
"loss": 1.1247,
|
16 |
+
"step": 10
|
17 |
+
},
|
18 |
+
{
|
19 |
+
"epoch": 0.7804878048780488,
|
20 |
+
"grad_norm": 1.1862331772202763,
|
21 |
+
"learning_rate": 5e-06,
|
22 |
+
"loss": 1.018,
|
23 |
+
"step": 20
|
24 |
+
},
|
25 |
+
{
|
26 |
+
"epoch": 0.975609756097561,
|
27 |
+
"eval_loss": 0.9846493005752563,
|
28 |
+
"eval_runtime": 18.0366,
|
29 |
+
"eval_samples_per_second": 38.311,
|
30 |
+
"eval_steps_per_second": 0.61,
|
31 |
+
"step": 25
|
32 |
+
},
|
33 |
+
{
|
34 |
+
"epoch": 1.170731707317073,
|
35 |
+
"grad_norm": 2.162191793586483,
|
36 |
+
"learning_rate": 5e-06,
|
37 |
+
"loss": 1.0085,
|
38 |
+
"step": 30
|
39 |
+
},
|
40 |
+
{
|
41 |
+
"epoch": 1.5609756097560976,
|
42 |
+
"grad_norm": 1.9535821700087348,
|
43 |
+
"learning_rate": 5e-06,
|
44 |
+
"loss": 0.9331,
|
45 |
+
"step": 40
|
46 |
+
},
|
47 |
+
{
|
48 |
+
"epoch": 1.951219512195122,
|
49 |
+
"grad_norm": 1.124539760660556,
|
50 |
+
"learning_rate": 5e-06,
|
51 |
+
"loss": 0.9171,
|
52 |
+
"step": 50
|
53 |
+
},
|
54 |
+
{
|
55 |
+
"epoch": 1.9902439024390244,
|
56 |
+
"eval_loss": 0.9493349194526672,
|
57 |
+
"eval_runtime": 18.0567,
|
58 |
+
"eval_samples_per_second": 38.268,
|
59 |
+
"eval_steps_per_second": 0.609,
|
60 |
+
"step": 51
|
61 |
+
},
|
62 |
+
{
|
63 |
+
"epoch": 2.341463414634146,
|
64 |
+
"grad_norm": 0.9405210346947177,
|
65 |
+
"learning_rate": 5e-06,
|
66 |
+
"loss": 0.9107,
|
67 |
+
"step": 60
|
68 |
+
},
|
69 |
+
{
|
70 |
+
"epoch": 2.7317073170731705,
|
71 |
+
"grad_norm": 1.8443181229715055,
|
72 |
+
"learning_rate": 5e-06,
|
73 |
+
"loss": 0.8594,
|
74 |
+
"step": 70
|
75 |
+
},
|
76 |
+
{
|
77 |
+
"epoch": 2.926829268292683,
|
78 |
+
"eval_loss": 0.9366496801376343,
|
79 |
+
"eval_runtime": 16.7048,
|
80 |
+
"eval_samples_per_second": 41.365,
|
81 |
+
"eval_steps_per_second": 0.658,
|
82 |
+
"step": 75
|
83 |
+
},
|
84 |
+
{
|
85 |
+
"epoch": 2.926829268292683,
|
86 |
+
"step": 75,
|
87 |
+
"total_flos": 125418413752320.0,
|
88 |
+
"train_loss": 0.9610127830505371,
|
89 |
+
"train_runtime": 2801.3394,
|
90 |
+
"train_samples_per_second": 14.047,
|
91 |
+
"train_steps_per_second": 0.027
|
92 |
+
}
|
93 |
+
],
|
94 |
+
"logging_steps": 10,
|
95 |
+
"max_steps": 75,
|
96 |
+
"num_input_tokens_seen": 0,
|
97 |
+
"num_train_epochs": 3,
|
98 |
+
"save_steps": 500,
|
99 |
+
"stateful_callbacks": {
|
100 |
+
"TrainerControl": {
|
101 |
+
"args": {
|
102 |
+
"should_epoch_stop": false,
|
103 |
+
"should_evaluate": false,
|
104 |
+
"should_log": false,
|
105 |
+
"should_save": true,
|
106 |
+
"should_training_stop": true
|
107 |
+
},
|
108 |
+
"attributes": {}
|
109 |
+
}
|
110 |
+
},
|
111 |
+
"total_flos": 125418413752320.0,
|
112 |
+
"train_batch_size": 8,
|
113 |
+
"trial_name": null,
|
114 |
+
"trial_params": null
|
115 |
+
}
|
training_eval_loss.png
ADDED
![]() |
training_loss.png
ADDED
![]() |