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 +150 -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_dsp
|
@@ -15,7 +16,7 @@ should probably proofread and complete it, then remove this comment. -->
|
|
15 |
|
16 |
# stackexchange_dsp
|
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: 0.8000
|
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_dsp
|
|
|
16 |
|
17 |
# stackexchange_dsp
|
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_dsp dataset.
|
20 |
It achieves the following results on the evaluation set:
|
21 |
- Loss: 0.8000
|
22 |
|
all_results.json
ADDED
@@ -0,0 +1,12 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
1 |
+
{
|
2 |
+
"epoch": 2.9822485207100593,
|
3 |
+
"eval_loss": 0.7999942302703857,
|
4 |
+
"eval_runtime": 28.6817,
|
5 |
+
"eval_samples_per_second": 39.642,
|
6 |
+
"eval_steps_per_second": 0.628,
|
7 |
+
"total_flos": 210845313269760.0,
|
8 |
+
"train_loss": 0.8213065730200874,
|
9 |
+
"train_runtime": 4456.5245,
|
10 |
+
"train_samples_per_second": 14.53,
|
11 |
+
"train_steps_per_second": 0.028
|
12 |
+
}
|
eval_results.json
ADDED
@@ -0,0 +1,7 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
1 |
+
{
|
2 |
+
"epoch": 2.9822485207100593,
|
3 |
+
"eval_loss": 0.7999942302703857,
|
4 |
+
"eval_runtime": 28.6817,
|
5 |
+
"eval_samples_per_second": 39.642,
|
6 |
+
"eval_steps_per_second": 0.628
|
7 |
+
}
|
train_results.json
ADDED
@@ -0,0 +1,8 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
1 |
+
{
|
2 |
+
"epoch": 2.9822485207100593,
|
3 |
+
"total_flos": 210845313269760.0,
|
4 |
+
"train_loss": 0.8213065730200874,
|
5 |
+
"train_runtime": 4456.5245,
|
6 |
+
"train_samples_per_second": 14.53,
|
7 |
+
"train_steps_per_second": 0.028
|
8 |
+
}
|
trainer_state.json
ADDED
@@ -0,0 +1,150 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
1 |
+
{
|
2 |
+
"best_metric": null,
|
3 |
+
"best_model_checkpoint": null,
|
4 |
+
"epoch": 2.9822485207100593,
|
5 |
+
"eval_steps": 500,
|
6 |
+
"global_step": 126,
|
7 |
+
"is_hyper_param_search": false,
|
8 |
+
"is_local_process_zero": true,
|
9 |
+
"is_world_process_zero": true,
|
10 |
+
"log_history": [
|
11 |
+
{
|
12 |
+
"epoch": 0.23668639053254437,
|
13 |
+
"grad_norm": 1.9313669247614162,
|
14 |
+
"learning_rate": 5e-06,
|
15 |
+
"loss": 1.018,
|
16 |
+
"step": 10
|
17 |
+
},
|
18 |
+
{
|
19 |
+
"epoch": 0.47337278106508873,
|
20 |
+
"grad_norm": 1.379427316659985,
|
21 |
+
"learning_rate": 5e-06,
|
22 |
+
"loss": 0.9191,
|
23 |
+
"step": 20
|
24 |
+
},
|
25 |
+
{
|
26 |
+
"epoch": 0.7100591715976331,
|
27 |
+
"grad_norm": 1.6079229642198343,
|
28 |
+
"learning_rate": 5e-06,
|
29 |
+
"loss": 0.8677,
|
30 |
+
"step": 30
|
31 |
+
},
|
32 |
+
{
|
33 |
+
"epoch": 0.9467455621301775,
|
34 |
+
"grad_norm": 0.734643863022887,
|
35 |
+
"learning_rate": 5e-06,
|
36 |
+
"loss": 0.8493,
|
37 |
+
"step": 40
|
38 |
+
},
|
39 |
+
{
|
40 |
+
"epoch": 0.9940828402366864,
|
41 |
+
"eval_loss": 0.8392977118492126,
|
42 |
+
"eval_runtime": 30.1115,
|
43 |
+
"eval_samples_per_second": 37.76,
|
44 |
+
"eval_steps_per_second": 0.598,
|
45 |
+
"step": 42
|
46 |
+
},
|
47 |
+
{
|
48 |
+
"epoch": 1.183431952662722,
|
49 |
+
"grad_norm": 1.005967406304035,
|
50 |
+
"learning_rate": 5e-06,
|
51 |
+
"loss": 0.8734,
|
52 |
+
"step": 50
|
53 |
+
},
|
54 |
+
{
|
55 |
+
"epoch": 1.4201183431952662,
|
56 |
+
"grad_norm": 0.9370549861399187,
|
57 |
+
"learning_rate": 5e-06,
|
58 |
+
"loss": 0.7941,
|
59 |
+
"step": 60
|
60 |
+
},
|
61 |
+
{
|
62 |
+
"epoch": 1.6568047337278107,
|
63 |
+
"grad_norm": 0.7056985035855818,
|
64 |
+
"learning_rate": 5e-06,
|
65 |
+
"loss": 0.787,
|
66 |
+
"step": 70
|
67 |
+
},
|
68 |
+
{
|
69 |
+
"epoch": 1.893491124260355,
|
70 |
+
"grad_norm": 1.2176250359078529,
|
71 |
+
"learning_rate": 5e-06,
|
72 |
+
"loss": 0.7843,
|
73 |
+
"step": 80
|
74 |
+
},
|
75 |
+
{
|
76 |
+
"epoch": 1.9881656804733727,
|
77 |
+
"eval_loss": 0.8069051504135132,
|
78 |
+
"eval_runtime": 29.6227,
|
79 |
+
"eval_samples_per_second": 38.383,
|
80 |
+
"eval_steps_per_second": 0.608,
|
81 |
+
"step": 84
|
82 |
+
},
|
83 |
+
{
|
84 |
+
"epoch": 2.1301775147928996,
|
85 |
+
"grad_norm": 1.0149726634752274,
|
86 |
+
"learning_rate": 5e-06,
|
87 |
+
"loss": 0.8153,
|
88 |
+
"step": 90
|
89 |
+
},
|
90 |
+
{
|
91 |
+
"epoch": 2.366863905325444,
|
92 |
+
"grad_norm": 1.1966616271031878,
|
93 |
+
"learning_rate": 5e-06,
|
94 |
+
"loss": 0.7379,
|
95 |
+
"step": 100
|
96 |
+
},
|
97 |
+
{
|
98 |
+
"epoch": 2.603550295857988,
|
99 |
+
"grad_norm": 1.28942414214322,
|
100 |
+
"learning_rate": 5e-06,
|
101 |
+
"loss": 0.7355,
|
102 |
+
"step": 110
|
103 |
+
},
|
104 |
+
{
|
105 |
+
"epoch": 2.8402366863905324,
|
106 |
+
"grad_norm": 0.6282115192314033,
|
107 |
+
"learning_rate": 5e-06,
|
108 |
+
"loss": 0.7323,
|
109 |
+
"step": 120
|
110 |
+
},
|
111 |
+
{
|
112 |
+
"epoch": 2.9822485207100593,
|
113 |
+
"eval_loss": 0.7999942302703857,
|
114 |
+
"eval_runtime": 27.42,
|
115 |
+
"eval_samples_per_second": 41.466,
|
116 |
+
"eval_steps_per_second": 0.656,
|
117 |
+
"step": 126
|
118 |
+
},
|
119 |
+
{
|
120 |
+
"epoch": 2.9822485207100593,
|
121 |
+
"step": 126,
|
122 |
+
"total_flos": 210845313269760.0,
|
123 |
+
"train_loss": 0.8213065730200874,
|
124 |
+
"train_runtime": 4456.5245,
|
125 |
+
"train_samples_per_second": 14.53,
|
126 |
+
"train_steps_per_second": 0.028
|
127 |
+
}
|
128 |
+
],
|
129 |
+
"logging_steps": 10,
|
130 |
+
"max_steps": 126,
|
131 |
+
"num_input_tokens_seen": 0,
|
132 |
+
"num_train_epochs": 3,
|
133 |
+
"save_steps": 500,
|
134 |
+
"stateful_callbacks": {
|
135 |
+
"TrainerControl": {
|
136 |
+
"args": {
|
137 |
+
"should_epoch_stop": false,
|
138 |
+
"should_evaluate": false,
|
139 |
+
"should_log": false,
|
140 |
+
"should_save": true,
|
141 |
+
"should_training_stop": true
|
142 |
+
},
|
143 |
+
"attributes": {}
|
144 |
+
}
|
145 |
+
},
|
146 |
+
"total_flos": 210845313269760.0,
|
147 |
+
"train_batch_size": 8,
|
148 |
+
"trial_name": null,
|
149 |
+
"trial_params": null
|
150 |
+
}
|
training_eval_loss.png
ADDED
![]() |
training_loss.png
ADDED
![]() |