Holmeister
commited on
Commit
•
cfefc93
1
Parent(s):
d628738
End of training
Browse files- README.md +3 -2
- all_results.json +12 -0
- eval_results.json +7 -0
- train_results.json +8 -0
- trainer_state.json +171 -0
README.md
CHANGED
@@ -2,6 +2,7 @@
|
|
2 |
library_name: peft
|
3 |
tags:
|
4 |
- llama-factory
|
|
|
5 |
- generated_from_trainer
|
6 |
base_model: mistralai/Mistral-7B-v0.3
|
7 |
model-index:
|
@@ -14,9 +15,9 @@ should probably proofread and complete it, then remove this comment. -->
|
|
14 |
|
15 |
# Mistral_AAID_new_mixed_train_final_2
|
16 |
|
17 |
-
This model is a fine-tuned version of [mistralai/Mistral-7B-v0.3](https://huggingface.co/mistralai/Mistral-7B-v0.3) on
|
18 |
It achieves the following results on the evaluation set:
|
19 |
-
- Loss: 0.
|
20 |
|
21 |
## Model description
|
22 |
|
|
|
2 |
library_name: peft
|
3 |
tags:
|
4 |
- llama-factory
|
5 |
+
- lora
|
6 |
- generated_from_trainer
|
7 |
base_model: mistralai/Mistral-7B-v0.3
|
8 |
model-index:
|
|
|
15 |
|
16 |
# Mistral_AAID_new_mixed_train_final_2
|
17 |
|
18 |
+
This model is a fine-tuned version of [mistralai/Mistral-7B-v0.3](https://huggingface.co/mistralai/Mistral-7B-v0.3) on the AAID_new_mixed dataset.
|
19 |
It achieves the following results on the evaluation set:
|
20 |
+
- Loss: 0.5347
|
21 |
|
22 |
## Model description
|
23 |
|
all_results.json
ADDED
@@ -0,0 +1,12 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
1 |
+
{
|
2 |
+
"epoch": 0.3500136724090785,
|
3 |
+
"eval_loss": 0.5347398519515991,
|
4 |
+
"eval_runtime": 232.3061,
|
5 |
+
"eval_samples_per_second": 20.284,
|
6 |
+
"eval_steps_per_second": 0.637,
|
7 |
+
"total_flos": 6.133543194454917e+17,
|
8 |
+
"train_loss": 0.37804515957832335,
|
9 |
+
"train_runtime": 17123.6385,
|
10 |
+
"train_samples_per_second": 41.003,
|
11 |
+
"train_steps_per_second": 0.16
|
12 |
+
}
|
eval_results.json
ADDED
@@ -0,0 +1,7 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
1 |
+
{
|
2 |
+
"epoch": 0.3500136724090785,
|
3 |
+
"eval_loss": 0.5347398519515991,
|
4 |
+
"eval_runtime": 232.3061,
|
5 |
+
"eval_samples_per_second": 20.284,
|
6 |
+
"eval_steps_per_second": 0.637
|
7 |
+
}
|
train_results.json
ADDED
@@ -0,0 +1,8 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
1 |
+
{
|
2 |
+
"epoch": 0.3500136724090785,
|
3 |
+
"total_flos": 6.133543194454917e+17,
|
4 |
+
"train_loss": 0.37804515957832335,
|
5 |
+
"train_runtime": 17123.6385,
|
6 |
+
"train_samples_per_second": 41.003,
|
7 |
+
"train_steps_per_second": 0.16
|
8 |
+
}
|
trainer_state.json
ADDED
@@ -0,0 +1,171 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
1 |
+
{
|
2 |
+
"best_metric": 0.5347398519515991,
|
3 |
+
"best_model_checkpoint": "saves/Mistral-7B/lora/train_1/checkpoint-120",
|
4 |
+
"epoch": 0.3500136724090785,
|
5 |
+
"eval_steps": 40,
|
6 |
+
"global_step": 320,
|
7 |
+
"is_hyper_param_search": false,
|
8 |
+
"is_local_process_zero": true,
|
9 |
+
"is_world_process_zero": true,
|
10 |
+
"log_history": [
|
11 |
+
{
|
12 |
+
"epoch": 0.04375170905113481,
|
13 |
+
"grad_norm": 0.8040962219238281,
|
14 |
+
"learning_rate": 0.00014457831325301204,
|
15 |
+
"loss": 0.9535,
|
16 |
+
"step": 40
|
17 |
+
},
|
18 |
+
{
|
19 |
+
"epoch": 0.04375170905113481,
|
20 |
+
"eval_loss": 0.5372358560562134,
|
21 |
+
"eval_runtime": 232.3992,
|
22 |
+
"eval_samples_per_second": 20.275,
|
23 |
+
"eval_steps_per_second": 0.637,
|
24 |
+
"step": 40
|
25 |
+
},
|
26 |
+
{
|
27 |
+
"epoch": 0.08750341810226962,
|
28 |
+
"grad_norm": 0.5753062963485718,
|
29 |
+
"learning_rate": 0.0002891566265060241,
|
30 |
+
"loss": 0.3461,
|
31 |
+
"step": 80
|
32 |
+
},
|
33 |
+
{
|
34 |
+
"epoch": 0.08750341810226962,
|
35 |
+
"eval_loss": 0.5532671213150024,
|
36 |
+
"eval_runtime": 232.5653,
|
37 |
+
"eval_samples_per_second": 20.261,
|
38 |
+
"eval_steps_per_second": 0.636,
|
39 |
+
"step": 80
|
40 |
+
},
|
41 |
+
{
|
42 |
+
"epoch": 0.13125512715340443,
|
43 |
+
"grad_norm": 0.6130059957504272,
|
44 |
+
"learning_rate": 0.00029985669589905196,
|
45 |
+
"loss": 0.3181,
|
46 |
+
"step": 120
|
47 |
+
},
|
48 |
+
{
|
49 |
+
"epoch": 0.13125512715340443,
|
50 |
+
"eval_loss": 0.5347398519515991,
|
51 |
+
"eval_runtime": 232.5299,
|
52 |
+
"eval_samples_per_second": 20.264,
|
53 |
+
"eval_steps_per_second": 0.636,
|
54 |
+
"step": 120
|
55 |
+
},
|
56 |
+
{
|
57 |
+
"epoch": 0.17500683620453925,
|
58 |
+
"grad_norm": 0.4800422787666321,
|
59 |
+
"learning_rate": 0.0002993796936474492,
|
60 |
+
"loss": 0.3057,
|
61 |
+
"step": 160
|
62 |
+
},
|
63 |
+
{
|
64 |
+
"epoch": 0.17500683620453925,
|
65 |
+
"eval_loss": 0.5564690828323364,
|
66 |
+
"eval_runtime": 232.384,
|
67 |
+
"eval_samples_per_second": 20.277,
|
68 |
+
"eval_steps_per_second": 0.637,
|
69 |
+
"step": 160
|
70 |
+
},
|
71 |
+
{
|
72 |
+
"epoch": 0.21875854525567404,
|
73 |
+
"grad_norm": 0.38192033767700195,
|
74 |
+
"learning_rate": 0.00029856911617379416,
|
75 |
+
"loss": 0.297,
|
76 |
+
"step": 200
|
77 |
+
},
|
78 |
+
{
|
79 |
+
"epoch": 0.21875854525567404,
|
80 |
+
"eval_loss": 0.5358652472496033,
|
81 |
+
"eval_runtime": 232.6578,
|
82 |
+
"eval_samples_per_second": 20.253,
|
83 |
+
"eval_steps_per_second": 0.636,
|
84 |
+
"step": 200
|
85 |
+
},
|
86 |
+
{
|
87 |
+
"epoch": 0.26251025430680885,
|
88 |
+
"grad_norm": 0.42881786823272705,
|
89 |
+
"learning_rate": 0.00029742677355383513,
|
90 |
+
"loss": 0.277,
|
91 |
+
"step": 240
|
92 |
+
},
|
93 |
+
{
|
94 |
+
"epoch": 0.26251025430680885,
|
95 |
+
"eval_loss": 0.5426477789878845,
|
96 |
+
"eval_runtime": 232.3465,
|
97 |
+
"eval_samples_per_second": 20.28,
|
98 |
+
"eval_steps_per_second": 0.637,
|
99 |
+
"step": 240
|
100 |
+
},
|
101 |
+
{
|
102 |
+
"epoch": 0.30626196335794365,
|
103 |
+
"grad_norm": 0.4693865180015564,
|
104 |
+
"learning_rate": 0.0002959552167179149,
|
105 |
+
"loss": 0.2707,
|
106 |
+
"step": 280
|
107 |
+
},
|
108 |
+
{
|
109 |
+
"epoch": 0.30626196335794365,
|
110 |
+
"eval_loss": 0.5676203370094299,
|
111 |
+
"eval_runtime": 232.3212,
|
112 |
+
"eval_samples_per_second": 20.282,
|
113 |
+
"eval_steps_per_second": 0.637,
|
114 |
+
"step": 280
|
115 |
+
},
|
116 |
+
{
|
117 |
+
"epoch": 0.3500136724090785,
|
118 |
+
"grad_norm": 0.5723518133163452,
|
119 |
+
"learning_rate": 0.00029415773175456614,
|
120 |
+
"loss": 0.2563,
|
121 |
+
"step": 320
|
122 |
+
},
|
123 |
+
{
|
124 |
+
"epoch": 0.3500136724090785,
|
125 |
+
"eval_loss": 0.566847562789917,
|
126 |
+
"eval_runtime": 232.4745,
|
127 |
+
"eval_samples_per_second": 20.269,
|
128 |
+
"eval_steps_per_second": 0.637,
|
129 |
+
"step": 320
|
130 |
+
},
|
131 |
+
{
|
132 |
+
"epoch": 0.3500136724090785,
|
133 |
+
"step": 320,
|
134 |
+
"total_flos": 6.133543194454917e+17,
|
135 |
+
"train_loss": 0.37804515957832335,
|
136 |
+
"train_runtime": 17123.6385,
|
137 |
+
"train_samples_per_second": 41.003,
|
138 |
+
"train_steps_per_second": 0.16
|
139 |
+
}
|
140 |
+
],
|
141 |
+
"logging_steps": 40,
|
142 |
+
"max_steps": 2742,
|
143 |
+
"num_input_tokens_seen": 0,
|
144 |
+
"num_train_epochs": 3,
|
145 |
+
"save_steps": 40,
|
146 |
+
"stateful_callbacks": {
|
147 |
+
"EarlyStoppingCallback": {
|
148 |
+
"args": {
|
149 |
+
"early_stopping_patience": 5,
|
150 |
+
"early_stopping_threshold": 0.0
|
151 |
+
},
|
152 |
+
"attributes": {
|
153 |
+
"early_stopping_patience_counter": 0
|
154 |
+
}
|
155 |
+
},
|
156 |
+
"TrainerControl": {
|
157 |
+
"args": {
|
158 |
+
"should_epoch_stop": false,
|
159 |
+
"should_evaluate": false,
|
160 |
+
"should_log": false,
|
161 |
+
"should_save": true,
|
162 |
+
"should_training_stop": true
|
163 |
+
},
|
164 |
+
"attributes": {}
|
165 |
+
}
|
166 |
+
},
|
167 |
+
"total_flos": 6.133543194454917e+17,
|
168 |
+
"train_batch_size": 32,
|
169 |
+
"trial_name": null,
|
170 |
+
"trial_params": null
|
171 |
+
}
|