End of training
Browse files- README.md +2 -4
- adapter_model.safetensors +1 -1
- all_results.json +5 -5
- runs/Apr01_01-06-45_83eaa54fbb08/events.out.tfevents.1711933606.83eaa54fbb08.5375.0 +2 -2
- train_results.json +5 -5
- trainer_state.json +62 -167
README.md
CHANGED
@@ -14,8 +14,6 @@ should probably proofread and complete it, then remove this comment. -->
|
|
14 |
# fine_tuned_llama2_7b
|
15 |
|
16 |
This model is a fine-tuned version of [meta-llama/Llama-2-7b-chat-hf](https://huggingface.co/meta-llama/Llama-2-7b-chat-hf) on an unknown dataset.
|
17 |
-
It achieves the following results on the evaluation set:
|
18 |
-
- Loss: 2.3496
|
19 |
|
20 |
## Model description
|
21 |
|
@@ -38,8 +36,8 @@ The following hyperparameters were used during training:
|
|
38 |
- train_batch_size: 1
|
39 |
- eval_batch_size: 8
|
40 |
- seed: 42
|
41 |
-
- gradient_accumulation_steps:
|
42 |
-
- total_train_batch_size:
|
43 |
- optimizer: Adam with betas=(0.9,0.999) and epsilon=1e-08
|
44 |
- lr_scheduler_type: cosine
|
45 |
- lr_scheduler_warmup_ratio: 0.1
|
|
|
14 |
# fine_tuned_llama2_7b
|
15 |
|
16 |
This model is a fine-tuned version of [meta-llama/Llama-2-7b-chat-hf](https://huggingface.co/meta-llama/Llama-2-7b-chat-hf) on an unknown dataset.
|
|
|
|
|
17 |
|
18 |
## Model description
|
19 |
|
|
|
36 |
- train_batch_size: 1
|
37 |
- eval_batch_size: 8
|
38 |
- seed: 42
|
39 |
+
- gradient_accumulation_steps: 8
|
40 |
+
- total_train_batch_size: 8
|
41 |
- optimizer: Adam with betas=(0.9,0.999) and epsilon=1e-08
|
42 |
- lr_scheduler_type: cosine
|
43 |
- lr_scheduler_warmup_ratio: 0.1
|
adapter_model.safetensors
CHANGED
@@ -1,3 +1,3 @@
|
|
1 |
version https://git-lfs.github.com/spec/v1
|
2 |
-
oid sha256:
|
3 |
size 639691872
|
|
|
1 |
version https://git-lfs.github.com/spec/v1
|
2 |
+
oid sha256:234f19e4eaf40da80182f856e7da52635d1e006d160f1eb4d153da9ba6891f03
|
3 |
size 639691872
|
all_results.json
CHANGED
@@ -1,8 +1,8 @@
|
|
1 |
{
|
2 |
"epoch": 1.0,
|
3 |
-
"total_flos":
|
4 |
-
"train_loss":
|
5 |
-
"train_runtime":
|
6 |
-
"train_samples_per_second": 1.
|
7 |
-
"train_steps_per_second": 0.
|
8 |
}
|
|
|
1 |
{
|
2 |
"epoch": 1.0,
|
3 |
+
"total_flos": 2.5585123528237056e+16,
|
4 |
+
"train_loss": 0.9235503957773034,
|
5 |
+
"train_runtime": 6177.2242,
|
6 |
+
"train_samples_per_second": 1.707,
|
7 |
+
"train_steps_per_second": 0.213
|
8 |
}
|
runs/Apr01_01-06-45_83eaa54fbb08/events.out.tfevents.1711933606.83eaa54fbb08.5375.0
CHANGED
@@ -1,3 +1,3 @@
|
|
1 |
version https://git-lfs.github.com/spec/v1
|
2 |
-
oid sha256:
|
3 |
-
size
|
|
|
1 |
version https://git-lfs.github.com/spec/v1
|
2 |
+
oid sha256:1a24416d7c619131464dad8de9f25e2e293f17a4406e36b610de7766635f3e34
|
3 |
+
size 8152
|
train_results.json
CHANGED
@@ -1,8 +1,8 @@
|
|
1 |
{
|
2 |
"epoch": 1.0,
|
3 |
-
"total_flos":
|
4 |
-
"train_loss":
|
5 |
-
"train_runtime":
|
6 |
-
"train_samples_per_second": 1.
|
7 |
-
"train_steps_per_second": 0.
|
8 |
}
|
|
|
1 |
{
|
2 |
"epoch": 1.0,
|
3 |
+
"total_flos": 2.5585123528237056e+16,
|
4 |
+
"train_loss": 0.9235503957773034,
|
5 |
+
"train_runtime": 6177.2242,
|
6 |
+
"train_samples_per_second": 1.707,
|
7 |
+
"train_steps_per_second": 0.213
|
8 |
}
|
trainer_state.json
CHANGED
@@ -1,225 +1,120 @@
|
|
1 |
{
|
2 |
"best_metric": null,
|
3 |
"best_model_checkpoint": null,
|
4 |
-
"epoch": 0.
|
5 |
"eval_steps": 500,
|
6 |
-
"global_step":
|
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.
|
13 |
-
"grad_norm":
|
14 |
-
"learning_rate":
|
15 |
-
"loss":
|
16 |
"step": 100
|
17 |
},
|
18 |
{
|
19 |
-
"epoch": 0.
|
20 |
-
"grad_norm":
|
21 |
-
"learning_rate": 1.
|
22 |
-
"loss":
|
23 |
"step": 200
|
24 |
},
|
25 |
{
|
26 |
-
"epoch": 0.
|
27 |
-
"grad_norm":
|
28 |
-
"learning_rate": 1.
|
29 |
-
"loss":
|
30 |
"step": 300
|
31 |
},
|
32 |
{
|
33 |
-
"epoch": 0.
|
34 |
-
"grad_norm":
|
35 |
-
"learning_rate": 1.
|
36 |
-
"loss":
|
37 |
"step": 400
|
38 |
},
|
39 |
{
|
40 |
-
"epoch": 0.
|
41 |
-
"grad_norm":
|
42 |
-
"learning_rate": 1.
|
43 |
-
"loss":
|
44 |
"step": 500
|
45 |
},
|
46 |
{
|
47 |
-
"epoch": 0.
|
48 |
-
"grad_norm":
|
49 |
-
"learning_rate": 1.
|
50 |
-
"loss":
|
51 |
"step": 600
|
52 |
},
|
53 |
{
|
54 |
-
"epoch": 0.
|
55 |
-
"grad_norm":
|
56 |
-
"learning_rate": 1.
|
57 |
-
"loss":
|
58 |
"step": 700
|
59 |
},
|
60 |
{
|
61 |
-
"epoch": 0.
|
62 |
-
"grad_norm":
|
63 |
-
"learning_rate":
|
64 |
-
"loss":
|
65 |
"step": 800
|
66 |
},
|
67 |
{
|
68 |
-
"epoch": 0.
|
69 |
-
"grad_norm":
|
70 |
-
"learning_rate":
|
71 |
-
"loss":
|
72 |
"step": 900
|
73 |
},
|
74 |
{
|
75 |
-
"epoch": 0.
|
76 |
-
"grad_norm":
|
77 |
-
"learning_rate":
|
78 |
-
"loss":
|
79 |
"step": 1000
|
80 |
},
|
81 |
{
|
82 |
-
"epoch": 0.
|
83 |
-
"grad_norm":
|
84 |
-
"learning_rate": 1.
|
85 |
-
"loss":
|
86 |
"step": 1100
|
87 |
},
|
88 |
{
|
89 |
-
"epoch": 0.
|
90 |
-
"grad_norm":
|
91 |
-
"learning_rate":
|
92 |
-
"loss":
|
93 |
"step": 1200
|
94 |
},
|
95 |
{
|
96 |
-
"epoch": 0.
|
97 |
-
"grad_norm":
|
98 |
-
"learning_rate": 1.
|
99 |
-
"loss":
|
100 |
"step": 1300
|
101 |
},
|
102 |
-
{
|
103 |
-
"epoch": 0.49,
|
104 |
-
"grad_norm": 10.425432205200195,
|
105 |
-
"learning_rate": 1.2314444308256605e-05,
|
106 |
-
"loss": 2.2865,
|
107 |
-
"step": 1400
|
108 |
-
},
|
109 |
-
{
|
110 |
-
"epoch": 0.52,
|
111 |
-
"grad_norm": 16.403301239013672,
|
112 |
-
"learning_rate": 1.1123080572287608e-05,
|
113 |
-
"loss": 2.2595,
|
114 |
-
"step": 1500
|
115 |
-
},
|
116 |
-
{
|
117 |
-
"epoch": 0.55,
|
118 |
-
"grad_norm": 11.935959815979004,
|
119 |
-
"learning_rate": 9.915289346843219e-06,
|
120 |
-
"loss": 2.3662,
|
121 |
-
"step": 1600
|
122 |
-
},
|
123 |
-
{
|
124 |
-
"epoch": 0.59,
|
125 |
-
"grad_norm": 18.410987854003906,
|
126 |
-
"learning_rate": 8.708737198449509e-06,
|
127 |
-
"loss": 2.2021,
|
128 |
-
"step": 1700
|
129 |
-
},
|
130 |
-
{
|
131 |
-
"epoch": 0.62,
|
132 |
-
"grad_norm": 15.293601036071777,
|
133 |
-
"learning_rate": 7.521072569442963e-06,
|
134 |
-
"loss": 2.2545,
|
135 |
-
"step": 1800
|
136 |
-
},
|
137 |
-
{
|
138 |
-
"epoch": 0.66,
|
139 |
-
"grad_norm": 16.34610939025879,
|
140 |
-
"learning_rate": 6.369667631219584e-06,
|
141 |
-
"loss": 2.3199,
|
142 |
-
"step": 1900
|
143 |
-
},
|
144 |
-
{
|
145 |
-
"epoch": 0.69,
|
146 |
-
"grad_norm": 15.948208808898926,
|
147 |
-
"learning_rate": 5.2713641785457504e-06,
|
148 |
-
"loss": 2.2029,
|
149 |
-
"step": 2000
|
150 |
-
},
|
151 |
-
{
|
152 |
-
"epoch": 0.73,
|
153 |
-
"grad_norm": 27.17706298828125,
|
154 |
-
"learning_rate": 4.242227281777747e-06,
|
155 |
-
"loss": 2.2861,
|
156 |
-
"step": 2100
|
157 |
-
},
|
158 |
-
{
|
159 |
-
"epoch": 0.76,
|
160 |
-
"grad_norm": 19.407489776611328,
|
161 |
-
"learning_rate": 3.297310300360622e-06,
|
162 |
-
"loss": 2.2157,
|
163 |
-
"step": 2200
|
164 |
-
},
|
165 |
-
{
|
166 |
-
"epoch": 0.8,
|
167 |
-
"grad_norm": 11.622710227966309,
|
168 |
-
"learning_rate": 2.450434694793621e-06,
|
169 |
-
"loss": 2.2724,
|
170 |
-
"step": 2300
|
171 |
-
},
|
172 |
-
{
|
173 |
-
"epoch": 0.83,
|
174 |
-
"grad_norm": 16.701732635498047,
|
175 |
-
"learning_rate": 1.7139878577898772e-06,
|
176 |
-
"loss": 2.1622,
|
177 |
-
"step": 2400
|
178 |
-
},
|
179 |
-
{
|
180 |
-
"epoch": 0.87,
|
181 |
-
"grad_norm": 10.720149040222168,
|
182 |
-
"learning_rate": 1.0987419217881333e-06,
|
183 |
-
"loss": 2.2026,
|
184 |
-
"step": 2500
|
185 |
-
},
|
186 |
-
{
|
187 |
-
"epoch": 0.9,
|
188 |
-
"grad_norm": 14.398381233215332,
|
189 |
-
"learning_rate": 6.136961931496943e-07,
|
190 |
-
"loss": 2.2619,
|
191 |
-
"step": 2600
|
192 |
-
},
|
193 |
-
{
|
194 |
-
"epoch": 0.94,
|
195 |
-
"grad_norm": 16.95086669921875,
|
196 |
-
"learning_rate": 2.6594551778223896e-07,
|
197 |
-
"loss": 2.2626,
|
198 |
-
"step": 2700
|
199 |
-
},
|
200 |
-
{
|
201 |
-
"epoch": 0.97,
|
202 |
-
"grad_norm": 12.132495880126953,
|
203 |
-
"learning_rate": 6.057650362879753e-08,
|
204 |
-
"loss": 2.1139,
|
205 |
-
"step": 2800
|
206 |
-
},
|
207 |
{
|
208 |
"epoch": 1.0,
|
209 |
-
"step":
|
210 |
-
"total_flos":
|
211 |
-
"train_loss":
|
212 |
-
"train_runtime":
|
213 |
-
"train_samples_per_second": 1.
|
214 |
-
"train_steps_per_second": 0.
|
215 |
}
|
216 |
],
|
217 |
"logging_steps": 100,
|
218 |
-
"max_steps":
|
219 |
"num_input_tokens_seen": 0,
|
220 |
"num_train_epochs": 1,
|
221 |
"save_steps": 500,
|
222 |
-
"total_flos":
|
223 |
"train_batch_size": 1,
|
224 |
"trial_name": null,
|
225 |
"trial_params": null
|
|
|
1 |
{
|
2 |
"best_metric": null,
|
3 |
"best_model_checkpoint": null,
|
4 |
+
"epoch": 0.9993360523570142,
|
5 |
"eval_steps": 500,
|
6 |
+
"global_step": 1317,
|
7 |
"is_hyper_param_search": false,
|
8 |
"is_local_process_zero": true,
|
9 |
"is_world_process_zero": true,
|
10 |
"log_history": [
|
11 |
{
|
12 |
+
"epoch": 0.08,
|
13 |
+
"grad_norm": 2.6666550636291504,
|
14 |
+
"learning_rate": 1.5151515151515153e-05,
|
15 |
+
"loss": 1.75,
|
16 |
"step": 100
|
17 |
},
|
18 |
{
|
19 |
+
"epoch": 0.15,
|
20 |
+
"grad_norm": 1.9126778841018677,
|
21 |
+
"learning_rate": 1.983794055463009e-05,
|
22 |
+
"loss": 0.9131,
|
23 |
"step": 200
|
24 |
},
|
25 |
{
|
26 |
+
"epoch": 0.23,
|
27 |
+
"grad_norm": 2.010279655456543,
|
28 |
+
"learning_rate": 1.90244256701717e-05,
|
29 |
+
"loss": 0.921,
|
30 |
"step": 300
|
31 |
},
|
32 |
{
|
33 |
+
"epoch": 0.3,
|
34 |
+
"grad_norm": 1.661656141281128,
|
35 |
+
"learning_rate": 1.7580334804873595e-05,
|
36 |
+
"loss": 0.8664,
|
37 |
"step": 400
|
38 |
},
|
39 |
{
|
40 |
+
"epoch": 0.38,
|
41 |
+
"grad_norm": 1.8409279584884644,
|
42 |
+
"learning_rate": 1.5606572885773613e-05,
|
43 |
+
"loss": 0.8562,
|
44 |
"step": 500
|
45 |
},
|
46 |
{
|
47 |
+
"epoch": 0.46,
|
48 |
+
"grad_norm": 1.8635607957839966,
|
49 |
+
"learning_rate": 1.324105526655396e-05,
|
50 |
+
"loss": 0.8561,
|
51 |
"step": 600
|
52 |
},
|
53 |
{
|
54 |
+
"epoch": 0.53,
|
55 |
+
"grad_norm": 1.4191073179244995,
|
56 |
+
"learning_rate": 1.0649070980273363e-05,
|
57 |
+
"loss": 0.8621,
|
58 |
"step": 700
|
59 |
},
|
60 |
{
|
61 |
+
"epoch": 0.61,
|
62 |
+
"grad_norm": 1.4062063694000244,
|
63 |
+
"learning_rate": 8.011733273733208e-06,
|
64 |
+
"loss": 0.8523,
|
65 |
"step": 800
|
66 |
},
|
67 |
{
|
68 |
+
"epoch": 0.68,
|
69 |
+
"grad_norm": 1.7386995553970337,
|
70 |
+
"learning_rate": 5.51332443501349e-06,
|
71 |
+
"loss": 0.8277,
|
72 |
"step": 900
|
73 |
},
|
74 |
{
|
75 |
+
"epoch": 0.76,
|
76 |
+
"grad_norm": 1.6414889097213745,
|
77 |
+
"learning_rate": 3.3284191862731585e-06,
|
78 |
+
"loss": 0.8406,
|
79 |
"step": 1000
|
80 |
},
|
81 |
{
|
82 |
+
"epoch": 0.83,
|
83 |
+
"grad_norm": 1.6184569597244263,
|
84 |
+
"learning_rate": 1.6096863865200606e-06,
|
85 |
+
"loss": 0.8189,
|
86 |
"step": 1100
|
87 |
},
|
88 |
{
|
89 |
+
"epoch": 0.91,
|
90 |
+
"grad_norm": 1.6727625131607056,
|
91 |
+
"learning_rate": 4.772213925798331e-07,
|
92 |
+
"loss": 0.837,
|
93 |
"step": 1200
|
94 |
},
|
95 |
{
|
96 |
+
"epoch": 0.99,
|
97 |
+
"grad_norm": 1.7798255681991577,
|
98 |
+
"learning_rate": 1.0154472728808318e-08,
|
99 |
+
"loss": 0.8195,
|
100 |
"step": 1300
|
101 |
},
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
102 |
{
|
103 |
"epoch": 1.0,
|
104 |
+
"step": 1317,
|
105 |
+
"total_flos": 2.5585123528237056e+16,
|
106 |
+
"train_loss": 0.9235503957773034,
|
107 |
+
"train_runtime": 6177.2242,
|
108 |
+
"train_samples_per_second": 1.707,
|
109 |
+
"train_steps_per_second": 0.213
|
110 |
}
|
111 |
],
|
112 |
"logging_steps": 100,
|
113 |
+
"max_steps": 1317,
|
114 |
"num_input_tokens_seen": 0,
|
115 |
"num_train_epochs": 1,
|
116 |
"save_steps": 500,
|
117 |
+
"total_flos": 2.5585123528237056e+16,
|
118 |
"train_batch_size": 1,
|
119 |
"trial_name": null,
|
120 |
"trial_params": null
|