Commit
•
c8bebbb
1
Parent(s):
12c6498
Model save
Browse files- README.md +6 -12
- adapter_config.json +3 -3
- adapter_model.safetensors +1 -1
- all_results.json +8 -8
- eval_results.json +5 -5
- runs/Dec14_19-31-24_s4311/events.out.tfevents.1702582363.s4311.897573.0 +3 -0
- runs/Dec14_19-31-24_s4311/events.out.tfevents.1702584022.s4311.897573.1 +3 -0
- train_results.json +4 -4
- trainer_state.json +28 -88
- training_args.bin +1 -1
README.md
CHANGED
@@ -15,7 +15,7 @@ should probably proofread and complete it, then remove this comment. -->
|
|
15 |
|
16 |
This model is a fine-tuned version of [mistralai/Mistral-7B-v0.1](https://huggingface.co/mistralai/Mistral-7B-v0.1) on an unknown dataset.
|
17 |
It achieves the following results on the evaluation set:
|
18 |
-
- Loss: 1.
|
19 |
|
20 |
## Model description
|
21 |
|
@@ -43,22 +43,16 @@ The following hyperparameters were used during training:
|
|
43 |
- total_train_batch_size: 512
|
44 |
- optimizer: Adam with betas=(0.9,0.999) and epsilon=1e-08
|
45 |
- lr_scheduler_type: cosine
|
46 |
-
- num_epochs:
|
47 |
|
48 |
### Training results
|
49 |
|
50 |
| Training Loss | Epoch | Step | Validation Loss |
|
51 |
|:-------------:|:-----:|:----:|:---------------:|
|
52 |
-
| 1.7095 | 0.09 | 1 | 1.
|
53 |
-
| 1.7095 | 1.07 | 2 | 1.
|
54 |
-
| 1.7095 | 2.06 | 3 | 1.
|
55 |
-
| 1.7095 | 3.04 | 4 | 1.
|
56 |
-
| 1.6715 | 4.02 | 5 | 1.6106 |
|
57 |
-
| 1.6715 | 5.1 | 7 | 1.5584 |
|
58 |
-
| 1.6715 | 6.08 | 8 | 1.5344 |
|
59 |
-
| 1.6715 | 7.06 | 9 | 1.5132 |
|
60 |
-
| 1.5588 | 8.05 | 10 | 1.4933 |
|
61 |
-
| 1.5588 | 9.03 | 11 | 1.4756 |
|
62 |
|
63 |
|
64 |
### Framework versions
|
|
|
15 |
|
16 |
This model is a fine-tuned version of [mistralai/Mistral-7B-v0.1](https://huggingface.co/mistralai/Mistral-7B-v0.1) on an unknown dataset.
|
17 |
It achieves the following results on the evaluation set:
|
18 |
+
- Loss: 1.6381
|
19 |
|
20 |
## Model description
|
21 |
|
|
|
43 |
- total_train_batch_size: 512
|
44 |
- optimizer: Adam with betas=(0.9,0.999) and epsilon=1e-08
|
45 |
- lr_scheduler_type: cosine
|
46 |
+
- num_epochs: 4
|
47 |
|
48 |
### Training results
|
49 |
|
50 |
| Training Loss | Epoch | Step | Validation Loss |
|
51 |
|:-------------:|:-----:|:----:|:---------------:|
|
52 |
+
| 1.7095 | 0.09 | 1 | 1.7014 |
|
53 |
+
| 1.7095 | 1.07 | 2 | 1.6855 |
|
54 |
+
| 1.7095 | 2.06 | 3 | 1.6637 |
|
55 |
+
| 1.7095 | 3.04 | 4 | 1.6382 |
|
|
|
|
|
|
|
|
|
|
|
|
|
56 |
|
57 |
|
58 |
### Framework versions
|
adapter_config.json
CHANGED
@@ -16,10 +16,10 @@
|
|
16 |
"rank_pattern": {},
|
17 |
"revision": null,
|
18 |
"target_modules": [
|
19 |
-
"q_proj",
|
20 |
-
"o_proj",
|
21 |
"v_proj",
|
22 |
-
"
|
|
|
|
|
23 |
],
|
24 |
"task_type": "CAUSAL_LM"
|
25 |
}
|
|
|
16 |
"rank_pattern": {},
|
17 |
"revision": null,
|
18 |
"target_modules": [
|
|
|
|
|
19 |
"v_proj",
|
20 |
+
"q_proj",
|
21 |
+
"k_proj",
|
22 |
+
"o_proj"
|
23 |
],
|
24 |
"task_type": "CAUSAL_LM"
|
25 |
}
|
adapter_model.safetensors
CHANGED
@@ -1,3 +1,3 @@
|
|
1 |
version https://git-lfs.github.com/spec/v1
|
2 |
-
oid sha256:
|
3 |
size 218138576
|
|
|
1 |
version https://git-lfs.github.com/spec/v1
|
2 |
+
oid sha256:0cbcd1726281528770049545897e4929fdbcd2e3c520f2cd8517683dc9413b7b
|
3 |
size 218138576
|
all_results.json
CHANGED
@@ -1,13 +1,13 @@
|
|
1 |
{
|
2 |
-
"epoch":
|
3 |
-
"eval_loss": 1.
|
4 |
-
"eval_runtime": 6.
|
5 |
"eval_samples": 294,
|
6 |
-
"eval_samples_per_second": 46.
|
7 |
-
"eval_steps_per_second": 5.
|
8 |
-
"train_loss": 1.
|
9 |
-
"train_runtime":
|
10 |
"train_samples": 5895,
|
11 |
-
"train_samples_per_second": 14.
|
12 |
"train_steps_per_second": 0.027
|
13 |
}
|
|
|
1 |
{
|
2 |
+
"epoch": 3.04,
|
3 |
+
"eval_loss": 1.6381348371505737,
|
4 |
+
"eval_runtime": 6.3457,
|
5 |
"eval_samples": 294,
|
6 |
+
"eval_samples_per_second": 46.331,
|
7 |
+
"eval_steps_per_second": 5.831,
|
8 |
+
"train_loss": 1.9840126037597656,
|
9 |
+
"train_runtime": 1653.0192,
|
10 |
"train_samples": 5895,
|
11 |
+
"train_samples_per_second": 14.265,
|
12 |
"train_steps_per_second": 0.027
|
13 |
}
|
eval_results.json
CHANGED
@@ -1,8 +1,8 @@
|
|
1 |
{
|
2 |
-
"epoch":
|
3 |
-
"eval_loss": 1.
|
4 |
-
"eval_runtime": 6.
|
5 |
"eval_samples": 294,
|
6 |
-
"eval_samples_per_second": 46.
|
7 |
-
"eval_steps_per_second": 5.
|
8 |
}
|
|
|
1 |
{
|
2 |
+
"epoch": 3.04,
|
3 |
+
"eval_loss": 1.6381348371505737,
|
4 |
+
"eval_runtime": 6.3457,
|
5 |
"eval_samples": 294,
|
6 |
+
"eval_samples_per_second": 46.331,
|
7 |
+
"eval_steps_per_second": 5.831
|
8 |
}
|
runs/Dec14_19-31-24_s4311/events.out.tfevents.1702582363.s4311.897573.0
ADDED
@@ -0,0 +1,3 @@
|
|
|
|
|
|
|
|
|
1 |
+
version https://git-lfs.github.com/spec/v1
|
2 |
+
oid sha256:b8e986c304032fa4314bd17cd1ec247ab0c55cf74221235993be6aa019bbb297
|
3 |
+
size 5819
|
runs/Dec14_19-31-24_s4311/events.out.tfevents.1702584022.s4311.897573.1
ADDED
@@ -0,0 +1,3 @@
|
|
|
|
|
|
|
|
|
1 |
+
version https://git-lfs.github.com/spec/v1
|
2 |
+
oid sha256:898d7be29463605ee33deb4c69f44f454a32dc33c9c1920f1dc705b0424861e3
|
3 |
+
size 354
|
train_results.json
CHANGED
@@ -1,8 +1,8 @@
|
|
1 |
{
|
2 |
-
"epoch":
|
3 |
-
"train_loss": 1.
|
4 |
-
"train_runtime":
|
5 |
"train_samples": 5895,
|
6 |
-
"train_samples_per_second": 14.
|
7 |
"train_steps_per_second": 0.027
|
8 |
}
|
|
|
1 |
{
|
2 |
+
"epoch": 3.04,
|
3 |
+
"train_loss": 1.9840126037597656,
|
4 |
+
"train_runtime": 1653.0192,
|
5 |
"train_samples": 5895,
|
6 |
+
"train_samples_per_second": 14.265,
|
7 |
"train_steps_per_second": 0.027
|
8 |
}
|
trainer_state.json
CHANGED
@@ -1,126 +1,66 @@
|
|
1 |
{
|
2 |
"best_metric": null,
|
3 |
"best_model_checkpoint": null,
|
4 |
-
"epoch":
|
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": 0.09,
|
13 |
-
"learning_rate": 1.
|
14 |
"loss": 1.7095,
|
15 |
"step": 1
|
16 |
},
|
17 |
{
|
18 |
"epoch": 0.09,
|
19 |
-
"eval_loss": 1.
|
20 |
-
"eval_runtime": 6.
|
21 |
-
"eval_samples_per_second": 46.
|
22 |
-
"eval_steps_per_second": 5.
|
23 |
"step": 1
|
24 |
},
|
25 |
{
|
26 |
"epoch": 1.07,
|
27 |
-
"eval_loss": 1.
|
28 |
-
"eval_runtime": 6.
|
29 |
-
"eval_samples_per_second": 46.
|
30 |
-
"eval_steps_per_second": 5.
|
31 |
"step": 2
|
32 |
},
|
33 |
{
|
34 |
"epoch": 2.06,
|
35 |
-
"eval_loss": 1.
|
36 |
-
"eval_runtime": 6.
|
37 |
-
"eval_samples_per_second": 46.
|
38 |
-
"eval_steps_per_second": 5.
|
39 |
"step": 3
|
40 |
},
|
41 |
{
|
42 |
"epoch": 3.04,
|
43 |
-
"eval_loss": 1.
|
44 |
-
"eval_runtime": 6.
|
45 |
-
"eval_samples_per_second": 46.
|
46 |
-
"eval_steps_per_second": 5.
|
47 |
"step": 4
|
48 |
},
|
49 |
{
|
50 |
-
"epoch":
|
51 |
-
"
|
52 |
-
"
|
53 |
-
"
|
54 |
-
|
55 |
-
|
56 |
-
"epoch": 4.02,
|
57 |
-
"eval_loss": 1.6106265783309937,
|
58 |
-
"eval_runtime": 6.3433,
|
59 |
-
"eval_samples_per_second": 46.348,
|
60 |
-
"eval_steps_per_second": 5.833,
|
61 |
-
"step": 5
|
62 |
-
},
|
63 |
-
{
|
64 |
-
"epoch": 5.1,
|
65 |
-
"eval_loss": 1.5584189891815186,
|
66 |
-
"eval_runtime": 6.3627,
|
67 |
-
"eval_samples_per_second": 46.206,
|
68 |
-
"eval_steps_per_second": 5.815,
|
69 |
-
"step": 7
|
70 |
-
},
|
71 |
-
{
|
72 |
-
"epoch": 6.08,
|
73 |
-
"eval_loss": 1.5343831777572632,
|
74 |
-
"eval_runtime": 6.3403,
|
75 |
-
"eval_samples_per_second": 46.37,
|
76 |
-
"eval_steps_per_second": 5.836,
|
77 |
-
"step": 8
|
78 |
-
},
|
79 |
-
{
|
80 |
-
"epoch": 7.06,
|
81 |
-
"eval_loss": 1.5131820440292358,
|
82 |
-
"eval_runtime": 6.3297,
|
83 |
-
"eval_samples_per_second": 46.448,
|
84 |
-
"eval_steps_per_second": 5.845,
|
85 |
-
"step": 9
|
86 |
-
},
|
87 |
-
{
|
88 |
-
"epoch": 8.05,
|
89 |
-
"learning_rate": 1.9594929736144978e-05,
|
90 |
-
"loss": 1.5588,
|
91 |
-
"step": 10
|
92 |
-
},
|
93 |
-
{
|
94 |
-
"epoch": 8.05,
|
95 |
-
"eval_loss": 1.4932821989059448,
|
96 |
-
"eval_runtime": 6.3196,
|
97 |
-
"eval_samples_per_second": 46.522,
|
98 |
-
"eval_steps_per_second": 5.855,
|
99 |
-
"step": 10
|
100 |
-
},
|
101 |
-
{
|
102 |
-
"epoch": 9.03,
|
103 |
-
"eval_loss": 1.4755805730819702,
|
104 |
-
"eval_runtime": 6.3502,
|
105 |
-
"eval_samples_per_second": 46.298,
|
106 |
-
"eval_steps_per_second": 5.827,
|
107 |
-
"step": 11
|
108 |
-
},
|
109 |
-
{
|
110 |
-
"epoch": 9.03,
|
111 |
-
"step": 11,
|
112 |
-
"total_flos": 5.3091621472200294e+17,
|
113 |
-
"train_loss": 1.7142409628087825,
|
114 |
-
"train_runtime": 4135.8024,
|
115 |
-
"train_samples_per_second": 14.254,
|
116 |
"train_steps_per_second": 0.027
|
117 |
}
|
118 |
],
|
119 |
"logging_steps": 5,
|
120 |
-
"max_steps":
|
121 |
-
"num_train_epochs":
|
122 |
"save_steps": 500,
|
123 |
-
"total_flos":
|
124 |
"trial_name": null,
|
125 |
"trial_params": null
|
126 |
}
|
|
|
1 |
{
|
2 |
"best_metric": null,
|
3 |
"best_model_checkpoint": null,
|
4 |
+
"epoch": 3.0400271370420624,
|
5 |
"eval_steps": 500,
|
6 |
+
"global_step": 4,
|
7 |
"is_hyper_param_search": false,
|
8 |
"is_local_process_zero": true,
|
9 |
"is_world_process_zero": true,
|
10 |
"log_history": [
|
11 |
{
|
12 |
"epoch": 0.09,
|
13 |
+
"learning_rate": 1.9974521146102535e-05,
|
14 |
"loss": 1.7095,
|
15 |
"step": 1
|
16 |
},
|
17 |
{
|
18 |
"epoch": 0.09,
|
19 |
+
"eval_loss": 1.7013664245605469,
|
20 |
+
"eval_runtime": 6.3791,
|
21 |
+
"eval_samples_per_second": 46.088,
|
22 |
+
"eval_steps_per_second": 5.8,
|
23 |
"step": 1
|
24 |
},
|
25 |
{
|
26 |
"epoch": 1.07,
|
27 |
+
"eval_loss": 1.6854876279830933,
|
28 |
+
"eval_runtime": 6.3439,
|
29 |
+
"eval_samples_per_second": 46.344,
|
30 |
+
"eval_steps_per_second": 5.832,
|
31 |
"step": 2
|
32 |
},
|
33 |
{
|
34 |
"epoch": 2.06,
|
35 |
+
"eval_loss": 1.663727045059204,
|
36 |
+
"eval_runtime": 6.3388,
|
37 |
+
"eval_samples_per_second": 46.381,
|
38 |
+
"eval_steps_per_second": 5.837,
|
39 |
"step": 3
|
40 |
},
|
41 |
{
|
42 |
"epoch": 3.04,
|
43 |
+
"eval_loss": 1.6381714344024658,
|
44 |
+
"eval_runtime": 6.3614,
|
45 |
+
"eval_samples_per_second": 46.216,
|
46 |
+
"eval_steps_per_second": 5.816,
|
47 |
"step": 4
|
48 |
},
|
49 |
{
|
50 |
+
"epoch": 3.04,
|
51 |
+
"step": 4,
|
52 |
+
"total_flos": 2.123664876067881e+17,
|
53 |
+
"train_loss": 1.9840126037597656,
|
54 |
+
"train_runtime": 1653.0192,
|
55 |
+
"train_samples_per_second": 14.265,
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
56 |
"train_steps_per_second": 0.027
|
57 |
}
|
58 |
],
|
59 |
"logging_steps": 5,
|
60 |
+
"max_steps": 44,
|
61 |
+
"num_train_epochs": 4,
|
62 |
"save_steps": 500,
|
63 |
+
"total_flos": 2.123664876067881e+17,
|
64 |
"trial_name": null,
|
65 |
"trial_params": null
|
66 |
}
|
training_args.bin
CHANGED
@@ -1,3 +1,3 @@
|
|
1 |
version https://git-lfs.github.com/spec/v1
|
2 |
-
oid sha256:
|
3 |
size 4664
|
|
|
1 |
version https://git-lfs.github.com/spec/v1
|
2 |
+
oid sha256:2b24e1f15eab76911260c95314f76d27bf8972d0e9bf4fdbcf3205fc1db47f9a
|
3 |
size 4664
|