End of training
Browse files- README.md +1 -55
- adapter_config.json +4 -4
- adapter_model.safetensors +1 -1
- training_args.bin +1 -1
README.md
CHANGED
@@ -17,8 +17,6 @@ should probably proofread and complete it, then remove this comment. -->
|
|
17 |
# gemma_FT
|
18 |
|
19 |
This model is a fine-tuned version of [google/gemma-2b-it](https://huggingface.co/google/gemma-2b-it) on the None dataset.
|
20 |
-
It achieves the following results on the evaluation set:
|
21 |
-
- Loss: 2.2294
|
22 |
|
23 |
## Model description
|
24 |
|
@@ -46,63 +44,11 @@ The following hyperparameters were used during training:
|
|
46 |
- optimizer: Adam with betas=(0.9,0.999) and epsilon=1e-08
|
47 |
- lr_scheduler_type: linear
|
48 |
- lr_scheduler_warmup_steps: 2
|
49 |
-
- training_steps:
|
50 |
- mixed_precision_training: Native AMP
|
51 |
|
52 |
### Training results
|
53 |
|
54 |
-
| Training Loss | Epoch | Step | Validation Loss |
|
55 |
-
|:-------------:|:-----:|:----:|:---------------:|
|
56 |
-
| 3.0284 | 0.0 | 1 | 3.5564 |
|
57 |
-
| 3.3693 | 0.0 | 2 | 3.4912 |
|
58 |
-
| 3.2853 | 0.0 | 3 | 3.3325 |
|
59 |
-
| 3.0263 | 0.0 | 4 | 3.1522 |
|
60 |
-
| 3.2309 | 0.0 | 5 | 2.9678 |
|
61 |
-
| 3.0045 | 0.0 | 6 | 2.8090 |
|
62 |
-
| 2.3177 | 0.0 | 7 | 2.7007 |
|
63 |
-
| 2.7699 | 0.0 | 8 | 2.6393 |
|
64 |
-
| 2.593 | 0.0 | 9 | 2.6119 |
|
65 |
-
| 2.6395 | 0.01 | 10 | 2.5837 |
|
66 |
-
| 2.2016 | 0.01 | 11 | 2.5492 |
|
67 |
-
| 2.4271 | 0.01 | 12 | 2.5134 |
|
68 |
-
| 2.6307 | 0.01 | 13 | 2.4833 |
|
69 |
-
| 2.5509 | 0.01 | 14 | 2.4618 |
|
70 |
-
| 2.4261 | 0.01 | 15 | 2.4458 |
|
71 |
-
| 2.468 | 0.01 | 16 | 2.4316 |
|
72 |
-
| 2.2545 | 0.01 | 17 | 2.4174 |
|
73 |
-
| 2.4923 | 0.01 | 18 | 2.4027 |
|
74 |
-
| 2.8034 | 0.01 | 19 | 2.3862 |
|
75 |
-
| 2.4174 | 0.01 | 20 | 2.3703 |
|
76 |
-
| 2.0761 | 0.01 | 21 | 2.3553 |
|
77 |
-
| 2.2617 | 0.01 | 22 | 2.3421 |
|
78 |
-
| 2.582 | 0.01 | 23 | 2.3314 |
|
79 |
-
| 1.9628 | 0.01 | 24 | 2.3239 |
|
80 |
-
| 2.4096 | 0.01 | 25 | 2.3175 |
|
81 |
-
| 2.1909 | 0.01 | 26 | 2.3110 |
|
82 |
-
| 2.3109 | 0.01 | 27 | 2.3045 |
|
83 |
-
| 2.4591 | 0.01 | 28 | 2.2984 |
|
84 |
-
| 2.2765 | 0.02 | 29 | 2.2930 |
|
85 |
-
| 2.0513 | 0.02 | 30 | 2.2875 |
|
86 |
-
| 2.0974 | 0.02 | 31 | 2.2813 |
|
87 |
-
| 2.5818 | 0.02 | 32 | 2.2755 |
|
88 |
-
| 2.3872 | 0.02 | 33 | 2.2701 |
|
89 |
-
| 2.1044 | 0.02 | 34 | 2.2654 |
|
90 |
-
| 2.322 | 0.02 | 35 | 2.2611 |
|
91 |
-
| 2.1259 | 0.02 | 36 | 2.2573 |
|
92 |
-
| 2.3434 | 0.02 | 37 | 2.2538 |
|
93 |
-
| 2.2724 | 0.02 | 38 | 2.2506 |
|
94 |
-
| 2.3195 | 0.02 | 39 | 2.2476 |
|
95 |
-
| 2.458 | 0.02 | 40 | 2.2448 |
|
96 |
-
| 2.2682 | 0.02 | 41 | 2.2423 |
|
97 |
-
| 2.0698 | 0.02 | 42 | 2.2397 |
|
98 |
-
| 1.803 | 0.02 | 43 | 2.2374 |
|
99 |
-
| 2.1955 | 0.02 | 44 | 2.2354 |
|
100 |
-
| 1.9671 | 0.02 | 45 | 2.2336 |
|
101 |
-
| 1.6402 | 0.02 | 46 | 2.2322 |
|
102 |
-
| 1.9659 | 0.02 | 47 | 2.2312 |
|
103 |
-
| 2.4895 | 0.03 | 48 | 2.2303 |
|
104 |
-
| 2.2568 | 0.03 | 49 | 2.2297 |
|
105 |
-
| 1.9725 | 0.03 | 50 | 2.2294 |
|
106 |
|
107 |
|
108 |
### Framework versions
|
|
|
17 |
# gemma_FT
|
18 |
|
19 |
This model is a fine-tuned version of [google/gemma-2b-it](https://huggingface.co/google/gemma-2b-it) on the None dataset.
|
|
|
|
|
20 |
|
21 |
## Model description
|
22 |
|
|
|
44 |
- optimizer: Adam with betas=(0.9,0.999) and epsilon=1e-08
|
45 |
- lr_scheduler_type: linear
|
46 |
- lr_scheduler_warmup_steps: 2
|
47 |
+
- training_steps: 12
|
48 |
- mixed_precision_training: Native AMP
|
49 |
|
50 |
### Training results
|
51 |
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
52 |
|
53 |
|
54 |
### Framework versions
|
adapter_config.json
CHANGED
@@ -19,13 +19,13 @@
|
|
19 |
"rank_pattern": {},
|
20 |
"revision": null,
|
21 |
"target_modules": [
|
22 |
-
"k_proj",
|
23 |
-
"up_proj",
|
24 |
-
"down_proj",
|
25 |
"o_proj",
|
26 |
"q_proj",
|
|
|
27 |
"gate_proj",
|
28 |
-
"
|
|
|
|
|
29 |
],
|
30 |
"task_type": "CAUSAL_LM",
|
31 |
"use_rslora": false
|
|
|
19 |
"rank_pattern": {},
|
20 |
"revision": null,
|
21 |
"target_modules": [
|
|
|
|
|
|
|
22 |
"o_proj",
|
23 |
"q_proj",
|
24 |
+
"k_proj",
|
25 |
"gate_proj",
|
26 |
+
"up_proj",
|
27 |
+
"v_proj",
|
28 |
+
"down_proj"
|
29 |
],
|
30 |
"task_type": "CAUSAL_LM",
|
31 |
"use_rslora": false
|
adapter_model.safetensors
CHANGED
@@ -1,3 +1,3 @@
|
|
1 |
version https://git-lfs.github.com/spec/v1
|
2 |
-
oid sha256:
|
3 |
size 29450584
|
|
|
1 |
version https://git-lfs.github.com/spec/v1
|
2 |
+
oid sha256:5e8edcfe2a8de2682dacf0a9f10557569be7affe05ac359928a869b5a3063cf3
|
3 |
size 29450584
|
training_args.bin
CHANGED
@@ -1,3 +1,3 @@
|
|
1 |
version https://git-lfs.github.com/spec/v1
|
2 |
-
oid sha256:
|
3 |
size 4856
|
|
|
1 |
version https://git-lfs.github.com/spec/v1
|
2 |
+
oid sha256:63d52faaacf58905f69df60603490dddb079e07179df233f9e34b467a3b80126
|
3 |
size 4856
|