Commit
•
ff29418
1
Parent(s):
ad7e0fb
Gemma2b-APPS
Browse files- README.md +30 -1
- training_args.bin +1 -1
README.md
CHANGED
|
@@ -15,6 +15,8 @@ should probably proofread and complete it, then remove this comment. -->
|
|
| 15 |
# Gemma2b-APPS
|
| 16 |
|
| 17 |
This model is a fine-tuned version of [google/gemma-2b](https://huggingface.co/google/gemma-2b) on an unknown dataset.
|
|
|
|
|
|
|
| 18 |
|
| 19 |
## Model description
|
| 20 |
|
|
@@ -42,10 +44,37 @@ The following hyperparameters were used during training:
|
|
| 42 |
- optimizer: Adam with betas=(0.9,0.999) and epsilon=1e-08
|
| 43 |
- lr_scheduler_type: cosine
|
| 44 |
- lr_scheduler_warmup_steps: 100
|
| 45 |
-
- training_steps:
|
| 46 |
|
| 47 |
### Training results
|
| 48 |
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 49 |
|
| 50 |
|
| 51 |
### Framework versions
|
|
|
|
| 15 |
# Gemma2b-APPS
|
| 16 |
|
| 17 |
This model is a fine-tuned version of [google/gemma-2b](https://huggingface.co/google/gemma-2b) on an unknown dataset.
|
| 18 |
+
It achieves the following results on the evaluation set:
|
| 19 |
+
- Loss: 1.2944
|
| 20 |
|
| 21 |
## Model description
|
| 22 |
|
|
|
|
| 44 |
- optimizer: Adam with betas=(0.9,0.999) and epsilon=1e-08
|
| 45 |
- lr_scheduler_type: cosine
|
| 46 |
- lr_scheduler_warmup_steps: 100
|
| 47 |
+
- training_steps: 2500
|
| 48 |
|
| 49 |
### Training results
|
| 50 |
|
| 51 |
+
| Training Loss | Epoch | Step | Validation Loss |
|
| 52 |
+
|:-------------:|:-----:|:----:|:---------------:|
|
| 53 |
+
| 1.2046 | 0.04 | 100 | 1.2184 |
|
| 54 |
+
| 1.0393 | 0.08 | 200 | 1.1178 |
|
| 55 |
+
| 1.064 | 0.12 | 300 | 1.1963 |
|
| 56 |
+
| 1.0558 | 0.16 | 400 | 1.2121 |
|
| 57 |
+
| 0.9774 | 0.2 | 500 | 1.1823 |
|
| 58 |
+
| 1.0325 | 0.24 | 600 | 1.1829 |
|
| 59 |
+
| 1.0572 | 0.28 | 700 | 1.1875 |
|
| 60 |
+
| 0.9972 | 0.32 | 800 | 1.1901 |
|
| 61 |
+
| 0.9901 | 0.36 | 900 | 1.2049 |
|
| 62 |
+
| 1.0587 | 0.4 | 1000 | 1.2187 |
|
| 63 |
+
| 1.0643 | 0.44 | 1100 | 1.2270 |
|
| 64 |
+
| 1.1095 | 0.48 | 1200 | 1.2266 |
|
| 65 |
+
| 1.0179 | 0.52 | 1300 | 1.2468 |
|
| 66 |
+
| 1.0581 | 0.56 | 1400 | 1.2480 |
|
| 67 |
+
| 1.0722 | 0.6 | 1500 | 1.2642 |
|
| 68 |
+
| 1.1467 | 0.64 | 1600 | 1.2622 |
|
| 69 |
+
| 1.0536 | 0.68 | 1700 | 1.2816 |
|
| 70 |
+
| 1.0757 | 0.72 | 1800 | 1.2840 |
|
| 71 |
+
| 1.1504 | 0.76 | 1900 | 1.2884 |
|
| 72 |
+
| 1.1494 | 0.8 | 2000 | 1.2870 |
|
| 73 |
+
| 1.1206 | 0.84 | 2100 | 1.2888 |
|
| 74 |
+
| 1.0454 | 0.88 | 2200 | 1.2938 |
|
| 75 |
+
| 1.1487 | 0.92 | 2300 | 1.2944 |
|
| 76 |
+
| 1.1521 | 0.96 | 2400 | 1.2945 |
|
| 77 |
+
| 1.1155 | 1.0 | 2500 | 1.2944 |
|
| 78 |
|
| 79 |
|
| 80 |
### Framework versions
|
training_args.bin
CHANGED
|
@@ -1,3 +1,3 @@
|
|
| 1 |
version https://git-lfs.github.com/spec/v1
|
| 2 |
-
oid sha256:
|
| 3 |
size 4792
|
|
|
|
| 1 |
version https://git-lfs.github.com/spec/v1
|
| 2 |
+
oid sha256:64b99def95b4443ef812418a00edbb682e9fcd536a93b32edf5a854f565c371a
|
| 3 |
size 4792
|