Commit
•
f9b552c
1
Parent(s):
9195435
End of training
Browse files- README.md +21 -2
- adapter_model.bin +1 -1
- tokenizer.json +1 -8
- train_with_llm_answers.csv +0 -0
- training_args.bin +1 -1
README.md
CHANGED
|
@@ -17,6 +17,9 @@ should probably proofread and complete it, then remove this comment. -->
|
|
| 17 |
# mistral-7b-llm-science-exam
|
| 18 |
|
| 19 |
This model is a fine-tuned version of [/kaggle/input/mistral-7b/Mistral-7B-v0.1](https://huggingface.co//kaggle/input/mistral-7b/Mistral-7B-v0.1) on the llm-science-exam dataset.
|
|
|
|
|
|
|
|
|
|
| 20 |
|
| 21 |
## Model description
|
| 22 |
|
|
@@ -39,14 +42,30 @@ The following hyperparameters were used during training:
|
|
| 39 |
- train_batch_size: 2
|
| 40 |
- eval_batch_size: 2
|
| 41 |
- seed: 42
|
|
|
|
|
|
|
| 42 |
- optimizer: Adam with betas=(0.9,0.999) and epsilon=1e-08
|
| 43 |
- lr_scheduler_type: cosine
|
| 44 |
-
- lr_scheduler_warmup_steps:
|
| 45 |
- num_epochs: 1
|
| 46 |
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 47 |
### Framework versions
|
| 48 |
|
| 49 |
- Transformers 4.34.0.dev0
|
| 50 |
- Pytorch 2.0.0
|
| 51 |
-
- Datasets 2.
|
| 52 |
- Tokenizers 0.14.0
|
|
|
|
| 17 |
# mistral-7b-llm-science-exam
|
| 18 |
|
| 19 |
This model is a fine-tuned version of [/kaggle/input/mistral-7b/Mistral-7B-v0.1](https://huggingface.co//kaggle/input/mistral-7b/Mistral-7B-v0.1) on the llm-science-exam dataset.
|
| 20 |
+
It achieves the following results on the evaluation set:
|
| 21 |
+
- Loss: 0.3951
|
| 22 |
+
- Map@3: 0.8976
|
| 23 |
|
| 24 |
## Model description
|
| 25 |
|
|
|
|
| 42 |
- train_batch_size: 2
|
| 43 |
- eval_batch_size: 2
|
| 44 |
- seed: 42
|
| 45 |
+
- gradient_accumulation_steps: 4
|
| 46 |
+
- total_train_batch_size: 8
|
| 47 |
- optimizer: Adam with betas=(0.9,0.999) and epsilon=1e-08
|
| 48 |
- lr_scheduler_type: cosine
|
| 49 |
+
- lr_scheduler_warmup_steps: 50
|
| 50 |
- num_epochs: 1
|
| 51 |
|
| 52 |
+
### Training results
|
| 53 |
+
|
| 54 |
+
| Training Loss | Epoch | Step | Validation Loss | Map@3 |
|
| 55 |
+
|:-------------:|:-----:|:----:|:---------------:|:------:|
|
| 56 |
+
| 3.3769 | 0.11 | 50 | 1.8621 | 0.9238 |
|
| 57 |
+
| 1.5772 | 0.23 | 100 | 0.5619 | 0.9119 |
|
| 58 |
+
| 0.9202 | 0.34 | 150 | 0.3942 | 0.9095 |
|
| 59 |
+
| 0.9485 | 0.45 | 200 | 0.4117 | 0.8976 |
|
| 60 |
+
| 0.9698 | 0.56 | 250 | 0.4145 | 0.9048 |
|
| 61 |
+
| 0.8731 | 0.68 | 300 | 0.4054 | 0.9048 |
|
| 62 |
+
| 0.8929 | 0.79 | 350 | 0.3967 | 0.8976 |
|
| 63 |
+
| 0.9737 | 0.9 | 400 | 0.3951 | 0.8976 |
|
| 64 |
+
|
| 65 |
+
|
| 66 |
### Framework versions
|
| 67 |
|
| 68 |
- Transformers 4.34.0.dev0
|
| 69 |
- Pytorch 2.0.0
|
| 70 |
+
- Datasets 2.14.4
|
| 71 |
- Tokenizers 0.14.0
|
adapter_model.bin
CHANGED
|
@@ -1,3 +1,3 @@
|
|
| 1 |
version https://git-lfs.github.com/spec/v1
|
| 2 |
-
oid sha256:
|
| 3 |
size 109097933
|
|
|
|
| 1 |
version https://git-lfs.github.com/spec/v1
|
| 2 |
+
oid sha256:2d86ce6f9f84cb26a1333f2846a479131f01f28104293d0fe5b57b944a02addb
|
| 3 |
size 109097933
|
tokenizer.json
CHANGED
|
@@ -1,14 +1,7 @@
|
|
| 1 |
{
|
| 2 |
"version": "1.0",
|
| 3 |
"truncation": null,
|
| 4 |
-
"padding":
|
| 5 |
-
"strategy": "BatchLongest",
|
| 6 |
-
"direction": "Left",
|
| 7 |
-
"pad_to_multiple_of": null,
|
| 8 |
-
"pad_id": 2,
|
| 9 |
-
"pad_type_id": 0,
|
| 10 |
-
"pad_token": "</s>"
|
| 11 |
-
},
|
| 12 |
"added_tokens": [
|
| 13 |
{
|
| 14 |
"id": 0,
|
|
|
|
| 1 |
{
|
| 2 |
"version": "1.0",
|
| 3 |
"truncation": null,
|
| 4 |
+
"padding": null,
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 5 |
"added_tokens": [
|
| 6 |
{
|
| 7 |
"id": 0,
|
train_with_llm_answers.csv
CHANGED
|
The diff for this file is too large to render.
See raw diff
|
|
|
training_args.bin
CHANGED
|
@@ -1,3 +1,3 @@
|
|
| 1 |
version https://git-lfs.github.com/spec/v1
|
| 2 |
-
oid sha256:
|
| 3 |
size 4091
|
|
|
|
| 1 |
version https://git-lfs.github.com/spec/v1
|
| 2 |
+
oid sha256:8edb6de5035d85d5594bfc396384890041683eb5dd7618eacf9061692226bffb
|
| 3 |
size 4091
|