Model save
Browse files- README.md +5 -20
- events.out.tfevents.1716992315.isl-gpu27.3581638.0 +3 -0
- model.safetensors +1 -1
- training_args.bin +2 -2
README.md
CHANGED
|
@@ -3,8 +3,6 @@ license: apache-2.0
|
|
| 3 |
base_model: JackFram/llama-68m
|
| 4 |
tags:
|
| 5 |
- generated_from_trainer
|
| 6 |
-
metrics:
|
| 7 |
-
- accuracy
|
| 8 |
model-index:
|
| 9 |
- name: recreate_llama_68M_vanilla
|
| 10 |
results: []
|
|
@@ -15,10 +13,7 @@ should probably proofread and complete it, then remove this comment. -->
|
|
| 15 |
|
| 16 |
# recreate_llama_68M_vanilla
|
| 17 |
|
| 18 |
-
This model is a fine-tuned version of [JackFram/llama-68m](https://huggingface.co/JackFram/llama-68m) on
|
| 19 |
-
It achieves the following results on the evaluation set:
|
| 20 |
-
- Loss: 9.5494
|
| 21 |
-
- Accuracy: 0.3512
|
| 22 |
|
| 23 |
## Model description
|
| 24 |
|
|
@@ -37,27 +32,17 @@ More information needed
|
|
| 37 |
### Training hyperparameters
|
| 38 |
|
| 39 |
The following hyperparameters were used during training:
|
| 40 |
-
- learning_rate: 0.
|
| 41 |
-
- train_batch_size:
|
| 42 |
-
- eval_batch_size:
|
| 43 |
- seed: 42
|
| 44 |
- optimizer: Adam with betas=(0.9,0.999) and epsilon=1e-08
|
| 45 |
- lr_scheduler_type: linear
|
|
|
|
| 46 |
- num_epochs: 3.0
|
| 47 |
|
| 48 |
### Training results
|
| 49 |
|
| 50 |
-
| Training Loss | Epoch | Step | Validation Loss | Accuracy |
|
| 51 |
-
|:-------------:|:------:|:----:|:---------------:|:--------:|
|
| 52 |
-
| No log | 0.3125 | 10 | 7.9370 | 0.3676 |
|
| 53 |
-
| No log | 0.625 | 20 | 8.6808 | 0.3478 |
|
| 54 |
-
| No log | 0.9375 | 30 | 10.9798 | 0.1029 |
|
| 55 |
-
| No log | 1.25 | 40 | 10.3023 | 0.2493 |
|
| 56 |
-
| No log | 1.5625 | 50 | 9.7688 | 0.3501 |
|
| 57 |
-
| No log | 1.875 | 60 | 9.6190 | 0.3510 |
|
| 58 |
-
| No log | 2.1875 | 70 | 9.5617 | 0.3510 |
|
| 59 |
-
| No log | 2.5 | 80 | 9.5470 | 0.3511 |
|
| 60 |
-
| No log | 2.8125 | 90 | 9.5487 | 0.3511 |
|
| 61 |
|
| 62 |
|
| 63 |
### Framework versions
|
|
|
|
| 3 |
base_model: JackFram/llama-68m
|
| 4 |
tags:
|
| 5 |
- generated_from_trainer
|
|
|
|
|
|
|
| 6 |
model-index:
|
| 7 |
- name: recreate_llama_68M_vanilla
|
| 8 |
results: []
|
|
|
|
| 13 |
|
| 14 |
# recreate_llama_68M_vanilla
|
| 15 |
|
| 16 |
+
This model is a fine-tuned version of [JackFram/llama-68m](https://huggingface.co/JackFram/llama-68m) on an unknown dataset.
|
|
|
|
|
|
|
|
|
|
| 17 |
|
| 18 |
## Model description
|
| 19 |
|
|
|
|
| 32 |
### Training hyperparameters
|
| 33 |
|
| 34 |
The following hyperparameters were used during training:
|
| 35 |
+
- learning_rate: 0.0001
|
| 36 |
+
- train_batch_size: 12
|
| 37 |
+
- eval_batch_size: 24
|
| 38 |
- seed: 42
|
| 39 |
- optimizer: Adam with betas=(0.9,0.999) and epsilon=1e-08
|
| 40 |
- lr_scheduler_type: linear
|
| 41 |
+
- lr_scheduler_warmup_ratio: 0.05
|
| 42 |
- num_epochs: 3.0
|
| 43 |
|
| 44 |
### Training results
|
| 45 |
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 46 |
|
| 47 |
|
| 48 |
### Framework versions
|
events.out.tfevents.1716992315.isl-gpu27.3581638.0
ADDED
|
@@ -0,0 +1,3 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
version https://git-lfs.github.com/spec/v1
|
| 2 |
+
oid sha256:a3a3ff99ea9c22da16589178c9795aac80195817c2259b7a173bb03508929dfc
|
| 3 |
+
size 5413
|
model.safetensors
CHANGED
|
@@ -1,3 +1,3 @@
|
|
| 1 |
version https://git-lfs.github.com/spec/v1
|
| 2 |
-
oid sha256:
|
| 3 |
size 272123144
|
|
|
|
| 1 |
version https://git-lfs.github.com/spec/v1
|
| 2 |
+
oid sha256:b29f07972e09b7e99ddb0e5808e0d88a17501a11c554772a8caa6699722d71ee
|
| 3 |
size 272123144
|
training_args.bin
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:b997f16f1f5bfcc99bd4bc2fc9d1821e67341b4093297f45f67ca5d81966f1ba
|
| 3 |
+
size 5240
|