Commit
•
1cef3af
1
Parent(s):
8190725
Model save
Browse files- README.md +11 -10
- adapter_model.safetensors +1 -1
- all_results.json +5 -10
- runs/May04_00-40-05_poseidon/events.out.tfevents.1714783215.poseidon.1378242.0 +2 -2
- train_results.json +5 -5
- trainer_state.json +0 -0
README.md
CHANGED
@@ -2,13 +2,12 @@
|
|
2 |
license: apache-2.0
|
3 |
library_name: peft
|
4 |
tags:
|
5 |
-
- alignment-handbook
|
6 |
- trl
|
7 |
- sft
|
8 |
- generated_from_trainer
|
9 |
base_model: TinyLlama/TinyLlama-1.1B-intermediate-step-1431k-3T
|
10 |
datasets:
|
11 |
-
-
|
12 |
model-index:
|
13 |
- name: tinyllama-1.1b-sum-sft-qlora
|
14 |
results: []
|
@@ -19,9 +18,9 @@ should probably proofread and complete it, then remove this comment. -->
|
|
19 |
|
20 |
# tinyllama-1.1b-sum-sft-qlora
|
21 |
|
22 |
-
This model is a fine-tuned version of [TinyLlama/TinyLlama-1.1B-intermediate-step-1431k-3T](https://huggingface.co/TinyLlama/TinyLlama-1.1B-intermediate-step-1431k-3T) on the
|
23 |
It achieves the following results on the evaluation set:
|
24 |
-
- Loss: 2.
|
25 |
|
26 |
## Model description
|
27 |
|
@@ -40,24 +39,26 @@ More information needed
|
|
40 |
### Training hyperparameters
|
41 |
|
42 |
The following hyperparameters were used during training:
|
43 |
-
- learning_rate: 0.
|
44 |
-
- train_batch_size:
|
45 |
- eval_batch_size: 8
|
46 |
- seed: 42
|
47 |
- distributed_type: multi-GPU
|
48 |
- gradient_accumulation_steps: 2
|
49 |
-
- total_train_batch_size:
|
50 |
- optimizer: Adam with betas=(0.9,0.999) and epsilon=1e-08
|
51 |
- lr_scheduler_type: cosine
|
52 |
- lr_scheduler_warmup_ratio: 0.1
|
53 |
-
- num_epochs:
|
54 |
|
55 |
### Training results
|
56 |
|
57 |
| Training Loss | Epoch | Step | Validation Loss |
|
58 |
|:-------------:|:-----:|:-----:|:---------------:|
|
59 |
-
| 2.
|
60 |
-
| 2.
|
|
|
|
|
61 |
|
62 |
|
63 |
### Framework versions
|
|
|
2 |
license: apache-2.0
|
3 |
library_name: peft
|
4 |
tags:
|
|
|
5 |
- trl
|
6 |
- sft
|
7 |
- generated_from_trainer
|
8 |
base_model: TinyLlama/TinyLlama-1.1B-intermediate-step-1431k-3T
|
9 |
datasets:
|
10 |
+
- generator
|
11 |
model-index:
|
12 |
- name: tinyllama-1.1b-sum-sft-qlora
|
13 |
results: []
|
|
|
18 |
|
19 |
# tinyllama-1.1b-sum-sft-qlora
|
20 |
|
21 |
+
This model is a fine-tuned version of [TinyLlama/TinyLlama-1.1B-intermediate-step-1431k-3T](https://huggingface.co/TinyLlama/TinyLlama-1.1B-intermediate-step-1431k-3T) on the generator dataset.
|
22 |
It achieves the following results on the evaluation set:
|
23 |
+
- Loss: 2.1479
|
24 |
|
25 |
## Model description
|
26 |
|
|
|
39 |
### Training hyperparameters
|
40 |
|
41 |
The following hyperparameters were used during training:
|
42 |
+
- learning_rate: 0.0002
|
43 |
+
- train_batch_size: 4
|
44 |
- eval_batch_size: 8
|
45 |
- seed: 42
|
46 |
- distributed_type: multi-GPU
|
47 |
- gradient_accumulation_steps: 2
|
48 |
+
- total_train_batch_size: 8
|
49 |
- optimizer: Adam with betas=(0.9,0.999) and epsilon=1e-08
|
50 |
- lr_scheduler_type: cosine
|
51 |
- lr_scheduler_warmup_ratio: 0.1
|
52 |
+
- num_epochs: 4
|
53 |
|
54 |
### Training results
|
55 |
|
56 |
| Training Loss | Epoch | Step | Validation Loss |
|
57 |
|:-------------:|:-----:|:-----:|:---------------:|
|
58 |
+
| 2.1066 | 1.0 | 10792 | 2.1614 |
|
59 |
+
| 2.144 | 2.0 | 21584 | 2.1491 |
|
60 |
+
| 2.1212 | 3.0 | 32376 | 2.1476 |
|
61 |
+
| 2.0543 | 4.0 | 43168 | 2.1479 |
|
62 |
|
63 |
|
64 |
### Framework versions
|
adapter_model.safetensors
CHANGED
@@ -1,3 +1,3 @@
|
|
1 |
version https://git-lfs.github.com/spec/v1
|
2 |
-
oid sha256:
|
3 |
size 25272360
|
|
|
1 |
version https://git-lfs.github.com/spec/v1
|
2 |
+
oid sha256:f4c1a3caa3685a65acbe6d90ee1bba0f15b694282fd68a959427c61afb59b1e9
|
3 |
size 25272360
|
all_results.json
CHANGED
@@ -1,13 +1,8 @@
|
|
1 |
{
|
2 |
-
"epoch":
|
3 |
-
"
|
4 |
-
"
|
5 |
-
"eval_samples": 6553,
|
6 |
-
"eval_samples_per_second": 25.555,
|
7 |
-
"eval_steps_per_second": 3.195,
|
8 |
-
"train_loss": 2.1470370752908954,
|
9 |
-
"train_runtime": 23954.8129,
|
10 |
"train_samples": 116722,
|
11 |
-
"train_samples_per_second":
|
12 |
-
"train_steps_per_second": 0.
|
13 |
}
|
|
|
1 |
{
|
2 |
+
"epoch": 4.0,
|
3 |
+
"train_loss": 2.0942285917837413,
|
4 |
+
"train_runtime": 51337.8051,
|
|
|
|
|
|
|
|
|
|
|
5 |
"train_samples": 116722,
|
6 |
+
"train_samples_per_second": 6.727,
|
7 |
+
"train_steps_per_second": 0.841
|
8 |
}
|
runs/May04_00-40-05_poseidon/events.out.tfevents.1714783215.poseidon.1378242.0
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 sha256:6c33a15ea2d21cc3c25c70c4331e3dd45282e0f3150270b51126984a03fefed5
|
3 |
+
size 1815973
|
train_results.json
CHANGED
@@ -1,8 +1,8 @@
|
|
1 |
{
|
2 |
-
"epoch":
|
3 |
-
"train_loss": 2.
|
4 |
-
"train_runtime":
|
5 |
"train_samples": 116722,
|
6 |
-
"train_samples_per_second":
|
7 |
-
"train_steps_per_second": 0.
|
8 |
}
|
|
|
1 |
{
|
2 |
+
"epoch": 4.0,
|
3 |
+
"train_loss": 2.0942285917837413,
|
4 |
+
"train_runtime": 51337.8051,
|
5 |
"train_samples": 116722,
|
6 |
+
"train_samples_per_second": 6.727,
|
7 |
+
"train_steps_per_second": 0.841
|
8 |
}
|
trainer_state.json
CHANGED
The diff for this file is too large to render.
See raw diff
|
|