End of training
Browse files
README.md
CHANGED
@@ -1,10 +1,15 @@
|
|
1 |
---
|
|
|
2 |
tags:
|
|
|
|
|
|
|
|
|
3 |
- trl
|
4 |
- sft
|
5 |
- generated_from_trainer
|
6 |
datasets:
|
7 |
-
-
|
8 |
model-index:
|
9 |
- name: zephyr-2b-gemma-dft
|
10 |
results: []
|
@@ -16,7 +21,9 @@ should probably proofread and complete it, then remove this comment. -->
|
|
16 |
[<img src="https://raw.githubusercontent.com/wandb/assets/main/wandb-github-badge-28.svg" alt="Visualize in Weights & Biases" width="200" height="32"/>](https://zebra.wandb.io/cto/distillm/runs/vxz2pnbq)
|
17 |
# zephyr-2b-gemma-dft
|
18 |
|
19 |
-
This model
|
|
|
|
|
20 |
|
21 |
## Model description
|
22 |
|
|
|
1 |
---
|
2 |
+
base_model: data/gemma-2b
|
3 |
tags:
|
4 |
+
- alignment-handbook
|
5 |
+
- trl
|
6 |
+
- sft
|
7 |
+
- generated_from_trainer
|
8 |
- trl
|
9 |
- sft
|
10 |
- generated_from_trainer
|
11 |
datasets:
|
12 |
+
- argilla/dpo-mix-7k
|
13 |
model-index:
|
14 |
- name: zephyr-2b-gemma-dft
|
15 |
results: []
|
|
|
21 |
[<img src="https://raw.githubusercontent.com/wandb/assets/main/wandb-github-badge-28.svg" alt="Visualize in Weights & Biases" width="200" height="32"/>](https://zebra.wandb.io/cto/distillm/runs/vxz2pnbq)
|
22 |
# zephyr-2b-gemma-dft
|
23 |
|
24 |
+
This model is a fine-tuned version of [data/gemma-2b](https://huggingface.co/data/gemma-2b) on the argilla/dpo-mix-7k dataset.
|
25 |
+
It achieves the following results on the evaluation set:
|
26 |
+
- Loss: 0.0001
|
27 |
|
28 |
## Model description
|
29 |
|
all_results.json
CHANGED
@@ -1,5 +1,10 @@
|
|
1 |
{
|
2 |
"epoch": 0,
|
|
|
|
|
|
|
|
|
|
|
3 |
"total_flos": 0,
|
4 |
"train_loss": 0.0,
|
5 |
"train_runtime": 5.1094,
|
|
|
1 |
{
|
2 |
"epoch": 0,
|
3 |
+
"eval_loss": 5.277253148960881e-05,
|
4 |
+
"eval_runtime": 76.218,
|
5 |
+
"eval_samples": 750,
|
6 |
+
"eval_samples_per_second": 12.74,
|
7 |
+
"eval_steps_per_second": 1.601,
|
8 |
"total_flos": 0,
|
9 |
"train_loss": 0.0,
|
10 |
"train_runtime": 5.1094,
|
config.json
CHANGED
@@ -24,6 +24,6 @@
|
|
24 |
"rope_theta": 10000.0,
|
25 |
"torch_dtype": "bfloat16",
|
26 |
"transformers_version": "4.42.4",
|
27 |
-
"use_cache":
|
28 |
"vocab_size": 256000
|
29 |
}
|
|
|
24 |
"rope_theta": 10000.0,
|
25 |
"torch_dtype": "bfloat16",
|
26 |
"transformers_version": "4.42.4",
|
27 |
+
"use_cache": true,
|
28 |
"vocab_size": 256000
|
29 |
}
|
eval_results.json
CHANGED
@@ -1,8 +1,8 @@
|
|
1 |
{
|
2 |
"epoch": 0,
|
3 |
"eval_loss": 5.277253148960881e-05,
|
4 |
-
"eval_runtime": 76.
|
5 |
"eval_samples": 750,
|
6 |
-
"eval_samples_per_second": 12.
|
7 |
-
"eval_steps_per_second": 1.
|
8 |
}
|
|
|
1 |
{
|
2 |
"epoch": 0,
|
3 |
"eval_loss": 5.277253148960881e-05,
|
4 |
+
"eval_runtime": 76.218,
|
5 |
"eval_samples": 750,
|
6 |
+
"eval_samples_per_second": 12.74,
|
7 |
+
"eval_steps_per_second": 1.601
|
8 |
}
|
runs/Jul23_16-11-48_ale-distillm-8-0-0/events.out.tfevents.1721747639.ale-distillm-8-0-0.69762.1
ADDED
@@ -0,0 +1,3 @@
|
|
|
|
|
|
|
|
|
1 |
+
version https://git-lfs.github.com/spec/v1
|
2 |
+
oid sha256:bebad7a947c42b08bd748f2d0e8f90adef828c9a03e02faecdf0737fef4f3af1
|
3 |
+
size 344
|