End of training
Browse files
README.md
CHANGED
@@ -1,11 +1,19 @@
|
|
1 |
---
|
2 |
base_model: loubnabnl/smollm-1.7b-16k
|
3 |
tags:
|
|
|
|
|
|
|
|
|
4 |
- trl
|
5 |
- sft
|
6 |
- generated_from_trainer
|
7 |
datasets:
|
8 |
-
-
|
|
|
|
|
|
|
|
|
9 |
model-index:
|
10 |
- name: smollm-1.7B-16k-instruct-v0
|
11 |
results: []
|
@@ -17,7 +25,7 @@ should probably proofread and complete it, then remove this comment. -->
|
|
17 |
[<img src="https://raw.githubusercontent.com/wandb/assets/main/wandb-github-badge-28.svg" alt="Visualize in Weights & Biases" width="200" height="32"/>](https://wandb.ai/loubnabnl/huggingface/runs/7rd4806f)
|
18 |
# smollm-1.7B-16k-instruct-v0
|
19 |
|
20 |
-
This model is a fine-tuned version of [loubnabnl/smollm-1.7b-16k](https://huggingface.co/loubnabnl/smollm-1.7b-16k) on the
|
21 |
It achieves the following results on the evaluation set:
|
22 |
- Loss: 1.0118
|
23 |
|
|
|
1 |
---
|
2 |
base_model: loubnabnl/smollm-1.7b-16k
|
3 |
tags:
|
4 |
+
- alignment-handbook
|
5 |
+
- trl
|
6 |
+
- sft
|
7 |
+
- generated_from_trainer
|
8 |
- trl
|
9 |
- sft
|
10 |
- generated_from_trainer
|
11 |
datasets:
|
12 |
+
- HuggingFaceTB/Magpie-Pro-300K-Filtered-H4
|
13 |
+
- HuggingFaceTB/self-oss-instruct-sc2-H4
|
14 |
+
- HuggingFaceTB/OpenHermes-2.5-H4
|
15 |
+
- HuggingFaceTB/everyday-conversations-llama3.1-2k
|
16 |
+
- HuggingFaceTB/instruct-data-basics-smollm-H4
|
17 |
model-index:
|
18 |
- name: smollm-1.7B-16k-instruct-v0
|
19 |
results: []
|
|
|
25 |
[<img src="https://raw.githubusercontent.com/wandb/assets/main/wandb-github-badge-28.svg" alt="Visualize in Weights & Biases" width="200" height="32"/>](https://wandb.ai/loubnabnl/huggingface/runs/7rd4806f)
|
26 |
# smollm-1.7B-16k-instruct-v0
|
27 |
|
28 |
+
This model is a fine-tuned version of [loubnabnl/smollm-1.7b-16k](https://huggingface.co/loubnabnl/smollm-1.7b-16k) on the HuggingFaceTB/Magpie-Pro-300K-Filtered-H4, the HuggingFaceTB/self-oss-instruct-sc2-H4, the HuggingFaceTB/OpenHermes-2.5-H4, the HuggingFaceTB/everyday-conversations-llama3.1-2k and the HuggingFaceTB/instruct-data-basics-smollm-H4 datasets.
|
29 |
It achieves the following results on the evaluation set:
|
30 |
- Loss: 1.0118
|
31 |
|
all_results.json
CHANGED
@@ -1,5 +1,10 @@
|
|
1 |
{
|
2 |
"epoch": 0.9951219512195122,
|
|
|
|
|
|
|
|
|
|
|
3 |
"total_flos": 128481799176192.0,
|
4 |
"train_loss": 0.7558732021088693,
|
5 |
"train_runtime": 2589.5829,
|
|
|
1 |
{
|
2 |
"epoch": 0.9951219512195122,
|
3 |
+
"eval_loss": 1.0117870569229126,
|
4 |
+
"eval_runtime": 124.7042,
|
5 |
+
"eval_samples": 82730,
|
6 |
+
"eval_samples_per_second": 21.026,
|
7 |
+
"eval_steps_per_second": 0.658,
|
8 |
"total_flos": 128481799176192.0,
|
9 |
"train_loss": 0.7558732021088693,
|
10 |
"train_runtime": 2589.5829,
|
config.json
CHANGED
@@ -25,6 +25,6 @@
|
|
25 |
"tie_word_embeddings": true,
|
26 |
"torch_dtype": "bfloat16",
|
27 |
"transformers_version": "4.42.3",
|
28 |
-
"use_cache":
|
29 |
"vocab_size": 49152
|
30 |
}
|
|
|
25 |
"tie_word_embeddings": true,
|
26 |
"torch_dtype": "bfloat16",
|
27 |
"transformers_version": "4.42.3",
|
28 |
+
"use_cache": true,
|
29 |
"vocab_size": 49152
|
30 |
}
|
eval_results.json
ADDED
@@ -0,0 +1,8 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
1 |
+
{
|
2 |
+
"epoch": 0.9951219512195122,
|
3 |
+
"eval_loss": 1.0117870569229126,
|
4 |
+
"eval_runtime": 124.7042,
|
5 |
+
"eval_samples": 82730,
|
6 |
+
"eval_samples_per_second": 21.026,
|
7 |
+
"eval_steps_per_second": 0.658
|
8 |
+
}
|
runs/Sep20_12-08-36_ip-26-0-168-120/events.out.tfevents.1726837978.ip-26-0-168-120.4123050.1
ADDED
@@ -0,0 +1,3 @@
|
|
|
|
|
|
|
|
|
1 |
+
version https://git-lfs.github.com/spec/v1
|
2 |
+
oid sha256:654f4999022be4056a01d98518a74d277017a67b16671f7dd7c1e9aa3d4f589e
|
3 |
+
size 354
|