Update README.md
Browse files
README.md
CHANGED
@@ -4,11 +4,12 @@ language:
|
|
4 |
- en
|
5 |
license: apache-2.0
|
6 |
base_model: google/flan-t5-xl
|
7 |
-
tags:
|
8 |
-
- generated_from_trainer
|
9 |
model-index:
|
10 |
- name: flan-t5-xl-summary-map-reduce-1024
|
11 |
results: []
|
|
|
|
|
|
|
12 |
---
|
13 |
|
14 |
<!-- This model card has been generated automatically according to the information the Trainer had access to. You
|
@@ -16,22 +17,11 @@ should probably proofread and complete it, then remove this comment. -->
|
|
16 |
|
17 |
# flan-t5-xl-summary-map-reduce-1024
|
18 |
|
19 |
-
This model is a fine-tuned version of [google/flan-t5-xl](https://huggingface.co/google/flan-t5-xl) on the pszemraj/summary-map-reduce dataset.
|
20 |
It achieves the following results on the evaluation set:
|
21 |
- Loss: 0.6039
|
22 |
- Num Input Tokens Seen: 7138765
|
23 |
|
24 |
-
## Model description
|
25 |
-
|
26 |
-
More information needed
|
27 |
-
|
28 |
-
## Intended uses & limitations
|
29 |
-
|
30 |
-
More information needed
|
31 |
-
|
32 |
-
## Training and evaluation data
|
33 |
-
|
34 |
-
More information needed
|
35 |
|
36 |
## Training procedure
|
37 |
|
@@ -48,21 +38,3 @@ The following hyperparameters were used during training:
|
|
48 |
- lr_scheduler_type: cosine
|
49 |
- lr_scheduler_warmup_ratio: 0.05
|
50 |
- num_epochs: 2.0
|
51 |
-
|
52 |
-
### Training results
|
53 |
-
|
54 |
-
| Training Loss | Epoch | Step | Validation Loss | Input Tokens Seen |
|
55 |
-
|:-------------:|:------:|:----:|:---------------:|:-----------------:|
|
56 |
-
| 0.8172 | 0.3851 | 100 | 0.6644 | 1364870 |
|
57 |
-
| 0.7664 | 0.7702 | 200 | 0.6271 | 2744502 |
|
58 |
-
| 0.6584 | 1.1552 | 300 | 0.6146 | 4137699 |
|
59 |
-
| 0.6348 | 1.5403 | 400 | 0.6049 | 5518719 |
|
60 |
-
| 0.6372 | 1.9254 | 500 | 0.6038 | 6895203 |
|
61 |
-
|
62 |
-
|
63 |
-
### Framework versions
|
64 |
-
|
65 |
-
- Transformers 4.46.0.dev0
|
66 |
-
- Pytorch 2.5.1+cu124
|
67 |
-
- Datasets 3.1.0
|
68 |
-
- Tokenizers 0.20.2
|
|
|
4 |
- en
|
5 |
license: apache-2.0
|
6 |
base_model: google/flan-t5-xl
|
|
|
|
|
7 |
model-index:
|
8 |
- name: flan-t5-xl-summary-map-reduce-1024
|
9 |
results: []
|
10 |
+
datasets:
|
11 |
+
- pszemraj/summary-map-reduce-v1
|
12 |
+
pipeline_tag: text2text-generation
|
13 |
---
|
14 |
|
15 |
<!-- This model card has been generated automatically according to the information the Trainer had access to. You
|
|
|
17 |
|
18 |
# flan-t5-xl-summary-map-reduce-1024
|
19 |
|
20 |
+
This model is a fine-tuned version of [google/flan-t5-xl](https://huggingface.co/google/flan-t5-xl) on the pszemraj/summary-map-reduce-v1 dataset.
|
21 |
It achieves the following results on the evaluation set:
|
22 |
- Loss: 0.6039
|
23 |
- Num Input Tokens Seen: 7138765
|
24 |
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
25 |
|
26 |
## Training procedure
|
27 |
|
|
|
38 |
- lr_scheduler_type: cosine
|
39 |
- lr_scheduler_warmup_ratio: 0.05
|
40 |
- num_epochs: 2.0
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|