Training in progress, epoch 4
Browse files
config.json
ADDED
@@ -0,0 +1,42 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
1 |
+
{
|
2 |
+
"_name_or_path": "PlanTL-GOB-ES/gpt2-large-bne",
|
3 |
+
"activation_function": "gelu_new",
|
4 |
+
"architectures": [
|
5 |
+
"GPT2ForSequenceClassification"
|
6 |
+
],
|
7 |
+
"attn_pdrop": 0.0,
|
8 |
+
"bos_token_id": 2,
|
9 |
+
"embd_pdrop": 0.0,
|
10 |
+
"eos_token_id": 2,
|
11 |
+
"initializer_range": 0.02,
|
12 |
+
"layer_norm_epsilon": 1e-06,
|
13 |
+
"model_type": "gpt2",
|
14 |
+
"n_ctx": 512,
|
15 |
+
"n_embd": 1280,
|
16 |
+
"n_head": 20,
|
17 |
+
"n_inner": null,
|
18 |
+
"n_layer": 36,
|
19 |
+
"n_positions": 512,
|
20 |
+
"pad_token_id": 2,
|
21 |
+
"prefix": "</s>",
|
22 |
+
"problem_type": "single_label_classification",
|
23 |
+
"reorder_and_upcast_attn": false,
|
24 |
+
"resid_pdrop": 0.0,
|
25 |
+
"scale_attn_by_inverse_layer_idx": false,
|
26 |
+
"scale_attn_weights": true,
|
27 |
+
"summary_activation": null,
|
28 |
+
"summary_first_dropout": 0.0,
|
29 |
+
"summary_proj_to_labels": true,
|
30 |
+
"summary_type": "cls_index",
|
31 |
+
"summary_use_proj": true,
|
32 |
+
"task_specific_params": {
|
33 |
+
"text-generation": {
|
34 |
+
"do_sample": true,
|
35 |
+
"max_length": 50
|
36 |
+
}
|
37 |
+
},
|
38 |
+
"torch_dtype": "float32",
|
39 |
+
"transformers_version": "4.39.3",
|
40 |
+
"use_cache": true,
|
41 |
+
"vocab_size": 50263
|
42 |
+
}
|
logs/events.out.tfevents.1714931938.gh01.198328.0
ADDED
@@ -0,0 +1,3 @@
|
|
|
|
|
|
|
|
|
1 |
+
version https://git-lfs.github.com/spec/v1
|
2 |
+
oid sha256:ad14ad2f6be06a28b1baf9029ea25f6c3e3ba384811cc0dd1a8c56f21eec4ace
|
3 |
+
size 45696
|
logs/events.out.tfevents.1714932442.gh01.198328.1
ADDED
@@ -0,0 +1,3 @@
|
|
|
|
|
|
|
|
|
1 |
+
version https://git-lfs.github.com/spec/v1
|
2 |
+
oid sha256:ceb73b4fadb323e0bef0b33032ce8534ecfa8cea39e9b538800a92e63b744d89
|
3 |
+
size 512
|
logs/events.out.tfevents.1714933609.gh01.199031.0
ADDED
@@ -0,0 +1,3 @@
|
|
|
|
|
|
|
|
|
1 |
+
version https://git-lfs.github.com/spec/v1
|
2 |
+
oid sha256:cec3d11b3a1bbb25d9c68b0b3c301c73380548d5423f38933c8866b629765b23
|
3 |
+
size 45373
|
model.safetensors
ADDED
@@ -0,0 +1,3 @@
|
|
|
|
|
|
|
|
|
1 |
+
version https://git-lfs.github.com/spec/v1
|
2 |
+
oid sha256:aab7f9498930dbefe5653e15ed541b862d92008d39561c15c0ee2fd4ad10f093
|
3 |
+
size 3093585520
|
training_args.bin
ADDED
@@ -0,0 +1,3 @@
|
|
|
|
|
|
|
|
|
1 |
+
version https://git-lfs.github.com/spec/v1
|
2 |
+
oid sha256:4913c6b4f3ac5d4186b917fa50fb1f183b601356cf97a086ff7191d88b6a7496
|
3 |
+
size 5048
|