Commit
·
2fb82b0
1
Parent(s):
607136a
Training in progress, step 500
Browse files- config.json +46 -0
- model.safetensors +3 -0
- runs/Nov07_02-07-47_7de20a233ed2/events.out.tfevents.1699322895.7de20a233ed2.1327.0 +3 -0
- runs/Nov07_02-07-47_7de20a233ed2/events.out.tfevents.1699323001.7de20a233ed2.1327.1 +3 -0
- runs/Nov07_02-11-42_7de20a233ed2/events.out.tfevents.1699323108.7de20a233ed2.1327.2 +3 -0
- runs/Nov07_02-17-44_7de20a233ed2/events.out.tfevents.1699323466.7de20a233ed2.1327.3 +3 -0
- runs/Nov07_02-19-12_7de20a233ed2/events.out.tfevents.1699323552.7de20a233ed2.1327.4 +3 -0
- runs/Nov07_02-22-32_7de20a233ed2/events.out.tfevents.1699323753.7de20a233ed2.1327.5 +3 -0
- runs/Nov07_02-26-40_7de20a233ed2/events.out.tfevents.1699324009.7de20a233ed2.1327.6 +3 -0
- runs/Nov07_02-35-58_7de20a233ed2/events.out.tfevents.1699324559.7de20a233ed2.1327.7 +3 -0
- runs/Nov07_02-36-14_7de20a233ed2/events.out.tfevents.1699324574.7de20a233ed2.1327.8 +3 -0
- runs/Nov07_02-37-13_7de20a233ed2/events.out.tfevents.1699324633.7de20a233ed2.1327.9 +3 -0
- training_args.bin +3 -0
config.json
ADDED
@@ -0,0 +1,46 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
1 |
+
{
|
2 |
+
"_name_or_path": "distilgpt2",
|
3 |
+
"_num_labels": 1,
|
4 |
+
"activation_function": "gelu_new",
|
5 |
+
"architectures": [
|
6 |
+
"GPT2LMHeadModel"
|
7 |
+
],
|
8 |
+
"attn_pdrop": 0.1,
|
9 |
+
"bos_token_id": 50256,
|
10 |
+
"embd_pdrop": 0.1,
|
11 |
+
"eos_token_id": 50256,
|
12 |
+
"id2label": {
|
13 |
+
"0": "LABEL_0"
|
14 |
+
},
|
15 |
+
"initializer_range": 0.02,
|
16 |
+
"label2id": {
|
17 |
+
"LABEL_0": 0
|
18 |
+
},
|
19 |
+
"layer_norm_epsilon": 1e-05,
|
20 |
+
"model_type": "gpt2",
|
21 |
+
"n_ctx": 1024,
|
22 |
+
"n_embd": 768,
|
23 |
+
"n_head": 12,
|
24 |
+
"n_inner": null,
|
25 |
+
"n_layer": 6,
|
26 |
+
"n_positions": 1024,
|
27 |
+
"reorder_and_upcast_attn": false,
|
28 |
+
"resid_pdrop": 0.1,
|
29 |
+
"scale_attn_by_inverse_layer_idx": false,
|
30 |
+
"scale_attn_weights": true,
|
31 |
+
"summary_activation": null,
|
32 |
+
"summary_first_dropout": 0.1,
|
33 |
+
"summary_proj_to_labels": true,
|
34 |
+
"summary_type": "cls_index",
|
35 |
+
"summary_use_proj": true,
|
36 |
+
"task_specific_params": {
|
37 |
+
"text-generation": {
|
38 |
+
"do_sample": true,
|
39 |
+
"max_length": 50
|
40 |
+
}
|
41 |
+
},
|
42 |
+
"torch_dtype": "float32",
|
43 |
+
"transformers_version": "4.35.0",
|
44 |
+
"use_cache": true,
|
45 |
+
"vocab_size": 50258
|
46 |
+
}
|
model.safetensors
ADDED
@@ -0,0 +1,3 @@
|
|
|
|
|
|
|
|
|
1 |
+
version https://git-lfs.github.com/spec/v1
|
2 |
+
oid sha256:f5f490303465c7df27d08f1a5cbd8df05473b51dca3ab8117e6fccc95aeb317c
|
3 |
+
size 327661000
|
runs/Nov07_02-07-47_7de20a233ed2/events.out.tfevents.1699322895.7de20a233ed2.1327.0
ADDED
@@ -0,0 +1,3 @@
|
|
|
|
|
|
|
|
|
1 |
+
version https://git-lfs.github.com/spec/v1
|
2 |
+
oid sha256:60432d82efd8f8d54aa8eb38e98b76d8d9c830986107b018b460709a983afd53
|
3 |
+
size 9046
|
runs/Nov07_02-07-47_7de20a233ed2/events.out.tfevents.1699323001.7de20a233ed2.1327.1
ADDED
@@ -0,0 +1,3 @@
|
|
|
|
|
|
|
|
|
1 |
+
version https://git-lfs.github.com/spec/v1
|
2 |
+
oid sha256:17e611dcac83c5f4c9e4d5fda20453aeacf9680a6a139b74aace905bc94675b1
|
3 |
+
size 4567
|
runs/Nov07_02-11-42_7de20a233ed2/events.out.tfevents.1699323108.7de20a233ed2.1327.2
ADDED
@@ -0,0 +1,3 @@
|
|
|
|
|
|
|
|
|
1 |
+
version https://git-lfs.github.com/spec/v1
|
2 |
+
oid sha256:834092be027b9c983f53ddf73b363f8c28ca213a08e28925c886db1f8b6c9743
|
3 |
+
size 31441
|
runs/Nov07_02-17-44_7de20a233ed2/events.out.tfevents.1699323466.7de20a233ed2.1327.3
ADDED
@@ -0,0 +1,3 @@
|
|
|
|
|
|
|
|
|
1 |
+
version https://git-lfs.github.com/spec/v1
|
2 |
+
oid sha256:da156dee0dc30131492e964f31a5da0f42067a425cd9c24283da3f2b3ec443ac
|
3 |
+
size 18004
|
runs/Nov07_02-19-12_7de20a233ed2/events.out.tfevents.1699323552.7de20a233ed2.1327.4
ADDED
@@ -0,0 +1,3 @@
|
|
|
|
|
|
|
|
|
1 |
+
version https://git-lfs.github.com/spec/v1
|
2 |
+
oid sha256:0c13def3a680b1e4b373f9a49401100521b375c3aeaa30fdf2a9cad20e9544e0
|
3 |
+
size 4567
|
runs/Nov07_02-22-32_7de20a233ed2/events.out.tfevents.1699323753.7de20a233ed2.1327.5
ADDED
@@ -0,0 +1,3 @@
|
|
|
|
|
|
|
|
|
1 |
+
version https://git-lfs.github.com/spec/v1
|
2 |
+
oid sha256:1332207f5d41b047e411e0df6dddae8ce54eceb112d24578508f1f21526ddf68
|
3 |
+
size 18004
|
runs/Nov07_02-26-40_7de20a233ed2/events.out.tfevents.1699324009.7de20a233ed2.1327.6
ADDED
@@ -0,0 +1,3 @@
|
|
|
|
|
|
|
|
|
1 |
+
version https://git-lfs.github.com/spec/v1
|
2 |
+
oid sha256:b04ad1e795474b673e382f4cccc17a1c22ea6728405a73827d9d98a949f2972a
|
3 |
+
size 4567
|
runs/Nov07_02-35-58_7de20a233ed2/events.out.tfevents.1699324559.7de20a233ed2.1327.7
ADDED
@@ -0,0 +1,3 @@
|
|
|
|
|
|
|
|
|
1 |
+
version https://git-lfs.github.com/spec/v1
|
2 |
+
oid sha256:6cd152593bd2817ea04b6d10e6a0af7e9130a3d03ce955c32be1c3ec9d1996ec
|
3 |
+
size 4567
|
runs/Nov07_02-36-14_7de20a233ed2/events.out.tfevents.1699324574.7de20a233ed2.1327.8
ADDED
@@ -0,0 +1,3 @@
|
|
|
|
|
|
|
|
|
1 |
+
version https://git-lfs.github.com/spec/v1
|
2 |
+
oid sha256:15448fc2a49936053fc3f9a39e11dd4c9ca3237db6b3ffc9b6ebc018a79f6fd6
|
3 |
+
size 4567
|
runs/Nov07_02-37-13_7de20a233ed2/events.out.tfevents.1699324633.7de20a233ed2.1327.9
ADDED
@@ -0,0 +1,3 @@
|
|
|
|
|
|
|
|
|
1 |
+
version https://git-lfs.github.com/spec/v1
|
2 |
+
oid sha256:d4f63f9be7df34e5c4ba037b9b9b0049a116c9a3e6cc16a622ca329c6ebd69ac
|
3 |
+
size 4995
|
training_args.bin
ADDED
@@ -0,0 +1,3 @@
|
|
|
|
|
|
|
|
|
1 |
+
version https://git-lfs.github.com/spec/v1
|
2 |
+
oid sha256:958298b49e164c9bbec3d696d728b9b91149e6f19a7b05c7fb85c4e11d9a3a90
|
3 |
+
size 4600
|