Training in progress, step 500
Browse files- last-checkpoint/config.json +54 -0
- last-checkpoint/optimizer.pt +3 -0
- last-checkpoint/pytorch_model.bin +3 -0
- last-checkpoint/rng_state.pth +3 -0
- last-checkpoint/scaler.pt +3 -0
- last-checkpoint/scheduler.pt +3 -0
- last-checkpoint/trainer_state.json +22 -0
- last-checkpoint/training_args.bin +3 -0
- pytorch_model.bin +1 -1
- runs/Oct22_08-25-04_c2bafc69294b/1666427318.536094/events.out.tfevents.1666427318.c2bafc69294b.24.1 +3 -0
- runs/Oct22_08-25-04_c2bafc69294b/1666427332.4427907/events.out.tfevents.1666427332.c2bafc69294b.24.2 +3 -0
- runs/Oct22_08-25-04_c2bafc69294b/1666427336.623432/events.out.tfevents.1666427336.c2bafc69294b.24.3 +3 -0
- runs/Oct22_08-25-04_c2bafc69294b/1666427338.9701047/events.out.tfevents.1666427338.c2bafc69294b.24.4 +3 -0
- runs/Oct22_08-25-04_c2bafc69294b/1666427340.18205/events.out.tfevents.1666427340.c2bafc69294b.24.5 +3 -0
- runs/Oct22_08-25-04_c2bafc69294b/1666427340.9704945/events.out.tfevents.1666427340.c2bafc69294b.24.6 +3 -0
- runs/Oct22_08-25-04_c2bafc69294b/1666427341.9964721/events.out.tfevents.1666427341.c2bafc69294b.24.7 +3 -0
- runs/Oct22_08-25-04_c2bafc69294b/1666427342.4280348/events.out.tfevents.1666427342.c2bafc69294b.24.8 +3 -0
- runs/Oct22_08-25-04_c2bafc69294b/1666427343.2115018/events.out.tfevents.1666427343.c2bafc69294b.24.9 +3 -0
- runs/Oct22_08-25-04_c2bafc69294b/1666427343.9733498/events.out.tfevents.1666427343.c2bafc69294b.24.10 +3 -0
- runs/Oct22_08-25-04_c2bafc69294b/1666427345.3813748/events.out.tfevents.1666427345.c2bafc69294b.24.11 +3 -0
- runs/Oct22_08-25-04_c2bafc69294b/events.out.tfevents.1666427318.c2bafc69294b.24.0 +3 -0
- training_args.bin +1 -1
last-checkpoint/config.json
ADDED
@@ -0,0 +1,54 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
1 |
+
{
|
2 |
+
"_name_or_path": "cedpsam/EleutherAI_gpt-neo-125M-stablediffionprompts",
|
3 |
+
"activation_function": "gelu_new",
|
4 |
+
"architectures": [
|
5 |
+
"GPTNeoForCausalLM"
|
6 |
+
],
|
7 |
+
"attention_dropout": 0,
|
8 |
+
"attention_layers": [
|
9 |
+
"global",
|
10 |
+
"local",
|
11 |
+
"global",
|
12 |
+
"local",
|
13 |
+
"global",
|
14 |
+
"local",
|
15 |
+
"global",
|
16 |
+
"local",
|
17 |
+
"global",
|
18 |
+
"local",
|
19 |
+
"global",
|
20 |
+
"local"
|
21 |
+
],
|
22 |
+
"attention_types": [
|
23 |
+
[
|
24 |
+
[
|
25 |
+
"global",
|
26 |
+
"local"
|
27 |
+
],
|
28 |
+
6
|
29 |
+
]
|
30 |
+
],
|
31 |
+
"bos_token_id": 50256,
|
32 |
+
"embed_dropout": 0,
|
33 |
+
"eos_token_id": 50256,
|
34 |
+
"gradient_checkpointing": false,
|
35 |
+
"hidden_size": 768,
|
36 |
+
"initializer_range": 0.02,
|
37 |
+
"intermediate_size": null,
|
38 |
+
"layer_norm_epsilon": 1e-05,
|
39 |
+
"max_position_embeddings": 2048,
|
40 |
+
"model_type": "gpt_neo",
|
41 |
+
"num_heads": 12,
|
42 |
+
"num_layers": 12,
|
43 |
+
"resid_dropout": 0,
|
44 |
+
"summary_activation": null,
|
45 |
+
"summary_first_dropout": 0.1,
|
46 |
+
"summary_proj_to_labels": true,
|
47 |
+
"summary_type": "cls_index",
|
48 |
+
"summary_use_proj": true,
|
49 |
+
"torch_dtype": "float32",
|
50 |
+
"transformers_version": "4.20.1",
|
51 |
+
"use_cache": true,
|
52 |
+
"vocab_size": 50257,
|
53 |
+
"window_size": 256
|
54 |
+
}
|
last-checkpoint/optimizer.pt
ADDED
@@ -0,0 +1,3 @@
|
|
|
|
|
|
|
|
|
1 |
+
version https://git-lfs.github.com/spec/v1
|
2 |
+
oid sha256:461cabf32d01c64c059826e739f859a183ff61a899c5bb3b5ece0165fef4a754
|
3 |
+
size 1001681601
|
last-checkpoint/pytorch_model.bin
ADDED
@@ -0,0 +1,3 @@
|
|
|
|
|
|
|
|
|
1 |
+
version https://git-lfs.github.com/spec/v1
|
2 |
+
oid sha256:224d0c9982a82b06722dc9427dd43a88136b9b487620e919d50604271f47a28f
|
3 |
+
size 551185105
|
last-checkpoint/rng_state.pth
ADDED
@@ -0,0 +1,3 @@
|
|
|
|
|
|
|
|
|
1 |
+
version https://git-lfs.github.com/spec/v1
|
2 |
+
oid sha256:62851f63d26f31f0098aaf22bfc590da9d5dd492f2a149b0c502c13be7100a3d
|
3 |
+
size 14503
|
last-checkpoint/scaler.pt
ADDED
@@ -0,0 +1,3 @@
|
|
|
|
|
|
|
|
|
1 |
+
version https://git-lfs.github.com/spec/v1
|
2 |
+
oid sha256:3978f33a9eccdea52f77a26979e6a0077a50cd239694fba6bc25cd0891e8edf1
|
3 |
+
size 559
|
last-checkpoint/scheduler.pt
ADDED
@@ -0,0 +1,3 @@
|
|
|
|
|
|
|
|
|
1 |
+
version https://git-lfs.github.com/spec/v1
|
2 |
+
oid sha256:4af5659a4ffafd6e5e4486d3da6802b3d89fc41fd3462038b6768c0dd81c78f7
|
3 |
+
size 623
|
last-checkpoint/trainer_state.json
ADDED
@@ -0,0 +1,22 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
1 |
+
{
|
2 |
+
"best_metric": null,
|
3 |
+
"best_model_checkpoint": null,
|
4 |
+
"epoch": 0.006782603977318972,
|
5 |
+
"global_step": 500,
|
6 |
+
"is_hyper_param_search": false,
|
7 |
+
"is_local_process_zero": true,
|
8 |
+
"is_world_process_zero": true,
|
9 |
+
"log_history": [
|
10 |
+
{
|
11 |
+
"epoch": 0.01,
|
12 |
+
"learning_rate": 9.886136363636363e-08,
|
13 |
+
"loss": 0.0261,
|
14 |
+
"step": 500
|
15 |
+
}
|
16 |
+
],
|
17 |
+
"max_steps": 44000,
|
18 |
+
"num_train_epochs": 1,
|
19 |
+
"total_flos": 524503851466752.0,
|
20 |
+
"trial_name": null,
|
21 |
+
"trial_params": null
|
22 |
+
}
|
last-checkpoint/training_args.bin
ADDED
@@ -0,0 +1,3 @@
|
|
|
|
|
|
|
|
|
1 |
+
version https://git-lfs.github.com/spec/v1
|
2 |
+
oid sha256:ba5cc09c321a679be90f03709cafea28721cc11a1abc076f191fcf60d47c338b
|
3 |
+
size 3439
|
pytorch_model.bin
CHANGED
@@ -1,3 +1,3 @@
|
|
1 |
version https://git-lfs.github.com/spec/v1
|
2 |
-
oid sha256:
|
3 |
size 551185105
|
|
|
1 |
version https://git-lfs.github.com/spec/v1
|
2 |
+
oid sha256:224d0c9982a82b06722dc9427dd43a88136b9b487620e919d50604271f47a28f
|
3 |
size 551185105
|
runs/Oct22_08-25-04_c2bafc69294b/1666427318.536094/events.out.tfevents.1666427318.c2bafc69294b.24.1
ADDED
@@ -0,0 +1,3 @@
|
|
|
|
|
|
|
|
|
1 |
+
version https://git-lfs.github.com/spec/v1
|
2 |
+
oid sha256:5a162168f46beb134b374ab1d9cf7ac5bc5ebafc5ce32b74455cef745dca356b
|
3 |
+
size 5438
|
runs/Oct22_08-25-04_c2bafc69294b/1666427332.4427907/events.out.tfevents.1666427332.c2bafc69294b.24.2
ADDED
@@ -0,0 +1,3 @@
|
|
|
|
|
|
|
|
|
1 |
+
version https://git-lfs.github.com/spec/v1
|
2 |
+
oid sha256:8ec93908be8b4f0522d5c515994b95006fd37e35910bd700f3734c6418b9b146
|
3 |
+
size 5438
|
runs/Oct22_08-25-04_c2bafc69294b/1666427336.623432/events.out.tfevents.1666427336.c2bafc69294b.24.3
ADDED
@@ -0,0 +1,3 @@
|
|
|
|
|
|
|
|
|
1 |
+
version https://git-lfs.github.com/spec/v1
|
2 |
+
oid sha256:8e79acbcef1af07f07568ed8e92f6e92e786d55b27512ab11016b6f060417530
|
3 |
+
size 5438
|
runs/Oct22_08-25-04_c2bafc69294b/1666427338.9701047/events.out.tfevents.1666427338.c2bafc69294b.24.4
ADDED
@@ -0,0 +1,3 @@
|
|
|
|
|
|
|
|
|
1 |
+
version https://git-lfs.github.com/spec/v1
|
2 |
+
oid sha256:8815e18fee8cf23184c001f444bb8db3cca239dc8b65ffdff55f4abfa6be4e8e
|
3 |
+
size 5438
|
runs/Oct22_08-25-04_c2bafc69294b/1666427340.18205/events.out.tfevents.1666427340.c2bafc69294b.24.5
ADDED
@@ -0,0 +1,3 @@
|
|
|
|
|
|
|
|
|
1 |
+
version https://git-lfs.github.com/spec/v1
|
2 |
+
oid sha256:403b74ec6a9d2f20cca9299073a7a1946407e75924cf2bf367959776d04b14f2
|
3 |
+
size 5438
|
runs/Oct22_08-25-04_c2bafc69294b/1666427340.9704945/events.out.tfevents.1666427340.c2bafc69294b.24.6
ADDED
@@ -0,0 +1,3 @@
|
|
|
|
|
|
|
|
|
1 |
+
version https://git-lfs.github.com/spec/v1
|
2 |
+
oid sha256:3e4dee12d7544790d7e986e2f92185690a92ffa30115524d4b351b28dc3d346a
|
3 |
+
size 5438
|
runs/Oct22_08-25-04_c2bafc69294b/1666427341.9964721/events.out.tfevents.1666427341.c2bafc69294b.24.7
ADDED
@@ -0,0 +1,3 @@
|
|
|
|
|
|
|
|
|
1 |
+
version https://git-lfs.github.com/spec/v1
|
2 |
+
oid sha256:b88ed860717d9976203b9b8a331731a122cdb9cc4a3ce2420fe189ccc34e7af6
|
3 |
+
size 5438
|
runs/Oct22_08-25-04_c2bafc69294b/1666427342.4280348/events.out.tfevents.1666427342.c2bafc69294b.24.8
ADDED
@@ -0,0 +1,3 @@
|
|
|
|
|
|
|
|
|
1 |
+
version https://git-lfs.github.com/spec/v1
|
2 |
+
oid sha256:d953fb50aa5dca2850ecd44af2bb3fedcd807eab41e219c699e6bcc91fb58547
|
3 |
+
size 5438
|
runs/Oct22_08-25-04_c2bafc69294b/1666427343.2115018/events.out.tfevents.1666427343.c2bafc69294b.24.9
ADDED
@@ -0,0 +1,3 @@
|
|
|
|
|
|
|
|
|
1 |
+
version https://git-lfs.github.com/spec/v1
|
2 |
+
oid sha256:c20f7a983c6723d8258a320e389ab68f52434649ff3e65cd923f3452ba8f89d1
|
3 |
+
size 5438
|
runs/Oct22_08-25-04_c2bafc69294b/1666427343.9733498/events.out.tfevents.1666427343.c2bafc69294b.24.10
ADDED
@@ -0,0 +1,3 @@
|
|
|
|
|
|
|
|
|
1 |
+
version https://git-lfs.github.com/spec/v1
|
2 |
+
oid sha256:41097aea0e5651bc66dd439f6755aab88ce0bb2037b5c323081716bb550ad974
|
3 |
+
size 5438
|
runs/Oct22_08-25-04_c2bafc69294b/1666427345.3813748/events.out.tfevents.1666427345.c2bafc69294b.24.11
ADDED
@@ -0,0 +1,3 @@
|
|
|
|
|
|
|
|
|
1 |
+
version https://git-lfs.github.com/spec/v1
|
2 |
+
oid sha256:a7485f1871947b152a1a0fcc95a9749dbaddf4f350653ad72f0e15aa070678d5
|
3 |
+
size 5438
|
runs/Oct22_08-25-04_c2bafc69294b/events.out.tfevents.1666427318.c2bafc69294b.24.0
ADDED
@@ -0,0 +1,3 @@
|
|
|
|
|
|
|
|
|
1 |
+
version https://git-lfs.github.com/spec/v1
|
2 |
+
oid sha256:6eeda11837e40dc31cd881bbd1ecf0cc0b8d0fe243a27afbabb7a0d66a367085
|
3 |
+
size 45803
|
training_args.bin
CHANGED
@@ -1,3 +1,3 @@
|
|
1 |
version https://git-lfs.github.com/spec/v1
|
2 |
-
oid sha256:
|
3 |
size 3439
|
|
|
1 |
version https://git-lfs.github.com/spec/v1
|
2 |
+
oid sha256:ba5cc09c321a679be90f03709cafea28721cc11a1abc076f191fcf60d47c338b
|
3 |
size 3439
|