Hugging Face's logo

mllm-dev
/
gpt2_f_experiment_3_1000

Text Classification
Transformers
Safetensors
gpt2
Inference Endpoints
text-generation-inference
Model card Files Community
mllm-dev commited on
Commit
126d8fc
•
1 Parent(s): 950ae7f

Upload folder using huggingface_hub

Browse files
Files changed (7) hide show
  1. model.safetensors +1 -1
  2. optimizer.pt +1 -1
  3. rng_state.pth +1 -1
  4. scheduler.pt +1 -1
  5. tokenizer.json +6 -1
  6. trainer_state.json +11 -11
  7. training_args.bin +1 -1
model.safetensors CHANGED
@@ -1,3 +1,3 @@
1
  version https://git-lfs.github.com/spec/v1
2
- oid sha256:bdd26392e733acc2bf07df4499675cff7c49cca3856bde1da7f12a201123c995
3
  size 497789648
 
1
  version https://git-lfs.github.com/spec/v1
2
+ oid sha256:b24102379d11eae252c467580205e7bc75a42a5181f37e2609bc08f77760f1dc
3
  size 497789648
optimizer.pt CHANGED
@@ -1,3 +1,3 @@
1
  version https://git-lfs.github.com/spec/v1
2
- oid sha256:afaffdb5a9c70a4c9418f3e115da46e1fd9b34022e32a476bee055c519a6b9f0
3
  size 995673413
 
1
  version https://git-lfs.github.com/spec/v1
2
+ oid sha256:d61276dc4beccfeac8f5a576b794611af91a0e1a29e03e465c0f6306c208115e
3
  size 995673413
rng_state.pth CHANGED
@@ -1,3 +1,3 @@
1
  version https://git-lfs.github.com/spec/v1
2
- oid sha256:797a471cad23917b9d403805f42f62679ccfae6a770a87408a15debfc23e4507
3
  size 14575
 
1
  version https://git-lfs.github.com/spec/v1
2
+ oid sha256:100aae511814c7709e33d2a7e5edec2138f73dddedc6401b8dd51e997c2a9d65
3
  size 14575
scheduler.pt CHANGED
@@ -1,3 +1,3 @@
1
  version https://git-lfs.github.com/spec/v1
2
- oid sha256:54f4f1b94cd49b030804b6e9c96d8faa07f56dfc10847cc9c67d1bcd0647a578
3
  size 627
 
1
  version https://git-lfs.github.com/spec/v1
2
+ oid sha256:25691735045c5749946dfe06471e1c5eb87438b273847d8c5a01062be5dc2158
3
  size 627
tokenizer.json CHANGED
@@ -1,6 +1,11 @@
1
  {
2
  "version": "1.0",
3
- "truncation": null,
 
 
 
 
 
4
  "padding": null,
5
  "added_tokens": [
6
  {
 
1
  {
2
  "version": "1.0",
3
+ "truncation": {
4
+ "direction": "Right",
5
+ "max_length": 1024,
6
+ "strategy": "LongestFirst",
7
+ "stride": 0
8
+ },
9
  "padding": null,
10
  "added_tokens": [
11
  {
trainer_state.json CHANGED
@@ -1,29 +1,29 @@
1
  {
2
- "best_metric": 1.4518036842346191,
3
- "best_model_checkpoint": "sean_test_out/checkpoint-42",
4
  "epoch": 1.0,
5
  "eval_steps": 500,
6
- "global_step": 42,
7
  "is_hyper_param_search": false,
8
  "is_local_process_zero": true,
9
  "is_world_process_zero": true,
10
  "log_history": [
11
  {
12
  "epoch": 1.0,
13
- "eval_accuracy": 0.353,
14
- "eval_loss": 1.4518036842346191,
15
- "eval_runtime": 58.1635,
16
- "eval_samples_per_second": 171.929,
17
- "eval_steps_per_second": 7.169,
18
- "step": 42
19
  }
20
  ],
21
  "logging_steps": 500,
22
- "max_steps": 42,
23
  "num_input_tokens_seen": 0,
24
  "num_train_epochs": 1,
25
  "save_steps": 500,
26
- "total_flos": 311359843123200.0,
27
  "train_batch_size": 24,
28
  "trial_name": null,
29
  "trial_params": null
 
1
  {
2
+ "best_metric": 1.1974165439605713,
3
+ "best_model_checkpoint": "sean_test_out/checkpoint-84",
4
  "epoch": 1.0,
5
  "eval_steps": 500,
6
+ "global_step": 84,
7
  "is_hyper_param_search": false,
8
  "is_local_process_zero": true,
9
  "is_world_process_zero": true,
10
  "log_history": [
11
  {
12
  "epoch": 1.0,
13
+ "eval_accuracy": 0.4583,
14
+ "eval_loss": 1.1974165439605713,
15
+ "eval_runtime": 60.9529,
16
+ "eval_samples_per_second": 164.061,
17
+ "eval_steps_per_second": 6.841,
18
+ "step": 84
19
  }
20
  ],
21
  "logging_steps": 500,
22
+ "max_steps": 84,
23
  "num_input_tokens_seen": 0,
24
  "num_train_epochs": 1,
25
  "save_steps": 500,
26
+ "total_flos": 650801682063360.0,
27
  "train_batch_size": 24,
28
  "trial_name": null,
29
  "trial_params": null
training_args.bin CHANGED
@@ -1,3 +1,3 @@
1
  version https://git-lfs.github.com/spec/v1
2
- oid sha256:e644ad3ad93c3cbe7dcd6662ac6c0e8467ec74e918b978d4cb18c413c6c07f11
3
  size 4539
 
1
  version https://git-lfs.github.com/spec/v1
2
+ oid sha256:93f80f261d91aaf59dea1c2a710f1dabe6385d9fee49b326957339a61484725e
3
  size 4539