manual push
Browse files- checkpoint-15000/config.json +55 -0
- checkpoint-15000/generation_config.json +6 -0
- checkpoint-15000/model.safetensors +3 -0
- checkpoint-15000/optimizer.pt +3 -0
- checkpoint-15000/rng_state.pth +3 -0
- checkpoint-15000/scheduler.pt +3 -0
- checkpoint-15000/special_tokens_map.json +16 -0
- checkpoint-15000/tokenizer.json +0 -0
- checkpoint-15000/tokenizer_config.json +50 -0
- checkpoint-15000/trainer_state.json +492 -0
- checkpoint-15000/training_args.bin +3 -0
- checkpoint-15500/config.json +55 -0
- checkpoint-15500/generation_config.json +6 -0
- checkpoint-15500/model.safetensors +3 -0
- checkpoint-15500/optimizer.pt +3 -0
- checkpoint-15500/rng_state.pth +3 -0
- checkpoint-15500/scheduler.pt +3 -0
- checkpoint-15500/special_tokens_map.json +16 -0
- checkpoint-15500/tokenizer.json +0 -0
- checkpoint-15500/tokenizer_config.json +50 -0
- checkpoint-15500/trainer_state.json +507 -0
- checkpoint-15500/training_args.bin +3 -0
checkpoint-15000/config.json
ADDED
|
@@ -0,0 +1,55 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
{
|
| 2 |
+
"_name_or_path": "EleutherAI/gpt-neo-125M",
|
| 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 |
+
"classifier_dropout": 0.1,
|
| 33 |
+
"embed_dropout": 0,
|
| 34 |
+
"eos_token_id": 50256,
|
| 35 |
+
"gradient_checkpointing": false,
|
| 36 |
+
"hidden_size": 768,
|
| 37 |
+
"initializer_range": 0.02,
|
| 38 |
+
"intermediate_size": null,
|
| 39 |
+
"layer_norm_epsilon": 1e-05,
|
| 40 |
+
"max_position_embeddings": 2048,
|
| 41 |
+
"model_type": "gpt_neo",
|
| 42 |
+
"num_heads": 12,
|
| 43 |
+
"num_layers": 12,
|
| 44 |
+
"resid_dropout": 0,
|
| 45 |
+
"summary_activation": null,
|
| 46 |
+
"summary_first_dropout": 0.1,
|
| 47 |
+
"summary_proj_to_labels": true,
|
| 48 |
+
"summary_type": "cls_index",
|
| 49 |
+
"summary_use_proj": true,
|
| 50 |
+
"torch_dtype": "float32",
|
| 51 |
+
"transformers_version": "4.48.3",
|
| 52 |
+
"use_cache": true,
|
| 53 |
+
"vocab_size": 257238,
|
| 54 |
+
"window_size": 256
|
| 55 |
+
}
|
checkpoint-15000/generation_config.json
ADDED
|
@@ -0,0 +1,6 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
{
|
| 2 |
+
"_from_model_config": true,
|
| 3 |
+
"bos_token_id": 50256,
|
| 4 |
+
"eos_token_id": 50256,
|
| 5 |
+
"transformers_version": "4.48.3"
|
| 6 |
+
}
|
checkpoint-15000/model.safetensors
ADDED
|
@@ -0,0 +1,3 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
version https://git-lfs.github.com/spec/v1
|
| 2 |
+
oid sha256:90ab723d394f28fa745a98ae07732ae15b69818bbe764089cae3d86323d23a1f
|
| 3 |
+
size 1136656968
|
checkpoint-15000/optimizer.pt
ADDED
|
@@ -0,0 +1,3 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
version https://git-lfs.github.com/spec/v1
|
| 2 |
+
oid sha256:446d8b45f226739d9c5599a85498f935910ae70abf9f130dcf55b0cc622a55bd
|
| 3 |
+
size 2273414202
|
checkpoint-15000/rng_state.pth
ADDED
|
@@ -0,0 +1,3 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
version https://git-lfs.github.com/spec/v1
|
| 2 |
+
oid sha256:77e2db468cf60d8fb2122f7d5380f04e37793ade84998ab16f53a921f9367a15
|
| 3 |
+
size 14244
|
checkpoint-15000/scheduler.pt
ADDED
|
@@ -0,0 +1,3 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
version https://git-lfs.github.com/spec/v1
|
| 2 |
+
oid sha256:2879a1fca2065a16e812c6277b3f3df3d31f3218056973e54d01475a39330694
|
| 3 |
+
size 1064
|
checkpoint-15000/special_tokens_map.json
ADDED
|
@@ -0,0 +1,16 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
{
|
| 2 |
+
"pad_token": {
|
| 3 |
+
"content": "[PAD]",
|
| 4 |
+
"lstrip": false,
|
| 5 |
+
"normalized": false,
|
| 6 |
+
"rstrip": false,
|
| 7 |
+
"single_word": false
|
| 8 |
+
},
|
| 9 |
+
"unk_token": {
|
| 10 |
+
"content": "[UNK]",
|
| 11 |
+
"lstrip": false,
|
| 12 |
+
"normalized": false,
|
| 13 |
+
"rstrip": false,
|
| 14 |
+
"single_word": false
|
| 15 |
+
}
|
| 16 |
+
}
|
checkpoint-15000/tokenizer.json
ADDED
|
The diff for this file is too large to render.
See raw diff
|
|
|
checkpoint-15000/tokenizer_config.json
ADDED
|
@@ -0,0 +1,50 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
{
|
| 2 |
+
"added_tokens_decoder": {
|
| 3 |
+
"0": {
|
| 4 |
+
"content": "[PAD]",
|
| 5 |
+
"lstrip": false,
|
| 6 |
+
"normalized": false,
|
| 7 |
+
"rstrip": false,
|
| 8 |
+
"single_word": false,
|
| 9 |
+
"special": true
|
| 10 |
+
},
|
| 11 |
+
"1": {
|
| 12 |
+
"content": "[UNK]",
|
| 13 |
+
"lstrip": false,
|
| 14 |
+
"normalized": false,
|
| 15 |
+
"rstrip": false,
|
| 16 |
+
"single_word": false,
|
| 17 |
+
"special": true
|
| 18 |
+
},
|
| 19 |
+
"2": {
|
| 20 |
+
"content": "[CLS]",
|
| 21 |
+
"lstrip": false,
|
| 22 |
+
"normalized": false,
|
| 23 |
+
"rstrip": false,
|
| 24 |
+
"single_word": false,
|
| 25 |
+
"special": true
|
| 26 |
+
},
|
| 27 |
+
"3": {
|
| 28 |
+
"content": "[SEP]",
|
| 29 |
+
"lstrip": false,
|
| 30 |
+
"normalized": false,
|
| 31 |
+
"rstrip": false,
|
| 32 |
+
"single_word": false,
|
| 33 |
+
"special": true
|
| 34 |
+
},
|
| 35 |
+
"4": {
|
| 36 |
+
"content": "[MASK]",
|
| 37 |
+
"lstrip": false,
|
| 38 |
+
"normalized": false,
|
| 39 |
+
"rstrip": false,
|
| 40 |
+
"single_word": false,
|
| 41 |
+
"special": true
|
| 42 |
+
}
|
| 43 |
+
},
|
| 44 |
+
"clean_up_tokenization_spaces": false,
|
| 45 |
+
"extra_special_tokens": {},
|
| 46 |
+
"model_max_length": 1000000000000000019884624838656,
|
| 47 |
+
"pad_token": "[PAD]",
|
| 48 |
+
"tokenizer_class": "PreTrainedTokenizerFast",
|
| 49 |
+
"unk_token": "[UNK]"
|
| 50 |
+
}
|
checkpoint-15000/trainer_state.json
ADDED
|
@@ -0,0 +1,492 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
{
|
| 2 |
+
"best_metric": 0.944929301738739,
|
| 3 |
+
"best_model_checkpoint": "./dhivehi_gpt_neo/checkpoint-15000",
|
| 4 |
+
"epoch": 0.524897644959233,
|
| 5 |
+
"eval_steps": 500,
|
| 6 |
+
"global_step": 15000,
|
| 7 |
+
"is_hyper_param_search": false,
|
| 8 |
+
"is_local_process_zero": true,
|
| 9 |
+
"is_world_process_zero": true,
|
| 10 |
+
"log_history": [
|
| 11 |
+
{
|
| 12 |
+
"epoch": 0.017496588165307764,
|
| 13 |
+
"grad_norm": 0.6128062605857849,
|
| 14 |
+
"learning_rate": 9.960000000000001e-05,
|
| 15 |
+
"loss": 1.7949,
|
| 16 |
+
"step": 500
|
| 17 |
+
},
|
| 18 |
+
{
|
| 19 |
+
"epoch": 0.017496588165307764,
|
| 20 |
+
"eval_loss": 1.5562938451766968,
|
| 21 |
+
"eval_runtime": 1838.2839,
|
| 22 |
+
"eval_samples_per_second": 27.637,
|
| 23 |
+
"eval_steps_per_second": 3.455,
|
| 24 |
+
"step": 500
|
| 25 |
+
},
|
| 26 |
+
{
|
| 27 |
+
"epoch": 0.03499317633061553,
|
| 28 |
+
"grad_norm": 1.0217710733413696,
|
| 29 |
+
"learning_rate": 9.94157055531438e-05,
|
| 30 |
+
"loss": 1.4832,
|
| 31 |
+
"step": 1000
|
| 32 |
+
},
|
| 33 |
+
{
|
| 34 |
+
"epoch": 0.03499317633061553,
|
| 35 |
+
"eval_loss": 1.4455698728561401,
|
| 36 |
+
"eval_runtime": 1838.2342,
|
| 37 |
+
"eval_samples_per_second": 27.637,
|
| 38 |
+
"eval_steps_per_second": 3.455,
|
| 39 |
+
"step": 1000
|
| 40 |
+
},
|
| 41 |
+
{
|
| 42 |
+
"epoch": 0.0524897644959233,
|
| 43 |
+
"grad_norm": 0.7153336405754089,
|
| 44 |
+
"learning_rate": 9.882906454224403e-05,
|
| 45 |
+
"loss": 1.4091,
|
| 46 |
+
"step": 1500
|
| 47 |
+
},
|
| 48 |
+
{
|
| 49 |
+
"epoch": 0.0524897644959233,
|
| 50 |
+
"eval_loss": 1.3654062747955322,
|
| 51 |
+
"eval_runtime": 1838.7353,
|
| 52 |
+
"eval_samples_per_second": 27.63,
|
| 53 |
+
"eval_steps_per_second": 3.454,
|
| 54 |
+
"step": 1500
|
| 55 |
+
},
|
| 56 |
+
{
|
| 57 |
+
"epoch": 0.06998635266123106,
|
| 58 |
+
"grad_norm": 2.103127956390381,
|
| 59 |
+
"learning_rate": 9.824242353134423e-05,
|
| 60 |
+
"loss": 1.3518,
|
| 61 |
+
"step": 2000
|
| 62 |
+
},
|
| 63 |
+
{
|
| 64 |
+
"epoch": 0.06998635266123106,
|
| 65 |
+
"eval_loss": 1.303843379020691,
|
| 66 |
+
"eval_runtime": 1838.4453,
|
| 67 |
+
"eval_samples_per_second": 27.634,
|
| 68 |
+
"eval_steps_per_second": 3.455,
|
| 69 |
+
"step": 2000
|
| 70 |
+
},
|
| 71 |
+
{
|
| 72 |
+
"epoch": 0.08748294082653882,
|
| 73 |
+
"grad_norm": 0.7201130986213684,
|
| 74 |
+
"learning_rate": 9.765578252044444e-05,
|
| 75 |
+
"loss": 1.3079,
|
| 76 |
+
"step": 2500
|
| 77 |
+
},
|
| 78 |
+
{
|
| 79 |
+
"epoch": 0.08748294082653882,
|
| 80 |
+
"eval_loss": 1.255767583847046,
|
| 81 |
+
"eval_runtime": 1839.5413,
|
| 82 |
+
"eval_samples_per_second": 27.618,
|
| 83 |
+
"eval_steps_per_second": 3.452,
|
| 84 |
+
"step": 2500
|
| 85 |
+
},
|
| 86 |
+
{
|
| 87 |
+
"epoch": 0.1049795289918466,
|
| 88 |
+
"grad_norm": 1.7693496942520142,
|
| 89 |
+
"learning_rate": 9.706914150954466e-05,
|
| 90 |
+
"loss": 1.2781,
|
| 91 |
+
"step": 3000
|
| 92 |
+
},
|
| 93 |
+
{
|
| 94 |
+
"epoch": 0.1049795289918466,
|
| 95 |
+
"eval_loss": 1.222457766532898,
|
| 96 |
+
"eval_runtime": 1838.7539,
|
| 97 |
+
"eval_samples_per_second": 27.63,
|
| 98 |
+
"eval_steps_per_second": 3.454,
|
| 99 |
+
"step": 3000
|
| 100 |
+
},
|
| 101 |
+
{
|
| 102 |
+
"epoch": 0.12247611715715435,
|
| 103 |
+
"grad_norm": 1.739169716835022,
|
| 104 |
+
"learning_rate": 9.648250049864486e-05,
|
| 105 |
+
"loss": 1.1931,
|
| 106 |
+
"step": 3500
|
| 107 |
+
},
|
| 108 |
+
{
|
| 109 |
+
"epoch": 0.12247611715715435,
|
| 110 |
+
"eval_loss": 1.1909434795379639,
|
| 111 |
+
"eval_runtime": 1838.4303,
|
| 112 |
+
"eval_samples_per_second": 27.634,
|
| 113 |
+
"eval_steps_per_second": 3.455,
|
| 114 |
+
"step": 3500
|
| 115 |
+
},
|
| 116 |
+
{
|
| 117 |
+
"epoch": 0.13997270532246212,
|
| 118 |
+
"grad_norm": 1.2461516857147217,
|
| 119 |
+
"learning_rate": 9.589585948774507e-05,
|
| 120 |
+
"loss": 1.1771,
|
| 121 |
+
"step": 4000
|
| 122 |
+
},
|
| 123 |
+
{
|
| 124 |
+
"epoch": 0.13997270532246212,
|
| 125 |
+
"eval_loss": 1.1642112731933594,
|
| 126 |
+
"eval_runtime": 1839.7956,
|
| 127 |
+
"eval_samples_per_second": 27.614,
|
| 128 |
+
"eval_steps_per_second": 3.452,
|
| 129 |
+
"step": 4000
|
| 130 |
+
},
|
| 131 |
+
{
|
| 132 |
+
"epoch": 0.1574692934877699,
|
| 133 |
+
"grad_norm": 1.5028084516525269,
|
| 134 |
+
"learning_rate": 9.530921847684529e-05,
|
| 135 |
+
"loss": 1.1496,
|
| 136 |
+
"step": 4500
|
| 137 |
+
},
|
| 138 |
+
{
|
| 139 |
+
"epoch": 0.1574692934877699,
|
| 140 |
+
"eval_loss": 1.141344428062439,
|
| 141 |
+
"eval_runtime": 1839.6492,
|
| 142 |
+
"eval_samples_per_second": 27.616,
|
| 143 |
+
"eval_steps_per_second": 3.452,
|
| 144 |
+
"step": 4500
|
| 145 |
+
},
|
| 146 |
+
{
|
| 147 |
+
"epoch": 0.17496588165307764,
|
| 148 |
+
"grad_norm": 1.0963733196258545,
|
| 149 |
+
"learning_rate": 9.47225774659455e-05,
|
| 150 |
+
"loss": 1.1325,
|
| 151 |
+
"step": 5000
|
| 152 |
+
},
|
| 153 |
+
{
|
| 154 |
+
"epoch": 0.17496588165307764,
|
| 155 |
+
"eval_loss": 1.12198007106781,
|
| 156 |
+
"eval_runtime": 1840.0292,
|
| 157 |
+
"eval_samples_per_second": 27.61,
|
| 158 |
+
"eval_steps_per_second": 3.452,
|
| 159 |
+
"step": 5000
|
| 160 |
+
},
|
| 161 |
+
{
|
| 162 |
+
"epoch": 0.1924624698183854,
|
| 163 |
+
"grad_norm": 0.9784343838691711,
|
| 164 |
+
"learning_rate": 9.41359364550457e-05,
|
| 165 |
+
"loss": 1.1301,
|
| 166 |
+
"step": 5500
|
| 167 |
+
},
|
| 168 |
+
{
|
| 169 |
+
"epoch": 0.1924624698183854,
|
| 170 |
+
"eval_loss": 1.1025567054748535,
|
| 171 |
+
"eval_runtime": 1840.7493,
|
| 172 |
+
"eval_samples_per_second": 27.6,
|
| 173 |
+
"eval_steps_per_second": 3.45,
|
| 174 |
+
"step": 5500
|
| 175 |
+
},
|
| 176 |
+
{
|
| 177 |
+
"epoch": 0.2099590579836932,
|
| 178 |
+
"grad_norm": 1.8819756507873535,
|
| 179 |
+
"learning_rate": 9.354929544414591e-05,
|
| 180 |
+
"loss": 1.1118,
|
| 181 |
+
"step": 6000
|
| 182 |
+
},
|
| 183 |
+
{
|
| 184 |
+
"epoch": 0.2099590579836932,
|
| 185 |
+
"eval_loss": 1.0877833366394043,
|
| 186 |
+
"eval_runtime": 1840.7767,
|
| 187 |
+
"eval_samples_per_second": 27.599,
|
| 188 |
+
"eval_steps_per_second": 3.45,
|
| 189 |
+
"step": 6000
|
| 190 |
+
},
|
| 191 |
+
{
|
| 192 |
+
"epoch": 0.22745564614900093,
|
| 193 |
+
"grad_norm": 1.390010118484497,
|
| 194 |
+
"learning_rate": 9.296382771526793e-05,
|
| 195 |
+
"loss": 1.1046,
|
| 196 |
+
"step": 6500
|
| 197 |
+
},
|
| 198 |
+
{
|
| 199 |
+
"epoch": 0.22745564614900093,
|
| 200 |
+
"eval_loss": 1.0721691846847534,
|
| 201 |
+
"eval_runtime": 1840.2034,
|
| 202 |
+
"eval_samples_per_second": 27.608,
|
| 203 |
+
"eval_steps_per_second": 3.451,
|
| 204 |
+
"step": 6500
|
| 205 |
+
},
|
| 206 |
+
{
|
| 207 |
+
"epoch": 0.2449522343143087,
|
| 208 |
+
"grad_norm": 1.1395955085754395,
|
| 209 |
+
"learning_rate": 9.237718670436814e-05,
|
| 210 |
+
"loss": 1.071,
|
| 211 |
+
"step": 7000
|
| 212 |
+
},
|
| 213 |
+
{
|
| 214 |
+
"epoch": 0.2449522343143087,
|
| 215 |
+
"eval_loss": 1.0588288307189941,
|
| 216 |
+
"eval_runtime": 1841.3244,
|
| 217 |
+
"eval_samples_per_second": 27.591,
|
| 218 |
+
"eval_steps_per_second": 3.449,
|
| 219 |
+
"step": 7000
|
| 220 |
+
},
|
| 221 |
+
{
|
| 222 |
+
"epoch": 0.2624488224796165,
|
| 223 |
+
"grad_norm": 1.557716965675354,
|
| 224 |
+
"learning_rate": 9.179054569346835e-05,
|
| 225 |
+
"loss": 1.0578,
|
| 226 |
+
"step": 7500
|
| 227 |
+
},
|
| 228 |
+
{
|
| 229 |
+
"epoch": 0.2624488224796165,
|
| 230 |
+
"eval_loss": 1.0490485429763794,
|
| 231 |
+
"eval_runtime": 1841.5042,
|
| 232 |
+
"eval_samples_per_second": 27.588,
|
| 233 |
+
"eval_steps_per_second": 3.449,
|
| 234 |
+
"step": 7500
|
| 235 |
+
},
|
| 236 |
+
{
|
| 237 |
+
"epoch": 0.27994541064492423,
|
| 238 |
+
"grad_norm": 1.6820220947265625,
|
| 239 |
+
"learning_rate": 9.120390468256855e-05,
|
| 240 |
+
"loss": 1.0293,
|
| 241 |
+
"step": 8000
|
| 242 |
+
},
|
| 243 |
+
{
|
| 244 |
+
"epoch": 0.27994541064492423,
|
| 245 |
+
"eval_loss": 1.0374187231063843,
|
| 246 |
+
"eval_runtime": 1841.7893,
|
| 247 |
+
"eval_samples_per_second": 27.584,
|
| 248 |
+
"eval_steps_per_second": 3.448,
|
| 249 |
+
"step": 8000
|
| 250 |
+
},
|
| 251 |
+
{
|
| 252 |
+
"epoch": 0.297441998810232,
|
| 253 |
+
"grad_norm": 1.325988531112671,
|
| 254 |
+
"learning_rate": 9.061726367166876e-05,
|
| 255 |
+
"loss": 1.0229,
|
| 256 |
+
"step": 8500
|
| 257 |
+
},
|
| 258 |
+
{
|
| 259 |
+
"epoch": 0.297441998810232,
|
| 260 |
+
"eval_loss": 1.0274155139923096,
|
| 261 |
+
"eval_runtime": 1842.3456,
|
| 262 |
+
"eval_samples_per_second": 27.576,
|
| 263 |
+
"eval_steps_per_second": 3.447,
|
| 264 |
+
"step": 8500
|
| 265 |
+
},
|
| 266 |
+
{
|
| 267 |
+
"epoch": 0.3149385869755398,
|
| 268 |
+
"grad_norm": 1.0609859228134155,
|
| 269 |
+
"learning_rate": 9.003179594279078e-05,
|
| 270 |
+
"loss": 1.0345,
|
| 271 |
+
"step": 9000
|
| 272 |
+
},
|
| 273 |
+
{
|
| 274 |
+
"epoch": 0.3149385869755398,
|
| 275 |
+
"eval_loss": 1.0168765783309937,
|
| 276 |
+
"eval_runtime": 1843.5677,
|
| 277 |
+
"eval_samples_per_second": 27.557,
|
| 278 |
+
"eval_steps_per_second": 3.445,
|
| 279 |
+
"step": 9000
|
| 280 |
+
},
|
| 281 |
+
{
|
| 282 |
+
"epoch": 0.3324351751408475,
|
| 283 |
+
"grad_norm": 0.6549689173698425,
|
| 284 |
+
"learning_rate": 8.944515493189099e-05,
|
| 285 |
+
"loss": 1.0255,
|
| 286 |
+
"step": 9500
|
| 287 |
+
},
|
| 288 |
+
{
|
| 289 |
+
"epoch": 0.3324351751408475,
|
| 290 |
+
"eval_loss": 1.007939338684082,
|
| 291 |
+
"eval_runtime": 1841.3868,
|
| 292 |
+
"eval_samples_per_second": 27.59,
|
| 293 |
+
"eval_steps_per_second": 3.449,
|
| 294 |
+
"step": 9500
|
| 295 |
+
},
|
| 296 |
+
{
|
| 297 |
+
"epoch": 0.3499317633061553,
|
| 298 |
+
"grad_norm": 1.5119966268539429,
|
| 299 |
+
"learning_rate": 8.88585139209912e-05,
|
| 300 |
+
"loss": 1.0412,
|
| 301 |
+
"step": 10000
|
| 302 |
+
},
|
| 303 |
+
{
|
| 304 |
+
"epoch": 0.3499317633061553,
|
| 305 |
+
"eval_loss": 1.0019127130508423,
|
| 306 |
+
"eval_runtime": 1841.6495,
|
| 307 |
+
"eval_samples_per_second": 27.586,
|
| 308 |
+
"eval_steps_per_second": 3.449,
|
| 309 |
+
"step": 10000
|
| 310 |
+
},
|
| 311 |
+
{
|
| 312 |
+
"epoch": 0.3674283514714631,
|
| 313 |
+
"grad_norm": 1.2304165363311768,
|
| 314 |
+
"learning_rate": 8.82718729100914e-05,
|
| 315 |
+
"loss": 0.9972,
|
| 316 |
+
"step": 10500
|
| 317 |
+
},
|
| 318 |
+
{
|
| 319 |
+
"epoch": 0.3674283514714631,
|
| 320 |
+
"eval_loss": 0.9952266216278076,
|
| 321 |
+
"eval_runtime": 1841.7158,
|
| 322 |
+
"eval_samples_per_second": 27.585,
|
| 323 |
+
"eval_steps_per_second": 3.448,
|
| 324 |
+
"step": 10500
|
| 325 |
+
},
|
| 326 |
+
{
|
| 327 |
+
"epoch": 0.3849249396367708,
|
| 328 |
+
"grad_norm": 1.0097540616989136,
|
| 329 |
+
"learning_rate": 8.768640518121341e-05,
|
| 330 |
+
"loss": 0.9861,
|
| 331 |
+
"step": 11000
|
| 332 |
+
},
|
| 333 |
+
{
|
| 334 |
+
"epoch": 0.3849249396367708,
|
| 335 |
+
"eval_loss": 0.9871318936347961,
|
| 336 |
+
"eval_runtime": 1841.3777,
|
| 337 |
+
"eval_samples_per_second": 27.59,
|
| 338 |
+
"eval_steps_per_second": 3.449,
|
| 339 |
+
"step": 11000
|
| 340 |
+
},
|
| 341 |
+
{
|
| 342 |
+
"epoch": 0.40242152780207857,
|
| 343 |
+
"grad_norm": 1.4742999076843262,
|
| 344 |
+
"learning_rate": 8.709976417031363e-05,
|
| 345 |
+
"loss": 0.979,
|
| 346 |
+
"step": 11500
|
| 347 |
+
},
|
| 348 |
+
{
|
| 349 |
+
"epoch": 0.40242152780207857,
|
| 350 |
+
"eval_loss": 0.9808717370033264,
|
| 351 |
+
"eval_runtime": 1841.6224,
|
| 352 |
+
"eval_samples_per_second": 27.587,
|
| 353 |
+
"eval_steps_per_second": 3.449,
|
| 354 |
+
"step": 11500
|
| 355 |
+
},
|
| 356 |
+
{
|
| 357 |
+
"epoch": 0.4199181159673864,
|
| 358 |
+
"grad_norm": 1.0650264024734497,
|
| 359 |
+
"learning_rate": 8.651312315941384e-05,
|
| 360 |
+
"loss": 0.9968,
|
| 361 |
+
"step": 12000
|
| 362 |
+
},
|
| 363 |
+
{
|
| 364 |
+
"epoch": 0.4199181159673864,
|
| 365 |
+
"eval_loss": 0.9748508930206299,
|
| 366 |
+
"eval_runtime": 1841.2412,
|
| 367 |
+
"eval_samples_per_second": 27.592,
|
| 368 |
+
"eval_steps_per_second": 3.449,
|
| 369 |
+
"step": 12000
|
| 370 |
+
},
|
| 371 |
+
{
|
| 372 |
+
"epoch": 0.4374147041326941,
|
| 373 |
+
"grad_norm": 0.9701998233795166,
|
| 374 |
+
"learning_rate": 8.592648214851404e-05,
|
| 375 |
+
"loss": 0.978,
|
| 376 |
+
"step": 12500
|
| 377 |
+
},
|
| 378 |
+
{
|
| 379 |
+
"epoch": 0.4374147041326941,
|
| 380 |
+
"eval_loss": 0.9688844680786133,
|
| 381 |
+
"eval_runtime": 1841.3182,
|
| 382 |
+
"eval_samples_per_second": 27.591,
|
| 383 |
+
"eval_steps_per_second": 3.449,
|
| 384 |
+
"step": 12500
|
| 385 |
+
},
|
| 386 |
+
{
|
| 387 |
+
"epoch": 0.45491129229800187,
|
| 388 |
+
"grad_norm": 1.7642956972122192,
|
| 389 |
+
"learning_rate": 8.533984113761425e-05,
|
| 390 |
+
"loss": 0.9682,
|
| 391 |
+
"step": 13000
|
| 392 |
+
},
|
| 393 |
+
{
|
| 394 |
+
"epoch": 0.45491129229800187,
|
| 395 |
+
"eval_loss": 0.9631832838058472,
|
| 396 |
+
"eval_runtime": 1842.1886,
|
| 397 |
+
"eval_samples_per_second": 27.578,
|
| 398 |
+
"eval_steps_per_second": 3.448,
|
| 399 |
+
"step": 13000
|
| 400 |
+
},
|
| 401 |
+
{
|
| 402 |
+
"epoch": 0.47240788046330967,
|
| 403 |
+
"grad_norm": 1.9703904390335083,
|
| 404 |
+
"learning_rate": 8.475320012671446e-05,
|
| 405 |
+
"loss": 0.9594,
|
| 406 |
+
"step": 13500
|
| 407 |
+
},
|
| 408 |
+
{
|
| 409 |
+
"epoch": 0.47240788046330967,
|
| 410 |
+
"eval_loss": 0.9580743908882141,
|
| 411 |
+
"eval_runtime": 1841.8156,
|
| 412 |
+
"eval_samples_per_second": 27.584,
|
| 413 |
+
"eval_steps_per_second": 3.448,
|
| 414 |
+
"step": 13500
|
| 415 |
+
},
|
| 416 |
+
{
|
| 417 |
+
"epoch": 0.4899044686286174,
|
| 418 |
+
"grad_norm": 1.6822909116744995,
|
| 419 |
+
"learning_rate": 8.416655911581468e-05,
|
| 420 |
+
"loss": 0.9565,
|
| 421 |
+
"step": 14000
|
| 422 |
+
},
|
| 423 |
+
{
|
| 424 |
+
"epoch": 0.4899044686286174,
|
| 425 |
+
"eval_loss": 0.952092707157135,
|
| 426 |
+
"eval_runtime": 1841.6575,
|
| 427 |
+
"eval_samples_per_second": 27.586,
|
| 428 |
+
"eval_steps_per_second": 3.449,
|
| 429 |
+
"step": 14000
|
| 430 |
+
},
|
| 431 |
+
{
|
| 432 |
+
"epoch": 0.5074010567939252,
|
| 433 |
+
"grad_norm": 1.2567087411880493,
|
| 434 |
+
"learning_rate": 8.357991810491488e-05,
|
| 435 |
+
"loss": 0.9328,
|
| 436 |
+
"step": 14500
|
| 437 |
+
},
|
| 438 |
+
{
|
| 439 |
+
"epoch": 0.5074010567939252,
|
| 440 |
+
"eval_loss": 0.9498482346534729,
|
| 441 |
+
"eval_runtime": 1841.8538,
|
| 442 |
+
"eval_samples_per_second": 27.583,
|
| 443 |
+
"eval_steps_per_second": 3.448,
|
| 444 |
+
"step": 14500
|
| 445 |
+
},
|
| 446 |
+
{
|
| 447 |
+
"epoch": 0.524897644959233,
|
| 448 |
+
"grad_norm": 1.645724892616272,
|
| 449 |
+
"learning_rate": 8.299445037603689e-05,
|
| 450 |
+
"loss": 0.9612,
|
| 451 |
+
"step": 15000
|
| 452 |
+
},
|
| 453 |
+
{
|
| 454 |
+
"epoch": 0.524897644959233,
|
| 455 |
+
"eval_loss": 0.944929301738739,
|
| 456 |
+
"eval_runtime": 1841.6649,
|
| 457 |
+
"eval_samples_per_second": 27.586,
|
| 458 |
+
"eval_steps_per_second": 3.449,
|
| 459 |
+
"step": 15000
|
| 460 |
+
}
|
| 461 |
+
],
|
| 462 |
+
"logging_steps": 500,
|
| 463 |
+
"max_steps": 85731,
|
| 464 |
+
"num_input_tokens_seen": 0,
|
| 465 |
+
"num_train_epochs": 3,
|
| 466 |
+
"save_steps": 500,
|
| 467 |
+
"stateful_callbacks": {
|
| 468 |
+
"EarlyStoppingCallback": {
|
| 469 |
+
"args": {
|
| 470 |
+
"early_stopping_patience": 3,
|
| 471 |
+
"early_stopping_threshold": 0.0
|
| 472 |
+
},
|
| 473 |
+
"attributes": {
|
| 474 |
+
"early_stopping_patience_counter": 0
|
| 475 |
+
}
|
| 476 |
+
},
|
| 477 |
+
"TrainerControl": {
|
| 478 |
+
"args": {
|
| 479 |
+
"should_epoch_stop": false,
|
| 480 |
+
"should_evaluate": false,
|
| 481 |
+
"should_log": false,
|
| 482 |
+
"should_save": true,
|
| 483 |
+
"should_training_stop": false
|
| 484 |
+
},
|
| 485 |
+
"attributes": {}
|
| 486 |
+
}
|
| 487 |
+
},
|
| 488 |
+
"total_flos": 6.268970336256e+16,
|
| 489 |
+
"train_batch_size": 4,
|
| 490 |
+
"trial_name": null,
|
| 491 |
+
"trial_params": null
|
| 492 |
+
}
|
checkpoint-15000/training_args.bin
ADDED
|
@@ -0,0 +1,3 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
version https://git-lfs.github.com/spec/v1
|
| 2 |
+
oid sha256:d1174044396b4e2f4cc6381a71fd810c6b5d8fb1d414d61315c45be1a1a8533a
|
| 3 |
+
size 5304
|
checkpoint-15500/config.json
ADDED
|
@@ -0,0 +1,55 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
{
|
| 2 |
+
"_name_or_path": "EleutherAI/gpt-neo-125M",
|
| 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 |
+
"classifier_dropout": 0.1,
|
| 33 |
+
"embed_dropout": 0,
|
| 34 |
+
"eos_token_id": 50256,
|
| 35 |
+
"gradient_checkpointing": false,
|
| 36 |
+
"hidden_size": 768,
|
| 37 |
+
"initializer_range": 0.02,
|
| 38 |
+
"intermediate_size": null,
|
| 39 |
+
"layer_norm_epsilon": 1e-05,
|
| 40 |
+
"max_position_embeddings": 2048,
|
| 41 |
+
"model_type": "gpt_neo",
|
| 42 |
+
"num_heads": 12,
|
| 43 |
+
"num_layers": 12,
|
| 44 |
+
"resid_dropout": 0,
|
| 45 |
+
"summary_activation": null,
|
| 46 |
+
"summary_first_dropout": 0.1,
|
| 47 |
+
"summary_proj_to_labels": true,
|
| 48 |
+
"summary_type": "cls_index",
|
| 49 |
+
"summary_use_proj": true,
|
| 50 |
+
"torch_dtype": "float32",
|
| 51 |
+
"transformers_version": "4.48.3",
|
| 52 |
+
"use_cache": true,
|
| 53 |
+
"vocab_size": 257238,
|
| 54 |
+
"window_size": 256
|
| 55 |
+
}
|
checkpoint-15500/generation_config.json
ADDED
|
@@ -0,0 +1,6 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
{
|
| 2 |
+
"_from_model_config": true,
|
| 3 |
+
"bos_token_id": 50256,
|
| 4 |
+
"eos_token_id": 50256,
|
| 5 |
+
"transformers_version": "4.48.3"
|
| 6 |
+
}
|
checkpoint-15500/model.safetensors
ADDED
|
@@ -0,0 +1,3 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
version https://git-lfs.github.com/spec/v1
|
| 2 |
+
oid sha256:76b001cfe4a152d5dcfa55e18d45cc53de7ad8dbf7e63a85604725daf7c2e418
|
| 3 |
+
size 1136656968
|
checkpoint-15500/optimizer.pt
ADDED
|
@@ -0,0 +1,3 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
version https://git-lfs.github.com/spec/v1
|
| 2 |
+
oid sha256:beffaf7698d6c120d20764d20a5502c6486322b9b685e5247593d6eaaa6ddc77
|
| 3 |
+
size 2273414202
|
checkpoint-15500/rng_state.pth
ADDED
|
@@ -0,0 +1,3 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
version https://git-lfs.github.com/spec/v1
|
| 2 |
+
oid sha256:27a79239f98d586c6d293becfe4724cb48ad892f743d1e770886cde54b3333d6
|
| 3 |
+
size 14244
|
checkpoint-15500/scheduler.pt
ADDED
|
@@ -0,0 +1,3 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
version https://git-lfs.github.com/spec/v1
|
| 2 |
+
oid sha256:3f3f55fb9dfd5c7efd324389470b763dda292d117c6f20c92a51034a7779553e
|
| 3 |
+
size 1064
|
checkpoint-15500/special_tokens_map.json
ADDED
|
@@ -0,0 +1,16 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
{
|
| 2 |
+
"pad_token": {
|
| 3 |
+
"content": "[PAD]",
|
| 4 |
+
"lstrip": false,
|
| 5 |
+
"normalized": false,
|
| 6 |
+
"rstrip": false,
|
| 7 |
+
"single_word": false
|
| 8 |
+
},
|
| 9 |
+
"unk_token": {
|
| 10 |
+
"content": "[UNK]",
|
| 11 |
+
"lstrip": false,
|
| 12 |
+
"normalized": false,
|
| 13 |
+
"rstrip": false,
|
| 14 |
+
"single_word": false
|
| 15 |
+
}
|
| 16 |
+
}
|
checkpoint-15500/tokenizer.json
ADDED
|
The diff for this file is too large to render.
See raw diff
|
|
|
checkpoint-15500/tokenizer_config.json
ADDED
|
@@ -0,0 +1,50 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
{
|
| 2 |
+
"added_tokens_decoder": {
|
| 3 |
+
"0": {
|
| 4 |
+
"content": "[PAD]",
|
| 5 |
+
"lstrip": false,
|
| 6 |
+
"normalized": false,
|
| 7 |
+
"rstrip": false,
|
| 8 |
+
"single_word": false,
|
| 9 |
+
"special": true
|
| 10 |
+
},
|
| 11 |
+
"1": {
|
| 12 |
+
"content": "[UNK]",
|
| 13 |
+
"lstrip": false,
|
| 14 |
+
"normalized": false,
|
| 15 |
+
"rstrip": false,
|
| 16 |
+
"single_word": false,
|
| 17 |
+
"special": true
|
| 18 |
+
},
|
| 19 |
+
"2": {
|
| 20 |
+
"content": "[CLS]",
|
| 21 |
+
"lstrip": false,
|
| 22 |
+
"normalized": false,
|
| 23 |
+
"rstrip": false,
|
| 24 |
+
"single_word": false,
|
| 25 |
+
"special": true
|
| 26 |
+
},
|
| 27 |
+
"3": {
|
| 28 |
+
"content": "[SEP]",
|
| 29 |
+
"lstrip": false,
|
| 30 |
+
"normalized": false,
|
| 31 |
+
"rstrip": false,
|
| 32 |
+
"single_word": false,
|
| 33 |
+
"special": true
|
| 34 |
+
},
|
| 35 |
+
"4": {
|
| 36 |
+
"content": "[MASK]",
|
| 37 |
+
"lstrip": false,
|
| 38 |
+
"normalized": false,
|
| 39 |
+
"rstrip": false,
|
| 40 |
+
"single_word": false,
|
| 41 |
+
"special": true
|
| 42 |
+
}
|
| 43 |
+
},
|
| 44 |
+
"clean_up_tokenization_spaces": false,
|
| 45 |
+
"extra_special_tokens": {},
|
| 46 |
+
"model_max_length": 1000000000000000019884624838656,
|
| 47 |
+
"pad_token": "[PAD]",
|
| 48 |
+
"tokenizer_class": "PreTrainedTokenizerFast",
|
| 49 |
+
"unk_token": "[UNK]"
|
| 50 |
+
}
|
checkpoint-15500/trainer_state.json
ADDED
|
@@ -0,0 +1,507 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
{
|
| 2 |
+
"best_metric": 0.9399999380111694,
|
| 3 |
+
"best_model_checkpoint": "./dhivehi_gpt_neo/checkpoint-15500",
|
| 4 |
+
"epoch": 0.5423942331245407,
|
| 5 |
+
"eval_steps": 500,
|
| 6 |
+
"global_step": 15500,
|
| 7 |
+
"is_hyper_param_search": false,
|
| 8 |
+
"is_local_process_zero": true,
|
| 9 |
+
"is_world_process_zero": true,
|
| 10 |
+
"log_history": [
|
| 11 |
+
{
|
| 12 |
+
"epoch": 0.017496588165307764,
|
| 13 |
+
"grad_norm": 0.6128062605857849,
|
| 14 |
+
"learning_rate": 9.960000000000001e-05,
|
| 15 |
+
"loss": 1.7949,
|
| 16 |
+
"step": 500
|
| 17 |
+
},
|
| 18 |
+
{
|
| 19 |
+
"epoch": 0.017496588165307764,
|
| 20 |
+
"eval_loss": 1.5562938451766968,
|
| 21 |
+
"eval_runtime": 1838.2839,
|
| 22 |
+
"eval_samples_per_second": 27.637,
|
| 23 |
+
"eval_steps_per_second": 3.455,
|
| 24 |
+
"step": 500
|
| 25 |
+
},
|
| 26 |
+
{
|
| 27 |
+
"epoch": 0.03499317633061553,
|
| 28 |
+
"grad_norm": 1.0217710733413696,
|
| 29 |
+
"learning_rate": 9.94157055531438e-05,
|
| 30 |
+
"loss": 1.4832,
|
| 31 |
+
"step": 1000
|
| 32 |
+
},
|
| 33 |
+
{
|
| 34 |
+
"epoch": 0.03499317633061553,
|
| 35 |
+
"eval_loss": 1.4455698728561401,
|
| 36 |
+
"eval_runtime": 1838.2342,
|
| 37 |
+
"eval_samples_per_second": 27.637,
|
| 38 |
+
"eval_steps_per_second": 3.455,
|
| 39 |
+
"step": 1000
|
| 40 |
+
},
|
| 41 |
+
{
|
| 42 |
+
"epoch": 0.0524897644959233,
|
| 43 |
+
"grad_norm": 0.7153336405754089,
|
| 44 |
+
"learning_rate": 9.882906454224403e-05,
|
| 45 |
+
"loss": 1.4091,
|
| 46 |
+
"step": 1500
|
| 47 |
+
},
|
| 48 |
+
{
|
| 49 |
+
"epoch": 0.0524897644959233,
|
| 50 |
+
"eval_loss": 1.3654062747955322,
|
| 51 |
+
"eval_runtime": 1838.7353,
|
| 52 |
+
"eval_samples_per_second": 27.63,
|
| 53 |
+
"eval_steps_per_second": 3.454,
|
| 54 |
+
"step": 1500
|
| 55 |
+
},
|
| 56 |
+
{
|
| 57 |
+
"epoch": 0.06998635266123106,
|
| 58 |
+
"grad_norm": 2.103127956390381,
|
| 59 |
+
"learning_rate": 9.824242353134423e-05,
|
| 60 |
+
"loss": 1.3518,
|
| 61 |
+
"step": 2000
|
| 62 |
+
},
|
| 63 |
+
{
|
| 64 |
+
"epoch": 0.06998635266123106,
|
| 65 |
+
"eval_loss": 1.303843379020691,
|
| 66 |
+
"eval_runtime": 1838.4453,
|
| 67 |
+
"eval_samples_per_second": 27.634,
|
| 68 |
+
"eval_steps_per_second": 3.455,
|
| 69 |
+
"step": 2000
|
| 70 |
+
},
|
| 71 |
+
{
|
| 72 |
+
"epoch": 0.08748294082653882,
|
| 73 |
+
"grad_norm": 0.7201130986213684,
|
| 74 |
+
"learning_rate": 9.765578252044444e-05,
|
| 75 |
+
"loss": 1.3079,
|
| 76 |
+
"step": 2500
|
| 77 |
+
},
|
| 78 |
+
{
|
| 79 |
+
"epoch": 0.08748294082653882,
|
| 80 |
+
"eval_loss": 1.255767583847046,
|
| 81 |
+
"eval_runtime": 1839.5413,
|
| 82 |
+
"eval_samples_per_second": 27.618,
|
| 83 |
+
"eval_steps_per_second": 3.452,
|
| 84 |
+
"step": 2500
|
| 85 |
+
},
|
| 86 |
+
{
|
| 87 |
+
"epoch": 0.1049795289918466,
|
| 88 |
+
"grad_norm": 1.7693496942520142,
|
| 89 |
+
"learning_rate": 9.706914150954466e-05,
|
| 90 |
+
"loss": 1.2781,
|
| 91 |
+
"step": 3000
|
| 92 |
+
},
|
| 93 |
+
{
|
| 94 |
+
"epoch": 0.1049795289918466,
|
| 95 |
+
"eval_loss": 1.222457766532898,
|
| 96 |
+
"eval_runtime": 1838.7539,
|
| 97 |
+
"eval_samples_per_second": 27.63,
|
| 98 |
+
"eval_steps_per_second": 3.454,
|
| 99 |
+
"step": 3000
|
| 100 |
+
},
|
| 101 |
+
{
|
| 102 |
+
"epoch": 0.12247611715715435,
|
| 103 |
+
"grad_norm": 1.739169716835022,
|
| 104 |
+
"learning_rate": 9.648250049864486e-05,
|
| 105 |
+
"loss": 1.1931,
|
| 106 |
+
"step": 3500
|
| 107 |
+
},
|
| 108 |
+
{
|
| 109 |
+
"epoch": 0.12247611715715435,
|
| 110 |
+
"eval_loss": 1.1909434795379639,
|
| 111 |
+
"eval_runtime": 1838.4303,
|
| 112 |
+
"eval_samples_per_second": 27.634,
|
| 113 |
+
"eval_steps_per_second": 3.455,
|
| 114 |
+
"step": 3500
|
| 115 |
+
},
|
| 116 |
+
{
|
| 117 |
+
"epoch": 0.13997270532246212,
|
| 118 |
+
"grad_norm": 1.2461516857147217,
|
| 119 |
+
"learning_rate": 9.589585948774507e-05,
|
| 120 |
+
"loss": 1.1771,
|
| 121 |
+
"step": 4000
|
| 122 |
+
},
|
| 123 |
+
{
|
| 124 |
+
"epoch": 0.13997270532246212,
|
| 125 |
+
"eval_loss": 1.1642112731933594,
|
| 126 |
+
"eval_runtime": 1839.7956,
|
| 127 |
+
"eval_samples_per_second": 27.614,
|
| 128 |
+
"eval_steps_per_second": 3.452,
|
| 129 |
+
"step": 4000
|
| 130 |
+
},
|
| 131 |
+
{
|
| 132 |
+
"epoch": 0.1574692934877699,
|
| 133 |
+
"grad_norm": 1.5028084516525269,
|
| 134 |
+
"learning_rate": 9.530921847684529e-05,
|
| 135 |
+
"loss": 1.1496,
|
| 136 |
+
"step": 4500
|
| 137 |
+
},
|
| 138 |
+
{
|
| 139 |
+
"epoch": 0.1574692934877699,
|
| 140 |
+
"eval_loss": 1.141344428062439,
|
| 141 |
+
"eval_runtime": 1839.6492,
|
| 142 |
+
"eval_samples_per_second": 27.616,
|
| 143 |
+
"eval_steps_per_second": 3.452,
|
| 144 |
+
"step": 4500
|
| 145 |
+
},
|
| 146 |
+
{
|
| 147 |
+
"epoch": 0.17496588165307764,
|
| 148 |
+
"grad_norm": 1.0963733196258545,
|
| 149 |
+
"learning_rate": 9.47225774659455e-05,
|
| 150 |
+
"loss": 1.1325,
|
| 151 |
+
"step": 5000
|
| 152 |
+
},
|
| 153 |
+
{
|
| 154 |
+
"epoch": 0.17496588165307764,
|
| 155 |
+
"eval_loss": 1.12198007106781,
|
| 156 |
+
"eval_runtime": 1840.0292,
|
| 157 |
+
"eval_samples_per_second": 27.61,
|
| 158 |
+
"eval_steps_per_second": 3.452,
|
| 159 |
+
"step": 5000
|
| 160 |
+
},
|
| 161 |
+
{
|
| 162 |
+
"epoch": 0.1924624698183854,
|
| 163 |
+
"grad_norm": 0.9784343838691711,
|
| 164 |
+
"learning_rate": 9.41359364550457e-05,
|
| 165 |
+
"loss": 1.1301,
|
| 166 |
+
"step": 5500
|
| 167 |
+
},
|
| 168 |
+
{
|
| 169 |
+
"epoch": 0.1924624698183854,
|
| 170 |
+
"eval_loss": 1.1025567054748535,
|
| 171 |
+
"eval_runtime": 1840.7493,
|
| 172 |
+
"eval_samples_per_second": 27.6,
|
| 173 |
+
"eval_steps_per_second": 3.45,
|
| 174 |
+
"step": 5500
|
| 175 |
+
},
|
| 176 |
+
{
|
| 177 |
+
"epoch": 0.2099590579836932,
|
| 178 |
+
"grad_norm": 1.8819756507873535,
|
| 179 |
+
"learning_rate": 9.354929544414591e-05,
|
| 180 |
+
"loss": 1.1118,
|
| 181 |
+
"step": 6000
|
| 182 |
+
},
|
| 183 |
+
{
|
| 184 |
+
"epoch": 0.2099590579836932,
|
| 185 |
+
"eval_loss": 1.0877833366394043,
|
| 186 |
+
"eval_runtime": 1840.7767,
|
| 187 |
+
"eval_samples_per_second": 27.599,
|
| 188 |
+
"eval_steps_per_second": 3.45,
|
| 189 |
+
"step": 6000
|
| 190 |
+
},
|
| 191 |
+
{
|
| 192 |
+
"epoch": 0.22745564614900093,
|
| 193 |
+
"grad_norm": 1.390010118484497,
|
| 194 |
+
"learning_rate": 9.296382771526793e-05,
|
| 195 |
+
"loss": 1.1046,
|
| 196 |
+
"step": 6500
|
| 197 |
+
},
|
| 198 |
+
{
|
| 199 |
+
"epoch": 0.22745564614900093,
|
| 200 |
+
"eval_loss": 1.0721691846847534,
|
| 201 |
+
"eval_runtime": 1840.2034,
|
| 202 |
+
"eval_samples_per_second": 27.608,
|
| 203 |
+
"eval_steps_per_second": 3.451,
|
| 204 |
+
"step": 6500
|
| 205 |
+
},
|
| 206 |
+
{
|
| 207 |
+
"epoch": 0.2449522343143087,
|
| 208 |
+
"grad_norm": 1.1395955085754395,
|
| 209 |
+
"learning_rate": 9.237718670436814e-05,
|
| 210 |
+
"loss": 1.071,
|
| 211 |
+
"step": 7000
|
| 212 |
+
},
|
| 213 |
+
{
|
| 214 |
+
"epoch": 0.2449522343143087,
|
| 215 |
+
"eval_loss": 1.0588288307189941,
|
| 216 |
+
"eval_runtime": 1841.3244,
|
| 217 |
+
"eval_samples_per_second": 27.591,
|
| 218 |
+
"eval_steps_per_second": 3.449,
|
| 219 |
+
"step": 7000
|
| 220 |
+
},
|
| 221 |
+
{
|
| 222 |
+
"epoch": 0.2624488224796165,
|
| 223 |
+
"grad_norm": 1.557716965675354,
|
| 224 |
+
"learning_rate": 9.179054569346835e-05,
|
| 225 |
+
"loss": 1.0578,
|
| 226 |
+
"step": 7500
|
| 227 |
+
},
|
| 228 |
+
{
|
| 229 |
+
"epoch": 0.2624488224796165,
|
| 230 |
+
"eval_loss": 1.0490485429763794,
|
| 231 |
+
"eval_runtime": 1841.5042,
|
| 232 |
+
"eval_samples_per_second": 27.588,
|
| 233 |
+
"eval_steps_per_second": 3.449,
|
| 234 |
+
"step": 7500
|
| 235 |
+
},
|
| 236 |
+
{
|
| 237 |
+
"epoch": 0.27994541064492423,
|
| 238 |
+
"grad_norm": 1.6820220947265625,
|
| 239 |
+
"learning_rate": 9.120390468256855e-05,
|
| 240 |
+
"loss": 1.0293,
|
| 241 |
+
"step": 8000
|
| 242 |
+
},
|
| 243 |
+
{
|
| 244 |
+
"epoch": 0.27994541064492423,
|
| 245 |
+
"eval_loss": 1.0374187231063843,
|
| 246 |
+
"eval_runtime": 1841.7893,
|
| 247 |
+
"eval_samples_per_second": 27.584,
|
| 248 |
+
"eval_steps_per_second": 3.448,
|
| 249 |
+
"step": 8000
|
| 250 |
+
},
|
| 251 |
+
{
|
| 252 |
+
"epoch": 0.297441998810232,
|
| 253 |
+
"grad_norm": 1.325988531112671,
|
| 254 |
+
"learning_rate": 9.061726367166876e-05,
|
| 255 |
+
"loss": 1.0229,
|
| 256 |
+
"step": 8500
|
| 257 |
+
},
|
| 258 |
+
{
|
| 259 |
+
"epoch": 0.297441998810232,
|
| 260 |
+
"eval_loss": 1.0274155139923096,
|
| 261 |
+
"eval_runtime": 1842.3456,
|
| 262 |
+
"eval_samples_per_second": 27.576,
|
| 263 |
+
"eval_steps_per_second": 3.447,
|
| 264 |
+
"step": 8500
|
| 265 |
+
},
|
| 266 |
+
{
|
| 267 |
+
"epoch": 0.3149385869755398,
|
| 268 |
+
"grad_norm": 1.0609859228134155,
|
| 269 |
+
"learning_rate": 9.003179594279078e-05,
|
| 270 |
+
"loss": 1.0345,
|
| 271 |
+
"step": 9000
|
| 272 |
+
},
|
| 273 |
+
{
|
| 274 |
+
"epoch": 0.3149385869755398,
|
| 275 |
+
"eval_loss": 1.0168765783309937,
|
| 276 |
+
"eval_runtime": 1843.5677,
|
| 277 |
+
"eval_samples_per_second": 27.557,
|
| 278 |
+
"eval_steps_per_second": 3.445,
|
| 279 |
+
"step": 9000
|
| 280 |
+
},
|
| 281 |
+
{
|
| 282 |
+
"epoch": 0.3324351751408475,
|
| 283 |
+
"grad_norm": 0.6549689173698425,
|
| 284 |
+
"learning_rate": 8.944515493189099e-05,
|
| 285 |
+
"loss": 1.0255,
|
| 286 |
+
"step": 9500
|
| 287 |
+
},
|
| 288 |
+
{
|
| 289 |
+
"epoch": 0.3324351751408475,
|
| 290 |
+
"eval_loss": 1.007939338684082,
|
| 291 |
+
"eval_runtime": 1841.3868,
|
| 292 |
+
"eval_samples_per_second": 27.59,
|
| 293 |
+
"eval_steps_per_second": 3.449,
|
| 294 |
+
"step": 9500
|
| 295 |
+
},
|
| 296 |
+
{
|
| 297 |
+
"epoch": 0.3499317633061553,
|
| 298 |
+
"grad_norm": 1.5119966268539429,
|
| 299 |
+
"learning_rate": 8.88585139209912e-05,
|
| 300 |
+
"loss": 1.0412,
|
| 301 |
+
"step": 10000
|
| 302 |
+
},
|
| 303 |
+
{
|
| 304 |
+
"epoch": 0.3499317633061553,
|
| 305 |
+
"eval_loss": 1.0019127130508423,
|
| 306 |
+
"eval_runtime": 1841.6495,
|
| 307 |
+
"eval_samples_per_second": 27.586,
|
| 308 |
+
"eval_steps_per_second": 3.449,
|
| 309 |
+
"step": 10000
|
| 310 |
+
},
|
| 311 |
+
{
|
| 312 |
+
"epoch": 0.3674283514714631,
|
| 313 |
+
"grad_norm": 1.2304165363311768,
|
| 314 |
+
"learning_rate": 8.82718729100914e-05,
|
| 315 |
+
"loss": 0.9972,
|
| 316 |
+
"step": 10500
|
| 317 |
+
},
|
| 318 |
+
{
|
| 319 |
+
"epoch": 0.3674283514714631,
|
| 320 |
+
"eval_loss": 0.9952266216278076,
|
| 321 |
+
"eval_runtime": 1841.7158,
|
| 322 |
+
"eval_samples_per_second": 27.585,
|
| 323 |
+
"eval_steps_per_second": 3.448,
|
| 324 |
+
"step": 10500
|
| 325 |
+
},
|
| 326 |
+
{
|
| 327 |
+
"epoch": 0.3849249396367708,
|
| 328 |
+
"grad_norm": 1.0097540616989136,
|
| 329 |
+
"learning_rate": 8.768640518121341e-05,
|
| 330 |
+
"loss": 0.9861,
|
| 331 |
+
"step": 11000
|
| 332 |
+
},
|
| 333 |
+
{
|
| 334 |
+
"epoch": 0.3849249396367708,
|
| 335 |
+
"eval_loss": 0.9871318936347961,
|
| 336 |
+
"eval_runtime": 1841.3777,
|
| 337 |
+
"eval_samples_per_second": 27.59,
|
| 338 |
+
"eval_steps_per_second": 3.449,
|
| 339 |
+
"step": 11000
|
| 340 |
+
},
|
| 341 |
+
{
|
| 342 |
+
"epoch": 0.40242152780207857,
|
| 343 |
+
"grad_norm": 1.4742999076843262,
|
| 344 |
+
"learning_rate": 8.709976417031363e-05,
|
| 345 |
+
"loss": 0.979,
|
| 346 |
+
"step": 11500
|
| 347 |
+
},
|
| 348 |
+
{
|
| 349 |
+
"epoch": 0.40242152780207857,
|
| 350 |
+
"eval_loss": 0.9808717370033264,
|
| 351 |
+
"eval_runtime": 1841.6224,
|
| 352 |
+
"eval_samples_per_second": 27.587,
|
| 353 |
+
"eval_steps_per_second": 3.449,
|
| 354 |
+
"step": 11500
|
| 355 |
+
},
|
| 356 |
+
{
|
| 357 |
+
"epoch": 0.4199181159673864,
|
| 358 |
+
"grad_norm": 1.0650264024734497,
|
| 359 |
+
"learning_rate": 8.651312315941384e-05,
|
| 360 |
+
"loss": 0.9968,
|
| 361 |
+
"step": 12000
|
| 362 |
+
},
|
| 363 |
+
{
|
| 364 |
+
"epoch": 0.4199181159673864,
|
| 365 |
+
"eval_loss": 0.9748508930206299,
|
| 366 |
+
"eval_runtime": 1841.2412,
|
| 367 |
+
"eval_samples_per_second": 27.592,
|
| 368 |
+
"eval_steps_per_second": 3.449,
|
| 369 |
+
"step": 12000
|
| 370 |
+
},
|
| 371 |
+
{
|
| 372 |
+
"epoch": 0.4374147041326941,
|
| 373 |
+
"grad_norm": 0.9701998233795166,
|
| 374 |
+
"learning_rate": 8.592648214851404e-05,
|
| 375 |
+
"loss": 0.978,
|
| 376 |
+
"step": 12500
|
| 377 |
+
},
|
| 378 |
+
{
|
| 379 |
+
"epoch": 0.4374147041326941,
|
| 380 |
+
"eval_loss": 0.9688844680786133,
|
| 381 |
+
"eval_runtime": 1841.3182,
|
| 382 |
+
"eval_samples_per_second": 27.591,
|
| 383 |
+
"eval_steps_per_second": 3.449,
|
| 384 |
+
"step": 12500
|
| 385 |
+
},
|
| 386 |
+
{
|
| 387 |
+
"epoch": 0.45491129229800187,
|
| 388 |
+
"grad_norm": 1.7642956972122192,
|
| 389 |
+
"learning_rate": 8.533984113761425e-05,
|
| 390 |
+
"loss": 0.9682,
|
| 391 |
+
"step": 13000
|
| 392 |
+
},
|
| 393 |
+
{
|
| 394 |
+
"epoch": 0.45491129229800187,
|
| 395 |
+
"eval_loss": 0.9631832838058472,
|
| 396 |
+
"eval_runtime": 1842.1886,
|
| 397 |
+
"eval_samples_per_second": 27.578,
|
| 398 |
+
"eval_steps_per_second": 3.448,
|
| 399 |
+
"step": 13000
|
| 400 |
+
},
|
| 401 |
+
{
|
| 402 |
+
"epoch": 0.47240788046330967,
|
| 403 |
+
"grad_norm": 1.9703904390335083,
|
| 404 |
+
"learning_rate": 8.475320012671446e-05,
|
| 405 |
+
"loss": 0.9594,
|
| 406 |
+
"step": 13500
|
| 407 |
+
},
|
| 408 |
+
{
|
| 409 |
+
"epoch": 0.47240788046330967,
|
| 410 |
+
"eval_loss": 0.9580743908882141,
|
| 411 |
+
"eval_runtime": 1841.8156,
|
| 412 |
+
"eval_samples_per_second": 27.584,
|
| 413 |
+
"eval_steps_per_second": 3.448,
|
| 414 |
+
"step": 13500
|
| 415 |
+
},
|
| 416 |
+
{
|
| 417 |
+
"epoch": 0.4899044686286174,
|
| 418 |
+
"grad_norm": 1.6822909116744995,
|
| 419 |
+
"learning_rate": 8.416655911581468e-05,
|
| 420 |
+
"loss": 0.9565,
|
| 421 |
+
"step": 14000
|
| 422 |
+
},
|
| 423 |
+
{
|
| 424 |
+
"epoch": 0.4899044686286174,
|
| 425 |
+
"eval_loss": 0.952092707157135,
|
| 426 |
+
"eval_runtime": 1841.6575,
|
| 427 |
+
"eval_samples_per_second": 27.586,
|
| 428 |
+
"eval_steps_per_second": 3.449,
|
| 429 |
+
"step": 14000
|
| 430 |
+
},
|
| 431 |
+
{
|
| 432 |
+
"epoch": 0.5074010567939252,
|
| 433 |
+
"grad_norm": 1.2567087411880493,
|
| 434 |
+
"learning_rate": 8.357991810491488e-05,
|
| 435 |
+
"loss": 0.9328,
|
| 436 |
+
"step": 14500
|
| 437 |
+
},
|
| 438 |
+
{
|
| 439 |
+
"epoch": 0.5074010567939252,
|
| 440 |
+
"eval_loss": 0.9498482346534729,
|
| 441 |
+
"eval_runtime": 1841.8538,
|
| 442 |
+
"eval_samples_per_second": 27.583,
|
| 443 |
+
"eval_steps_per_second": 3.448,
|
| 444 |
+
"step": 14500
|
| 445 |
+
},
|
| 446 |
+
{
|
| 447 |
+
"epoch": 0.524897644959233,
|
| 448 |
+
"grad_norm": 1.645724892616272,
|
| 449 |
+
"learning_rate": 8.299445037603689e-05,
|
| 450 |
+
"loss": 0.9612,
|
| 451 |
+
"step": 15000
|
| 452 |
+
},
|
| 453 |
+
{
|
| 454 |
+
"epoch": 0.524897644959233,
|
| 455 |
+
"eval_loss": 0.944929301738739,
|
| 456 |
+
"eval_runtime": 1841.6649,
|
| 457 |
+
"eval_samples_per_second": 27.586,
|
| 458 |
+
"eval_steps_per_second": 3.449,
|
| 459 |
+
"step": 15000
|
| 460 |
+
},
|
| 461 |
+
{
|
| 462 |
+
"epoch": 0.5423942331245407,
|
| 463 |
+
"grad_norm": 1.0592772960662842,
|
| 464 |
+
"learning_rate": 8.24089826471589e-05,
|
| 465 |
+
"loss": 0.9532,
|
| 466 |
+
"step": 15500
|
| 467 |
+
},
|
| 468 |
+
{
|
| 469 |
+
"epoch": 0.5423942331245407,
|
| 470 |
+
"eval_loss": 0.9399999380111694,
|
| 471 |
+
"eval_runtime": 1841.4529,
|
| 472 |
+
"eval_samples_per_second": 27.589,
|
| 473 |
+
"eval_steps_per_second": 3.449,
|
| 474 |
+
"step": 15500
|
| 475 |
+
}
|
| 476 |
+
],
|
| 477 |
+
"logging_steps": 500,
|
| 478 |
+
"max_steps": 85731,
|
| 479 |
+
"num_input_tokens_seen": 0,
|
| 480 |
+
"num_train_epochs": 3,
|
| 481 |
+
"save_steps": 500,
|
| 482 |
+
"stateful_callbacks": {
|
| 483 |
+
"EarlyStoppingCallback": {
|
| 484 |
+
"args": {
|
| 485 |
+
"early_stopping_patience": 3,
|
| 486 |
+
"early_stopping_threshold": 0.0
|
| 487 |
+
},
|
| 488 |
+
"attributes": {
|
| 489 |
+
"early_stopping_patience_counter": 0
|
| 490 |
+
}
|
| 491 |
+
},
|
| 492 |
+
"TrainerControl": {
|
| 493 |
+
"args": {
|
| 494 |
+
"should_epoch_stop": false,
|
| 495 |
+
"should_evaluate": false,
|
| 496 |
+
"should_log": false,
|
| 497 |
+
"should_save": true,
|
| 498 |
+
"should_training_stop": false
|
| 499 |
+
},
|
| 500 |
+
"attributes": {}
|
| 501 |
+
}
|
| 502 |
+
},
|
| 503 |
+
"total_flos": 6.4779360141312e+16,
|
| 504 |
+
"train_batch_size": 4,
|
| 505 |
+
"trial_name": null,
|
| 506 |
+
"trial_params": null
|
| 507 |
+
}
|
checkpoint-15500/training_args.bin
ADDED
|
@@ -0,0 +1,3 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
version https://git-lfs.github.com/spec/v1
|
| 2 |
+
oid sha256:d1174044396b4e2f4cc6381a71fd810c6b5d8fb1d414d61315c45be1a1a8533a
|
| 3 |
+
size 5304
|