Jiminiya commited on
Commit
f0e6d73
·
verified ·
1 Parent(s): d3dbd13

Upload folder using huggingface_hub

Browse files
README.md ADDED
@@ -0,0 +1,145 @@
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
1
+ ---
2
+ library_name: transformers
3
+ license: apache-2.0
4
+ base_model: Jiminiya/tiny_step1
5
+ tags:
6
+ - generated_from_trainer
7
+ model-index:
8
+ - name: outputs_continue_2
9
+ results: []
10
+ ---
11
+
12
+ <!-- This model card has been generated automatically according to the information the Trainer had access to. You
13
+ should probably proofread and complete it, then remove this comment. -->
14
+
15
+ [<img src="https://raw.githubusercontent.com/axolotl-ai-cloud/axolotl/main/image/axolotl-badge-web.png" alt="Built with Axolotl" width="200" height="32"/>](https://github.com/axolotl-ai-cloud/axolotl)
16
+ <details><summary>See axolotl config</summary>
17
+
18
+ axolotl version: `0.6.0`
19
+ ```yaml
20
+ #base_model: meta-llama/Llama-3.1-8B
21
+ base_model: Jiminiya/tiny_step1
22
+ model_type: LlamaForCausalLM
23
+ tokenizer_type: AutoTokenizer
24
+ special_tokens:
25
+ pad_token: "</s>"
26
+
27
+ load_in_8bit: false
28
+ load_in_4bit: false
29
+ strict: false
30
+
31
+ # max_steps:학습할 step. 논문 상에서는 400이라고 표기
32
+ # 단, 학습 환경의 테스트를 위해 50으로 변경
33
+ max_steps: 180
34
+ pretraining_dataset:
35
+ - path: Jiminiya/INU
36
+ type: pretrain
37
+ val_set_size: 0.0
38
+ output_dir: ./outputs_continue_2
39
+ #dataset_prepared_path:
40
+
41
+ unfrozen_parameters:
42
+ - ^lm_head.*
43
+ # - ^model.embed_tokens.weight
44
+ # ^model.layers.*
45
+
46
+ sequence_len: 2048
47
+ sample_packing: false
48
+ pad_to_sequence_len: false
49
+
50
+ wandb_project:
51
+ wandb_entity:
52
+ wandb_watch:
53
+ wandb_name:
54
+ wandb_log_model:
55
+
56
+ # gradient_accumulation_steps: 4
57
+ # micro_batch_size: 8
58
+ gradient_accumulation_steps: 4
59
+ micro_batch_size: 2
60
+ optimizer: adamw_torch
61
+ lr_scheduler: cosine
62
+ learning_rate: 4e-5
63
+
64
+ train_on_inputs: false
65
+ group_by_length: false
66
+ bf16: auto
67
+ fp16:
68
+ tf32:
69
+
70
+ gradient_checkpointing: true
71
+ early_stopping_patience:
72
+ resume_from_checkpoint:
73
+ logging_steps: 1
74
+ xformers_attention:
75
+ flash_attention: false
76
+
77
+ warmup_steps: 10
78
+ save_steps: 200
79
+ debug:
80
+ deepspeed:
81
+ weight_decay: 0.0
82
+ fsdp:
83
+ - full_shard
84
+ - auto_wrap
85
+ fsdp_config:
86
+ fsdp_limit_all_gathers: true
87
+ fsdp_sync_module_states: true
88
+ fsdp_offload_params: true
89
+ fsdp_use_orig_params: true
90
+ fsdp_cpu_ram_efficient_loading: false
91
+ fsdp_auto_wrap_policy: TRANSFORMER_BASED_WRAP
92
+ fsdp_transformer_layer_cls_to_wrap: LlamaDecoderLayer
93
+ fsdp_state_dict_type: FULL_STATE_DICT
94
+ fsdp_sharding_strategy: FULL_SHARD
95
+ fsdp_backward_prefetch: BACKWARD_POST
96
+
97
+ ```
98
+
99
+ </details><br>
100
+
101
+ # outputs_continue_2
102
+
103
+ This model is a fine-tuned version of [Jiminiya/tiny_step1](https://huggingface.co/Jiminiya/tiny_step1) on an unknown dataset.
104
+
105
+ ## Model description
106
+
107
+ More information needed
108
+
109
+ ## Intended uses & limitations
110
+
111
+ More information needed
112
+
113
+ ## Training and evaluation data
114
+
115
+ More information needed
116
+
117
+ ## Training procedure
118
+
119
+ ### Training hyperparameters
120
+
121
+ The following hyperparameters were used during training:
122
+ - learning_rate: 4e-05
123
+ - train_batch_size: 2
124
+ - eval_batch_size: 2
125
+ - seed: 42
126
+ - distributed_type: multi-GPU
127
+ - num_devices: 4
128
+ - gradient_accumulation_steps: 4
129
+ - total_train_batch_size: 32
130
+ - total_eval_batch_size: 8
131
+ - optimizer: Use adamw_torch with betas=(0.9,0.999) and epsilon=1e-08 and optimizer_args=No additional optimizer arguments
132
+ - lr_scheduler_type: cosine
133
+ - lr_scheduler_warmup_steps: 10
134
+ - training_steps: 180
135
+
136
+ ### Training results
137
+
138
+
139
+
140
+ ### Framework versions
141
+
142
+ - Transformers 4.48.1
143
+ - Pytorch 2.5.1+cu124
144
+ - Datasets 3.2.0
145
+ - Tokenizers 0.21.0
checkpoint-180/config.json ADDED
@@ -0,0 +1,30 @@
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
1
+ {
2
+ "_name_or_path": "Jiminiya/tiny_step1",
3
+ "architectures": [
4
+ "LlamaForCausalLM"
5
+ ],
6
+ "attention_bias": false,
7
+ "attention_dropout": 0.0,
8
+ "bos_token_id": 1,
9
+ "eos_token_id": 2,
10
+ "head_dim": 64,
11
+ "hidden_act": "silu",
12
+ "hidden_size": 2048,
13
+ "initializer_range": 0.02,
14
+ "intermediate_size": 5632,
15
+ "max_position_embeddings": 2048,
16
+ "mlp_bias": false,
17
+ "model_type": "llama",
18
+ "num_attention_heads": 32,
19
+ "num_hidden_layers": 22,
20
+ "num_key_value_heads": 4,
21
+ "pretraining_tp": 1,
22
+ "rms_norm_eps": 1e-05,
23
+ "rope_scaling": null,
24
+ "rope_theta": 10000.0,
25
+ "tie_word_embeddings": false,
26
+ "torch_dtype": "float32",
27
+ "transformers_version": "4.48.1",
28
+ "use_cache": false,
29
+ "vocab_size": 32000
30
+ }
checkpoint-180/generation_config.json ADDED
@@ -0,0 +1,8 @@
 
 
 
 
 
 
 
 
 
1
+ {
2
+ "bos_token_id": 1,
3
+ "do_sample": true,
4
+ "eos_token_id": 2,
5
+ "max_length": 2048,
6
+ "pad_token_id": 0,
7
+ "transformers_version": "4.48.1"
8
+ }
checkpoint-180/model.safetensors ADDED
@@ -0,0 +1,3 @@
 
 
 
 
1
+ version https://git-lfs.github.com/spec/v1
2
+ oid sha256:bc8a36d88735d6b99d596818ff20cd98dd222f4e4670290782996cce83e8f042
3
+ size 2462268008
checkpoint-180/optimizer.bin ADDED
@@ -0,0 +1,3 @@
 
 
 
 
1
+ version https://git-lfs.github.com/spec/v1
2
+ oid sha256:a1078fae233fa0f997c10c7598abecc8c91aa63d3944eced6f5d37f5a18462e2
3
+ size 524290146
checkpoint-180/pytorch_model_fsdp.bin ADDED
@@ -0,0 +1,3 @@
 
 
 
 
1
+ version https://git-lfs.github.com/spec/v1
2
+ oid sha256:196af0794ca205d81a1b3f358ef781b25a9ef1452a4cd755c7ae0c07cd8176e4
3
+ size 2462343983
checkpoint-180/rng_state_0.pth ADDED
@@ -0,0 +1,3 @@
 
 
 
 
1
+ version https://git-lfs.github.com/spec/v1
2
+ oid sha256:7ffb639768c91708df695c585429e90af6ac36cc2a44704cd2bc84a935ad31f2
3
+ size 15024
checkpoint-180/rng_state_1.pth ADDED
@@ -0,0 +1,3 @@
 
 
 
 
1
+ version https://git-lfs.github.com/spec/v1
2
+ oid sha256:8be46b342c87cbcabffa42238e3556ee82e64446303b89f7a49d99beb8067148
3
+ size 15024
checkpoint-180/rng_state_2.pth ADDED
@@ -0,0 +1,3 @@
 
 
 
 
1
+ version https://git-lfs.github.com/spec/v1
2
+ oid sha256:f67cb6d3c987c5e9d82aa648dc8aee43609b458f8baef5e18f7f72b31fea647d
3
+ size 15024
checkpoint-180/rng_state_3.pth ADDED
@@ -0,0 +1,3 @@
 
 
 
 
1
+ version https://git-lfs.github.com/spec/v1
2
+ oid sha256:625033414f26c16149a3c8e0f791602030231cb8d72adbbaf0f9abe58b8ecffb
3
+ size 15024
checkpoint-180/scheduler.pt ADDED
@@ -0,0 +1,3 @@
 
 
 
 
1
+ version https://git-lfs.github.com/spec/v1
2
+ oid sha256:6335a47e089fab921060fa4128968c70aa54db276ec3393a5c66c98e57e0dd4b
3
+ size 1064
checkpoint-180/special_tokens_map.json ADDED
@@ -0,0 +1,30 @@
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
1
+ {
2
+ "bos_token": {
3
+ "content": "<s>",
4
+ "lstrip": false,
5
+ "normalized": false,
6
+ "rstrip": false,
7
+ "single_word": false
8
+ },
9
+ "eos_token": {
10
+ "content": "</s>",
11
+ "lstrip": false,
12
+ "normalized": false,
13
+ "rstrip": false,
14
+ "single_word": false
15
+ },
16
+ "pad_token": {
17
+ "content": "</s>",
18
+ "lstrip": false,
19
+ "normalized": false,
20
+ "rstrip": false,
21
+ "single_word": false
22
+ },
23
+ "unk_token": {
24
+ "content": "<unk>",
25
+ "lstrip": false,
26
+ "normalized": false,
27
+ "rstrip": false,
28
+ "single_word": false
29
+ }
30
+ }
checkpoint-180/tokenizer.json ADDED
The diff for this file is too large to render. See raw diff
 
checkpoint-180/tokenizer.model ADDED
@@ -0,0 +1,3 @@
 
 
 
 
1
+ version https://git-lfs.github.com/spec/v1
2
+ oid sha256:9e556afd44213b6bd1be2b850ebbbd98f5481437a8021afaf58ee7fb1818d347
3
+ size 499723
checkpoint-180/tokenizer_config.json ADDED
@@ -0,0 +1,43 @@
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
1
+ {
2
+ "add_bos_token": true,
3
+ "add_eos_token": false,
4
+ "add_prefix_space": null,
5
+ "added_tokens_decoder": {
6
+ "0": {
7
+ "content": "<unk>",
8
+ "lstrip": false,
9
+ "normalized": false,
10
+ "rstrip": false,
11
+ "single_word": false,
12
+ "special": true
13
+ },
14
+ "1": {
15
+ "content": "<s>",
16
+ "lstrip": false,
17
+ "normalized": false,
18
+ "rstrip": false,
19
+ "single_word": false,
20
+ "special": true
21
+ },
22
+ "2": {
23
+ "content": "</s>",
24
+ "lstrip": false,
25
+ "normalized": false,
26
+ "rstrip": false,
27
+ "single_word": false,
28
+ "special": true
29
+ }
30
+ },
31
+ "bos_token": "<s>",
32
+ "clean_up_tokenization_spaces": false,
33
+ "eos_token": "</s>",
34
+ "extra_special_tokens": {},
35
+ "legacy": false,
36
+ "model_max_length": 1000000000000000019884624838656,
37
+ "pad_token": "</s>",
38
+ "padding_side": "right",
39
+ "sp_model_kwargs": {},
40
+ "tokenizer_class": "LlamaTokenizer",
41
+ "unk_token": "<unk>",
42
+ "use_default_system_prompt": false
43
+ }
checkpoint-180/trainer_state.json ADDED
@@ -0,0 +1,1293 @@
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
1
+ {
2
+ "best_metric": null,
3
+ "best_model_checkpoint": null,
4
+ "epoch": 89.0111111111111,
5
+ "eval_steps": 500,
6
+ "global_step": 180,
7
+ "is_hyper_param_search": false,
8
+ "is_local_process_zero": true,
9
+ "is_world_process_zero": true,
10
+ "log_history": [
11
+ {
12
+ "epoch": 0.005555555555555556,
13
+ "grad_norm": 2.4661359786987305,
14
+ "learning_rate": 4.000000000000001e-06,
15
+ "loss": 1.6655,
16
+ "step": 1
17
+ },
18
+ {
19
+ "epoch": 0.011111111111111112,
20
+ "grad_norm": 2.436788320541382,
21
+ "learning_rate": 8.000000000000001e-06,
22
+ "loss": 1.4955,
23
+ "step": 2
24
+ },
25
+ {
26
+ "epoch": 1.0055555555555555,
27
+ "grad_norm": 2.709148645401001,
28
+ "learning_rate": 1.2e-05,
29
+ "loss": 3.176,
30
+ "step": 3
31
+ },
32
+ {
33
+ "epoch": 1.011111111111111,
34
+ "grad_norm": 2.531345844268799,
35
+ "learning_rate": 1.6000000000000003e-05,
36
+ "loss": 1.4436,
37
+ "step": 4
38
+ },
39
+ {
40
+ "epoch": 2.0055555555555555,
41
+ "grad_norm": 2.1641957759857178,
42
+ "learning_rate": 2e-05,
43
+ "loss": 3.2386,
44
+ "step": 5
45
+ },
46
+ {
47
+ "epoch": 2.011111111111111,
48
+ "grad_norm": 2.5990405082702637,
49
+ "learning_rate": 2.4e-05,
50
+ "loss": 1.5247,
51
+ "step": 6
52
+ },
53
+ {
54
+ "epoch": 3.0055555555555555,
55
+ "grad_norm": 2.355940818786621,
56
+ "learning_rate": 2.8e-05,
57
+ "loss": 2.946,
58
+ "step": 7
59
+ },
60
+ {
61
+ "epoch": 3.011111111111111,
62
+ "grad_norm": 2.312788963317871,
63
+ "learning_rate": 3.2000000000000005e-05,
64
+ "loss": 1.5807,
65
+ "step": 8
66
+ },
67
+ {
68
+ "epoch": 4.0055555555555555,
69
+ "grad_norm": 2.0773563385009766,
70
+ "learning_rate": 3.6e-05,
71
+ "loss": 2.7929,
72
+ "step": 9
73
+ },
74
+ {
75
+ "epoch": 4.011111111111111,
76
+ "grad_norm": 2.6093788146972656,
77
+ "learning_rate": 4e-05,
78
+ "loss": 1.5504,
79
+ "step": 10
80
+ },
81
+ {
82
+ "epoch": 5.0055555555555555,
83
+ "grad_norm": 2.414492130279541,
84
+ "learning_rate": 3.9996585009161056e-05,
85
+ "loss": 2.8795,
86
+ "step": 11
87
+ },
88
+ {
89
+ "epoch": 5.011111111111111,
90
+ "grad_norm": 2.4428486824035645,
91
+ "learning_rate": 3.9986341202860467e-05,
92
+ "loss": 1.4725,
93
+ "step": 12
94
+ },
95
+ {
96
+ "epoch": 6.0055555555555555,
97
+ "grad_norm": 2.1731374263763428,
98
+ "learning_rate": 3.9969272079348685e-05,
99
+ "loss": 2.7246,
100
+ "step": 13
101
+ },
102
+ {
103
+ "epoch": 6.011111111111111,
104
+ "grad_norm": 2.0872628688812256,
105
+ "learning_rate": 3.994538346771576e-05,
106
+ "loss": 1.3387,
107
+ "step": 14
108
+ },
109
+ {
110
+ "epoch": 7.0055555555555555,
111
+ "grad_norm": 2.120208740234375,
112
+ "learning_rate": 3.991468352590069e-05,
113
+ "loss": 2.5507,
114
+ "step": 15
115
+ },
116
+ {
117
+ "epoch": 7.011111111111111,
118
+ "grad_norm": 2.0171964168548584,
119
+ "learning_rate": 3.987718273790548e-05,
120
+ "loss": 1.1936,
121
+ "step": 16
122
+ },
123
+ {
124
+ "epoch": 8.005555555555556,
125
+ "grad_norm": 1.9372913837432861,
126
+ "learning_rate": 3.983289391021486e-05,
127
+ "loss": 2.4933,
128
+ "step": 17
129
+ },
130
+ {
131
+ "epoch": 8.011111111111111,
132
+ "grad_norm": 1.9526296854019165,
133
+ "learning_rate": 3.9781832167422926e-05,
134
+ "loss": 1.2765,
135
+ "step": 18
136
+ },
137
+ {
138
+ "epoch": 9.005555555555556,
139
+ "grad_norm": 2.050081968307495,
140
+ "learning_rate": 3.972401494706805e-05,
141
+ "loss": 2.3595,
142
+ "step": 19
143
+ },
144
+ {
145
+ "epoch": 9.011111111111111,
146
+ "grad_norm": 1.751308798789978,
147
+ "learning_rate": 3.965946199367804e-05,
148
+ "loss": 1.1996,
149
+ "step": 20
150
+ },
151
+ {
152
+ "epoch": 10.005555555555556,
153
+ "grad_norm": 1.745150089263916,
154
+ "learning_rate": 3.958819535202732e-05,
155
+ "loss": 2.2313,
156
+ "step": 21
157
+ },
158
+ {
159
+ "epoch": 10.011111111111111,
160
+ "grad_norm": 1.9988559484481812,
161
+ "learning_rate": 3.951023935960874e-05,
162
+ "loss": 1.2053,
163
+ "step": 22
164
+ },
165
+ {
166
+ "epoch": 11.005555555555556,
167
+ "grad_norm": 1.741628646850586,
168
+ "learning_rate": 3.942562063832228e-05,
169
+ "loss": 2.0621,
170
+ "step": 23
171
+ },
172
+ {
173
+ "epoch": 11.011111111111111,
174
+ "grad_norm": 1.809770107269287,
175
+ "learning_rate": 3.933436808538375e-05,
176
+ "loss": 1.0686,
177
+ "step": 24
178
+ },
179
+ {
180
+ "epoch": 12.005555555555556,
181
+ "grad_norm": 1.723798155784607,
182
+ "learning_rate": 3.923651286345638e-05,
183
+ "loss": 2.377,
184
+ "step": 25
185
+ },
186
+ {
187
+ "epoch": 12.011111111111111,
188
+ "grad_norm": 1.8615117073059082,
189
+ "learning_rate": 3.913208839000882e-05,
190
+ "loss": 1.0577,
191
+ "step": 26
192
+ },
193
+ {
194
+ "epoch": 13.005555555555556,
195
+ "grad_norm": 1.5270243883132935,
196
+ "learning_rate": 3.9021130325903076e-05,
197
+ "loss": 1.7282,
198
+ "step": 27
199
+ },
200
+ {
201
+ "epoch": 13.011111111111111,
202
+ "grad_norm": 1.9072110652923584,
203
+ "learning_rate": 3.89036765632164e-05,
204
+ "loss": 1.0955,
205
+ "step": 28
206
+ },
207
+ {
208
+ "epoch": 14.005555555555556,
209
+ "grad_norm": 1.666566252708435,
210
+ "learning_rate": 3.877976721230114e-05,
211
+ "loss": 1.9452,
212
+ "step": 29
213
+ },
214
+ {
215
+ "epoch": 14.011111111111111,
216
+ "grad_norm": 1.6667906045913696,
217
+ "learning_rate": 3.864944458808712e-05,
218
+ "loss": 1.0229,
219
+ "step": 30
220
+ },
221
+ {
222
+ "epoch": 15.005555555555556,
223
+ "grad_norm": 1.6143485307693481,
224
+ "learning_rate": 3.851275319563113e-05,
225
+ "loss": 1.959,
226
+ "step": 31
227
+ },
228
+ {
229
+ "epoch": 15.011111111111111,
230
+ "grad_norm": 1.6035016775131226,
231
+ "learning_rate": 3.836973971491847e-05,
232
+ "loss": 0.9805,
233
+ "step": 32
234
+ },
235
+ {
236
+ "epoch": 16.005555555555556,
237
+ "grad_norm": 1.7784571647644043,
238
+ "learning_rate": 3.822045298492177e-05,
239
+ "loss": 1.7729,
240
+ "step": 33
241
+ },
242
+ {
243
+ "epoch": 16.011111111111113,
244
+ "grad_norm": 1.4132834672927856,
245
+ "learning_rate": 3.806494398692258e-05,
246
+ "loss": 0.8385,
247
+ "step": 34
248
+ },
249
+ {
250
+ "epoch": 17.005555555555556,
251
+ "grad_norm": 1.7934105396270752,
252
+ "learning_rate": 3.790326582710125e-05,
253
+ "loss": 1.9456,
254
+ "step": 35
255
+ },
256
+ {
257
+ "epoch": 17.011111111111113,
258
+ "grad_norm": 1.3573366403579712,
259
+ "learning_rate": 3.773547371840124e-05,
260
+ "loss": 0.8648,
261
+ "step": 36
262
+ },
263
+ {
264
+ "epoch": 18.005555555555556,
265
+ "grad_norm": 1.8247311115264893,
266
+ "learning_rate": 3.756162496167396e-05,
267
+ "loss": 1.7399,
268
+ "step": 37
269
+ },
270
+ {
271
+ "epoch": 18.011111111111113,
272
+ "grad_norm": 1.4213778972625732,
273
+ "learning_rate": 3.738177892611057e-05,
274
+ "loss": 0.8886,
275
+ "step": 38
276
+ },
277
+ {
278
+ "epoch": 19.005555555555556,
279
+ "grad_norm": 1.5431638956069946,
280
+ "learning_rate": 3.719599702896745e-05,
281
+ "loss": 1.6771,
282
+ "step": 39
283
+ },
284
+ {
285
+ "epoch": 19.011111111111113,
286
+ "grad_norm": 1.3147789239883423,
287
+ "learning_rate": 3.700434271459229e-05,
288
+ "loss": 0.7712,
289
+ "step": 40
290
+ },
291
+ {
292
+ "epoch": 20.005555555555556,
293
+ "grad_norm": 1.2570959329605103,
294
+ "learning_rate": 3.680688143275786e-05,
295
+ "loss": 1.6678,
296
+ "step": 41
297
+ },
298
+ {
299
+ "epoch": 20.011111111111113,
300
+ "grad_norm": 1.4288440942764282,
301
+ "learning_rate": 3.6603680616311013e-05,
302
+ "loss": 0.7298,
303
+ "step": 42
304
+ },
305
+ {
306
+ "epoch": 21.005555555555556,
307
+ "grad_norm": 1.330296516418457,
308
+ "learning_rate": 3.639480965814443e-05,
309
+ "loss": 1.4595,
310
+ "step": 43
311
+ },
312
+ {
313
+ "epoch": 21.011111111111113,
314
+ "grad_norm": 1.5652707815170288,
315
+ "learning_rate": 3.6180339887498953e-05,
316
+ "loss": 0.8478,
317
+ "step": 44
318
+ },
319
+ {
320
+ "epoch": 22.005555555555556,
321
+ "grad_norm": 1.2386178970336914,
322
+ "learning_rate": 3.5960344545604796e-05,
323
+ "loss": 1.4772,
324
+ "step": 45
325
+ },
326
+ {
327
+ "epoch": 22.011111111111113,
328
+ "grad_norm": 1.4858933687210083,
329
+ "learning_rate": 3.573489876066967e-05,
330
+ "loss": 0.7526,
331
+ "step": 46
332
+ },
333
+ {
334
+ "epoch": 23.005555555555556,
335
+ "grad_norm": 1.356368064880371,
336
+ "learning_rate": 3.55040795222226e-05,
337
+ "loss": 1.4374,
338
+ "step": 47
339
+ },
340
+ {
341
+ "epoch": 23.011111111111113,
342
+ "grad_norm": 1.3958989381790161,
343
+ "learning_rate": 3.526796565482206e-05,
344
+ "loss": 0.7137,
345
+ "step": 48
346
+ },
347
+ {
348
+ "epoch": 24.005555555555556,
349
+ "grad_norm": 1.2687575817108154,
350
+ "learning_rate": 3.502663779113747e-05,
351
+ "loss": 1.506,
352
+ "step": 49
353
+ },
354
+ {
355
+ "epoch": 24.011111111111113,
356
+ "grad_norm": 1.1373575925827026,
357
+ "learning_rate": 3.478017834441319e-05,
358
+ "loss": 0.6985,
359
+ "step": 50
360
+ },
361
+ {
362
+ "epoch": 25.005555555555556,
363
+ "grad_norm": 1.2077847719192505,
364
+ "learning_rate": 3.452867148032449e-05,
365
+ "loss": 1.3914,
366
+ "step": 51
367
+ },
368
+ {
369
+ "epoch": 25.011111111111113,
370
+ "grad_norm": 1.3798729181289673,
371
+ "learning_rate": 3.427220308823505e-05,
372
+ "loss": 0.7729,
373
+ "step": 52
374
+ },
375
+ {
376
+ "epoch": 26.005555555555556,
377
+ "grad_norm": 1.2104331254959106,
378
+ "learning_rate": 3.401086075186582e-05,
379
+ "loss": 1.257,
380
+ "step": 53
381
+ },
382
+ {
383
+ "epoch": 26.011111111111113,
384
+ "grad_norm": 1.277470588684082,
385
+ "learning_rate": 3.374473371938526e-05,
386
+ "loss": 0.7026,
387
+ "step": 54
388
+ },
389
+ {
390
+ "epoch": 27.005555555555556,
391
+ "grad_norm": 1.319318175315857,
392
+ "learning_rate": 3.347391287293115e-05,
393
+ "loss": 1.4463,
394
+ "step": 55
395
+ },
396
+ {
397
+ "epoch": 27.011111111111113,
398
+ "grad_norm": 1.1588152647018433,
399
+ "learning_rate": 3.319849069757446e-05,
400
+ "loss": 0.6308,
401
+ "step": 56
402
+ },
403
+ {
404
+ "epoch": 28.005555555555556,
405
+ "grad_norm": 1.316746473312378,
406
+ "learning_rate": 3.291856124973575e-05,
407
+ "loss": 1.3443,
408
+ "step": 57
409
+ },
410
+ {
411
+ "epoch": 28.011111111111113,
412
+ "grad_norm": 1.1471470594406128,
413
+ "learning_rate": 3.263422012506502e-05,
414
+ "loss": 0.614,
415
+ "step": 58
416
+ },
417
+ {
418
+ "epoch": 29.005555555555556,
419
+ "grad_norm": 1.3653746843338013,
420
+ "learning_rate": 3.234556442579586e-05,
421
+ "loss": 1.3079,
422
+ "step": 59
423
+ },
424
+ {
425
+ "epoch": 29.011111111111113,
426
+ "grad_norm": 1.275138020515442,
427
+ "learning_rate": 3.205269272758513e-05,
428
+ "loss": 0.5984,
429
+ "step": 60
430
+ },
431
+ {
432
+ "epoch": 30.005555555555556,
433
+ "grad_norm": 0.9681165814399719,
434
+ "learning_rate": 3.1755705045849465e-05,
435
+ "loss": 1.2511,
436
+ "step": 61
437
+ },
438
+ {
439
+ "epoch": 30.011111111111113,
440
+ "grad_norm": 1.3781417608261108,
441
+ "learning_rate": 3.145470280161011e-05,
442
+ "loss": 0.584,
443
+ "step": 62
444
+ },
445
+ {
446
+ "epoch": 31.005555555555556,
447
+ "grad_norm": 1.1990811824798584,
448
+ "learning_rate": 3.114978878685771e-05,
449
+ "loss": 1.1897,
450
+ "step": 63
451
+ },
452
+ {
453
+ "epoch": 31.011111111111113,
454
+ "grad_norm": 1.2162977457046509,
455
+ "learning_rate": 3.084106712944899e-05,
456
+ "loss": 0.6222,
457
+ "step": 64
458
+ },
459
+ {
460
+ "epoch": 32.00555555555555,
461
+ "grad_norm": 1.2030653953552246,
462
+ "learning_rate": 3.052864325754712e-05,
463
+ "loss": 1.2274,
464
+ "step": 65
465
+ },
466
+ {
467
+ "epoch": 32.01111111111111,
468
+ "grad_norm": 0.975091278553009,
469
+ "learning_rate": 3.021262386361814e-05,
470
+ "loss": 0.5352,
471
+ "step": 66
472
+ },
473
+ {
474
+ "epoch": 33.00555555555555,
475
+ "grad_norm": 1.3519790172576904,
476
+ "learning_rate": 2.9893116867995583e-05,
477
+ "loss": 1.2279,
478
+ "step": 67
479
+ },
480
+ {
481
+ "epoch": 33.01111111111111,
482
+ "grad_norm": 0.9604120254516602,
483
+ "learning_rate": 2.9570231382025732e-05,
484
+ "loss": 0.5505,
485
+ "step": 68
486
+ },
487
+ {
488
+ "epoch": 34.00555555555555,
489
+ "grad_norm": 1.1351654529571533,
490
+ "learning_rate": 2.924407767080627e-05,
491
+ "loss": 1.2232,
492
+ "step": 69
493
+ },
494
+ {
495
+ "epoch": 34.01111111111111,
496
+ "grad_norm": 1.191280484199524,
497
+ "learning_rate": 2.891476711553077e-05,
498
+ "loss": 0.526,
499
+ "step": 70
500
+ },
501
+ {
502
+ "epoch": 35.00555555555555,
503
+ "grad_norm": 1.213642954826355,
504
+ "learning_rate": 2.858241217545218e-05,
505
+ "loss": 1.1888,
506
+ "step": 71
507
+ },
508
+ {
509
+ "epoch": 35.01111111111111,
510
+ "grad_norm": 0.8892165422439575,
511
+ "learning_rate": 2.8247126349478073e-05,
512
+ "loss": 0.5471,
513
+ "step": 72
514
+ },
515
+ {
516
+ "epoch": 36.00555555555555,
517
+ "grad_norm": 1.0873414278030396,
518
+ "learning_rate": 2.790902413741085e-05,
519
+ "loss": 1.0696,
520
+ "step": 73
521
+ },
522
+ {
523
+ "epoch": 36.01111111111111,
524
+ "grad_norm": 1.070792555809021,
525
+ "learning_rate": 2.756822100084621e-05,
526
+ "loss": 0.558,
527
+ "step": 74
528
+ },
529
+ {
530
+ "epoch": 37.00555555555555,
531
+ "grad_norm": 1.0861364603042603,
532
+ "learning_rate": 2.7224833323743064e-05,
533
+ "loss": 1.019,
534
+ "step": 75
535
+ },
536
+ {
537
+ "epoch": 37.01111111111111,
538
+ "grad_norm": 1.0294417142868042,
539
+ "learning_rate": 2.6878978372678567e-05,
540
+ "loss": 0.5205,
541
+ "step": 76
542
+ },
543
+ {
544
+ "epoch": 38.00555555555555,
545
+ "grad_norm": 0.934058666229248,
546
+ "learning_rate": 2.6530774256801666e-05,
547
+ "loss": 0.983,
548
+ "step": 77
549
+ },
550
+ {
551
+ "epoch": 38.01111111111111,
552
+ "grad_norm": 1.0980716943740845,
553
+ "learning_rate": 2.618033988749895e-05,
554
+ "loss": 0.5306,
555
+ "step": 78
556
+ },
557
+ {
558
+ "epoch": 39.00555555555555,
559
+ "grad_norm": 1.4806180000305176,
560
+ "learning_rate": 2.5827794937786497e-05,
561
+ "loss": 1.2091,
562
+ "step": 79
563
+ },
564
+ {
565
+ "epoch": 39.01111111111111,
566
+ "grad_norm": 0.8035812377929688,
567
+ "learning_rate": 2.5473259801441663e-05,
568
+ "loss": 0.4874,
569
+ "step": 80
570
+ },
571
+ {
572
+ "epoch": 40.00555555555555,
573
+ "grad_norm": 0.9660578370094299,
574
+ "learning_rate": 2.5116855551888715e-05,
575
+ "loss": 1.0444,
576
+ "step": 81
577
+ },
578
+ {
579
+ "epoch": 40.01111111111111,
580
+ "grad_norm": 1.2110332250595093,
581
+ "learning_rate": 2.4758703900852376e-05,
582
+ "loss": 0.5502,
583
+ "step": 82
584
+ },
585
+ {
586
+ "epoch": 41.00555555555555,
587
+ "grad_norm": 1.1508184671401978,
588
+ "learning_rate": 2.4398927156793376e-05,
589
+ "loss": 0.9574,
590
+ "step": 83
591
+ },
592
+ {
593
+ "epoch": 41.01111111111111,
594
+ "grad_norm": 1.0473322868347168,
595
+ "learning_rate": 2.4037648183140205e-05,
596
+ "loss": 0.47,
597
+ "step": 84
598
+ },
599
+ {
600
+ "epoch": 42.00555555555555,
601
+ "grad_norm": 0.9665315747261047,
602
+ "learning_rate": 2.367499035633141e-05,
603
+ "loss": 1.0049,
604
+ "step": 85
605
+ },
606
+ {
607
+ "epoch": 42.01111111111111,
608
+ "grad_norm": 0.9815846681594849,
609
+ "learning_rate": 2.33110775236826e-05,
610
+ "loss": 0.4856,
611
+ "step": 86
612
+ },
613
+ {
614
+ "epoch": 43.00555555555555,
615
+ "grad_norm": 0.9438784122467041,
616
+ "learning_rate": 2.2946033961092754e-05,
617
+ "loss": 0.9671,
618
+ "step": 87
619
+ },
620
+ {
621
+ "epoch": 43.01111111111111,
622
+ "grad_norm": 1.1706269979476929,
623
+ "learning_rate": 2.257998433060407e-05,
624
+ "loss": 0.4977,
625
+ "step": 88
626
+ },
627
+ {
628
+ "epoch": 44.00555555555555,
629
+ "grad_norm": 1.1595220565795898,
630
+ "learning_rate": 2.2213053637830016e-05,
631
+ "loss": 0.9762,
632
+ "step": 89
633
+ },
634
+ {
635
+ "epoch": 44.01111111111111,
636
+ "grad_norm": 1.511170744895935,
637
+ "learning_rate": 2.184536718926604e-05,
638
+ "loss": 0.5609,
639
+ "step": 90
640
+ },
641
+ {
642
+ "epoch": 45.00555555555555,
643
+ "grad_norm": 0.9717811346054077,
644
+ "learning_rate": 2.147705054949748e-05,
645
+ "loss": 0.9186,
646
+ "step": 91
647
+ },
648
+ {
649
+ "epoch": 45.01111111111111,
650
+ "grad_norm": 0.8693075180053711,
651
+ "learning_rate": 2.11082294983194e-05,
652
+ "loss": 0.4779,
653
+ "step": 92
654
+ },
655
+ {
656
+ "epoch": 46.00555555555555,
657
+ "grad_norm": 0.9417635202407837,
658
+ "learning_rate": 2.0739029987782903e-05,
659
+ "loss": 0.9108,
660
+ "step": 93
661
+ },
662
+ {
663
+ "epoch": 46.01111111111111,
664
+ "grad_norm": 0.8994781970977783,
665
+ "learning_rate": 2.03695780991826e-05,
666
+ "loss": 0.4855,
667
+ "step": 94
668
+ },
669
+ {
670
+ "epoch": 47.00555555555555,
671
+ "grad_norm": 1.1585125923156738,
672
+ "learning_rate": 2e-05,
673
+ "loss": 0.9531,
674
+ "step": 95
675
+ },
676
+ {
677
+ "epoch": 47.01111111111111,
678
+ "grad_norm": 0.9600257873535156,
679
+ "learning_rate": 1.9630421900817407e-05,
680
+ "loss": 0.494,
681
+ "step": 96
682
+ },
683
+ {
684
+ "epoch": 48.00555555555555,
685
+ "grad_norm": 1.2095404863357544,
686
+ "learning_rate": 1.9260970012217107e-05,
687
+ "loss": 0.8899,
688
+ "step": 97
689
+ },
690
+ {
691
+ "epoch": 48.01111111111111,
692
+ "grad_norm": 0.8960286974906921,
693
+ "learning_rate": 1.8891770501680602e-05,
694
+ "loss": 0.4368,
695
+ "step": 98
696
+ },
697
+ {
698
+ "epoch": 49.00555555555555,
699
+ "grad_norm": 0.968328595161438,
700
+ "learning_rate": 1.8522949450502522e-05,
701
+ "loss": 1.0359,
702
+ "step": 99
703
+ },
704
+ {
705
+ "epoch": 49.01111111111111,
706
+ "grad_norm": 1.0718151330947876,
707
+ "learning_rate": 1.815463281073396e-05,
708
+ "loss": 0.4903,
709
+ "step": 100
710
+ },
711
+ {
712
+ "epoch": 50.00555555555555,
713
+ "grad_norm": 1.1338093280792236,
714
+ "learning_rate": 1.7786946362169987e-05,
715
+ "loss": 0.9169,
716
+ "step": 101
717
+ },
718
+ {
719
+ "epoch": 50.01111111111111,
720
+ "grad_norm": 0.9321633577346802,
721
+ "learning_rate": 1.742001566939594e-05,
722
+ "loss": 0.4071,
723
+ "step": 102
724
+ },
725
+ {
726
+ "epoch": 51.00555555555555,
727
+ "grad_norm": 1.1823705434799194,
728
+ "learning_rate": 1.705396603890725e-05,
729
+ "loss": 0.8915,
730
+ "step": 103
731
+ },
732
+ {
733
+ "epoch": 51.01111111111111,
734
+ "grad_norm": 0.8543993234634399,
735
+ "learning_rate": 1.66889224763174e-05,
736
+ "loss": 0.4676,
737
+ "step": 104
738
+ },
739
+ {
740
+ "epoch": 52.00555555555555,
741
+ "grad_norm": 1.0003684759140015,
742
+ "learning_rate": 1.6325009643668592e-05,
743
+ "loss": 0.9276,
744
+ "step": 105
745
+ },
746
+ {
747
+ "epoch": 52.01111111111111,
748
+ "grad_norm": 0.879921555519104,
749
+ "learning_rate": 1.59623518168598e-05,
750
+ "loss": 0.4429,
751
+ "step": 106
752
+ },
753
+ {
754
+ "epoch": 53.00555555555555,
755
+ "grad_norm": 0.93624347448349,
756
+ "learning_rate": 1.5601072843206634e-05,
757
+ "loss": 0.8572,
758
+ "step": 107
759
+ },
760
+ {
761
+ "epoch": 53.01111111111111,
762
+ "grad_norm": 1.0318535566329956,
763
+ "learning_rate": 1.524129609914763e-05,
764
+ "loss": 0.4265,
765
+ "step": 108
766
+ },
767
+ {
768
+ "epoch": 54.00555555555555,
769
+ "grad_norm": 1.082194209098816,
770
+ "learning_rate": 1.4883144448111288e-05,
771
+ "loss": 0.9541,
772
+ "step": 109
773
+ },
774
+ {
775
+ "epoch": 54.01111111111111,
776
+ "grad_norm": 0.960639476776123,
777
+ "learning_rate": 1.4526740198558345e-05,
778
+ "loss": 0.401,
779
+ "step": 110
780
+ },
781
+ {
782
+ "epoch": 55.00555555555555,
783
+ "grad_norm": 1.0439647436141968,
784
+ "learning_rate": 1.417220506221351e-05,
785
+ "loss": 0.9309,
786
+ "step": 111
787
+ },
788
+ {
789
+ "epoch": 55.01111111111111,
790
+ "grad_norm": 0.982650876045227,
791
+ "learning_rate": 1.3819660112501054e-05,
792
+ "loss": 0.4102,
793
+ "step": 112
794
+ },
795
+ {
796
+ "epoch": 56.00555555555555,
797
+ "grad_norm": 0.8614187836647034,
798
+ "learning_rate": 1.3469225743198337e-05,
799
+ "loss": 0.8775,
800
+ "step": 113
801
+ },
802
+ {
803
+ "epoch": 56.01111111111111,
804
+ "grad_norm": 1.095910668373108,
805
+ "learning_rate": 1.3121021627321438e-05,
806
+ "loss": 0.435,
807
+ "step": 114
808
+ },
809
+ {
810
+ "epoch": 57.00555555555555,
811
+ "grad_norm": 0.9055922627449036,
812
+ "learning_rate": 1.2775166676256942e-05,
813
+ "loss": 0.8587,
814
+ "step": 115
815
+ },
816
+ {
817
+ "epoch": 57.01111111111111,
818
+ "grad_norm": 1.1506797075271606,
819
+ "learning_rate": 1.2431778999153796e-05,
820
+ "loss": 0.4459,
821
+ "step": 116
822
+ },
823
+ {
824
+ "epoch": 58.00555555555555,
825
+ "grad_norm": 0.9050757884979248,
826
+ "learning_rate": 1.2090975862589151e-05,
827
+ "loss": 0.8311,
828
+ "step": 117
829
+ },
830
+ {
831
+ "epoch": 58.01111111111111,
832
+ "grad_norm": 0.846161425113678,
833
+ "learning_rate": 1.1752873650521934e-05,
834
+ "loss": 0.4002,
835
+ "step": 118
836
+ },
837
+ {
838
+ "epoch": 59.00555555555555,
839
+ "grad_norm": 1.0134700536727905,
840
+ "learning_rate": 1.1417587824547822e-05,
841
+ "loss": 0.9202,
842
+ "step": 119
843
+ },
844
+ {
845
+ "epoch": 59.01111111111111,
846
+ "grad_norm": 0.8739385008811951,
847
+ "learning_rate": 1.1085232884469236e-05,
848
+ "loss": 0.4268,
849
+ "step": 120
850
+ },
851
+ {
852
+ "epoch": 60.00555555555555,
853
+ "grad_norm": 0.9328306317329407,
854
+ "learning_rate": 1.0755922329193739e-05,
855
+ "loss": 0.8797,
856
+ "step": 121
857
+ },
858
+ {
859
+ "epoch": 60.01111111111111,
860
+ "grad_norm": 1.0130794048309326,
861
+ "learning_rate": 1.0429768617974271e-05,
862
+ "loss": 0.4173,
863
+ "step": 122
864
+ },
865
+ {
866
+ "epoch": 61.00555555555555,
867
+ "grad_norm": 0.9908707737922668,
868
+ "learning_rate": 1.0106883132004428e-05,
869
+ "loss": 0.7407,
870
+ "step": 123
871
+ },
872
+ {
873
+ "epoch": 61.01111111111111,
874
+ "grad_norm": 0.7447740435600281,
875
+ "learning_rate": 9.787376136381866e-06,
876
+ "loss": 0.4214,
877
+ "step": 124
878
+ },
879
+ {
880
+ "epoch": 62.00555555555555,
881
+ "grad_norm": 1.0056525468826294,
882
+ "learning_rate": 9.471356742452881e-06,
883
+ "loss": 0.832,
884
+ "step": 125
885
+ },
886
+ {
887
+ "epoch": 62.01111111111111,
888
+ "grad_norm": 0.9604623317718506,
889
+ "learning_rate": 9.158932870551012e-06,
890
+ "loss": 0.407,
891
+ "step": 126
892
+ },
893
+ {
894
+ "epoch": 63.00555555555555,
895
+ "grad_norm": 0.8304843306541443,
896
+ "learning_rate": 8.85021121314229e-06,
897
+ "loss": 0.784,
898
+ "step": 127
899
+ },
900
+ {
901
+ "epoch": 63.01111111111111,
902
+ "grad_norm": 0.8695037961006165,
903
+ "learning_rate": 8.545297198389896e-06,
904
+ "loss": 0.4088,
905
+ "step": 128
906
+ },
907
+ {
908
+ "epoch": 64.00555555555556,
909
+ "grad_norm": 0.8668813109397888,
910
+ "learning_rate": 8.24429495415054e-06,
911
+ "loss": 0.8336,
912
+ "step": 129
913
+ },
914
+ {
915
+ "epoch": 64.0111111111111,
916
+ "grad_norm": 1.1250050067901611,
917
+ "learning_rate": 7.947307272414874e-06,
918
+ "loss": 0.4529,
919
+ "step": 130
920
+ },
921
+ {
922
+ "epoch": 65.00555555555556,
923
+ "grad_norm": 1.0635528564453125,
924
+ "learning_rate": 7.654435574204145e-06,
925
+ "loss": 0.8638,
926
+ "step": 131
927
+ },
928
+ {
929
+ "epoch": 65.0111111111111,
930
+ "grad_norm": 0.7998731732368469,
931
+ "learning_rate": 7.365779874934987e-06,
932
+ "loss": 0.3651,
933
+ "step": 132
934
+ },
935
+ {
936
+ "epoch": 66.00555555555556,
937
+ "grad_norm": 1.1683666706085205,
938
+ "learning_rate": 7.081438750264258e-06,
939
+ "loss": 0.8704,
940
+ "step": 133
941
+ },
942
+ {
943
+ "epoch": 66.0111111111111,
944
+ "grad_norm": 0.911605715751648,
945
+ "learning_rate": 6.801509302425553e-06,
946
+ "loss": 0.4264,
947
+ "step": 134
948
+ },
949
+ {
950
+ "epoch": 67.00555555555556,
951
+ "grad_norm": 0.7795685529708862,
952
+ "learning_rate": 6.526087127068857e-06,
953
+ "loss": 0.7953,
954
+ "step": 135
955
+ },
956
+ {
957
+ "epoch": 67.0111111111111,
958
+ "grad_norm": 1.0329912900924683,
959
+ "learning_rate": 6.255266280614747e-06,
960
+ "loss": 0.4025,
961
+ "step": 136
962
+ },
963
+ {
964
+ "epoch": 68.00555555555556,
965
+ "grad_norm": 1.1942418813705444,
966
+ "learning_rate": 5.989139248134181e-06,
967
+ "loss": 0.8963,
968
+ "step": 137
969
+ },
970
+ {
971
+ "epoch": 68.0111111111111,
972
+ "grad_norm": 1.1305779218673706,
973
+ "learning_rate": 5.727796911764955e-06,
974
+ "loss": 0.4206,
975
+ "step": 138
976
+ },
977
+ {
978
+ "epoch": 69.00555555555556,
979
+ "grad_norm": 0.8273850679397583,
980
+ "learning_rate": 5.471328519675521e-06,
981
+ "loss": 0.798,
982
+ "step": 139
983
+ },
984
+ {
985
+ "epoch": 69.0111111111111,
986
+ "grad_norm": 0.9187830090522766,
987
+ "learning_rate": 5.219821655586821e-06,
988
+ "loss": 0.3705,
989
+ "step": 140
990
+ },
991
+ {
992
+ "epoch": 70.00555555555556,
993
+ "grad_norm": 0.9986073970794678,
994
+ "learning_rate": 4.9733622088625335e-06,
995
+ "loss": 0.7755,
996
+ "step": 141
997
+ },
998
+ {
999
+ "epoch": 70.0111111111111,
1000
+ "grad_norm": 0.7928145527839661,
1001
+ "learning_rate": 4.732034345177941e-06,
1002
+ "loss": 0.4097,
1003
+ "step": 142
1004
+ },
1005
+ {
1006
+ "epoch": 71.00555555555556,
1007
+ "grad_norm": 1.0464165210723877,
1008
+ "learning_rate": 4.495920477777403e-06,
1009
+ "loss": 0.8362,
1010
+ "step": 143
1011
+ },
1012
+ {
1013
+ "epoch": 71.0111111111111,
1014
+ "grad_norm": 0.8963899612426758,
1015
+ "learning_rate": 4.265101239330336e-06,
1016
+ "loss": 0.4154,
1017
+ "step": 144
1018
+ },
1019
+ {
1020
+ "epoch": 72.00555555555556,
1021
+ "grad_norm": 0.9925701022148132,
1022
+ "learning_rate": 4.03965545439521e-06,
1023
+ "loss": 0.7776,
1024
+ "step": 145
1025
+ },
1026
+ {
1027
+ "epoch": 72.0111111111111,
1028
+ "grad_norm": 0.8580424785614014,
1029
+ "learning_rate": 3.819660112501053e-06,
1030
+ "loss": 0.4027,
1031
+ "step": 146
1032
+ },
1033
+ {
1034
+ "epoch": 73.00555555555556,
1035
+ "grad_norm": 0.7751114368438721,
1036
+ "learning_rate": 3.60519034185558e-06,
1037
+ "loss": 0.6957,
1038
+ "step": 147
1039
+ },
1040
+ {
1041
+ "epoch": 73.0111111111111,
1042
+ "grad_norm": 1.1692156791687012,
1043
+ "learning_rate": 3.3963193836889907e-06,
1044
+ "loss": 0.4551,
1045
+ "step": 148
1046
+ },
1047
+ {
1048
+ "epoch": 74.00555555555556,
1049
+ "grad_norm": 1.0253629684448242,
1050
+ "learning_rate": 3.193118567242148e-06,
1051
+ "loss": 0.8051,
1052
+ "step": 149
1053
+ },
1054
+ {
1055
+ "epoch": 74.0111111111111,
1056
+ "grad_norm": 0.9178398251533508,
1057
+ "learning_rate": 2.9956572854077205e-06,
1058
+ "loss": 0.4135,
1059
+ "step": 150
1060
+ },
1061
+ {
1062
+ "epoch": 75.00555555555556,
1063
+ "grad_norm": 0.8287302851676941,
1064
+ "learning_rate": 2.804002971032551e-06,
1065
+ "loss": 0.7577,
1066
+ "step": 151
1067
+ },
1068
+ {
1069
+ "epoch": 75.0111111111111,
1070
+ "grad_norm": 0.8081830143928528,
1071
+ "learning_rate": 2.618221073889433e-06,
1072
+ "loss": 0.374,
1073
+ "step": 152
1074
+ },
1075
+ {
1076
+ "epoch": 76.00555555555556,
1077
+ "grad_norm": 0.8116244673728943,
1078
+ "learning_rate": 2.4383750383260417e-06,
1079
+ "loss": 0.9303,
1080
+ "step": 153
1081
+ },
1082
+ {
1083
+ "epoch": 76.0111111111111,
1084
+ "grad_norm": 0.9563033580780029,
1085
+ "learning_rate": 2.264526281598762e-06,
1086
+ "loss": 0.3772,
1087
+ "step": 154
1088
+ },
1089
+ {
1090
+ "epoch": 77.00555555555556,
1091
+ "grad_norm": 0.9266010522842407,
1092
+ "learning_rate": 2.0967341728987554e-06,
1093
+ "loss": 0.7978,
1094
+ "step": 155
1095
+ },
1096
+ {
1097
+ "epoch": 77.0111111111111,
1098
+ "grad_norm": 0.8907399773597717,
1099
+ "learning_rate": 1.9350560130774234e-06,
1100
+ "loss": 0.3479,
1101
+ "step": 156
1102
+ },
1103
+ {
1104
+ "epoch": 78.00555555555556,
1105
+ "grad_norm": 0.8672574758529663,
1106
+ "learning_rate": 1.7795470150782312e-06,
1107
+ "loss": 0.7473,
1108
+ "step": 157
1109
+ },
1110
+ {
1111
+ "epoch": 78.0111111111111,
1112
+ "grad_norm": 0.8016725182533264,
1113
+ "learning_rate": 1.6302602850815397e-06,
1114
+ "loss": 0.3785,
1115
+ "step": 158
1116
+ },
1117
+ {
1118
+ "epoch": 79.00555555555556,
1119
+ "grad_norm": 0.8856578469276428,
1120
+ "learning_rate": 1.487246804368876e-06,
1121
+ "loss": 0.8051,
1122
+ "step": 159
1123
+ },
1124
+ {
1125
+ "epoch": 79.0111111111111,
1126
+ "grad_norm": 0.8816702961921692,
1127
+ "learning_rate": 1.3505554119128861e-06,
1128
+ "loss": 0.4356,
1129
+ "step": 160
1130
+ },
1131
+ {
1132
+ "epoch": 80.00555555555556,
1133
+ "grad_norm": 0.8219400644302368,
1134
+ "learning_rate": 1.2202327876988719e-06,
1135
+ "loss": 0.7872,
1136
+ "step": 161
1137
+ },
1138
+ {
1139
+ "epoch": 80.0111111111111,
1140
+ "grad_norm": 0.9030799269676208,
1141
+ "learning_rate": 1.0963234367836106e-06,
1142
+ "loss": 0.4088,
1143
+ "step": 162
1144
+ },
1145
+ {
1146
+ "epoch": 81.00555555555556,
1147
+ "grad_norm": 0.7858496308326721,
1148
+ "learning_rate": 9.788696740969295e-07,
1149
+ "loss": 0.7501,
1150
+ "step": 163
1151
+ },
1152
+ {
1153
+ "epoch": 81.0111111111111,
1154
+ "grad_norm": 0.8430002331733704,
1155
+ "learning_rate": 8.679116099911855e-07,
1156
+ "loss": 0.3966,
1157
+ "step": 164
1158
+ },
1159
+ {
1160
+ "epoch": 82.00555555555556,
1161
+ "grad_norm": 0.9193421602249146,
1162
+ "learning_rate": 7.634871365436192e-07,
1163
+ "loss": 0.8171,
1164
+ "step": 165
1165
+ },
1166
+ {
1167
+ "epoch": 82.0111111111111,
1168
+ "grad_norm": 0.8523241281509399,
1169
+ "learning_rate": 6.656319146162516e-07,
1170
+ "loss": 0.3779,
1171
+ "step": 166
1172
+ },
1173
+ {
1174
+ "epoch": 83.00555555555556,
1175
+ "grad_norm": 0.839622974395752,
1176
+ "learning_rate": 5.743793616777216e-07,
1177
+ "loss": 0.8187,
1178
+ "step": 167
1179
+ },
1180
+ {
1181
+ "epoch": 83.0111111111111,
1182
+ "grad_norm": 0.9939560890197754,
1183
+ "learning_rate": 4.89760640391268e-07,
1184
+ "loss": 0.4129,
1185
+ "step": 168
1186
+ },
1187
+ {
1188
+ "epoch": 84.00555555555556,
1189
+ "grad_norm": 0.9763411283493042,
1190
+ "learning_rate": 4.118046479726823e-07,
1191
+ "loss": 0.7785,
1192
+ "step": 169
1193
+ },
1194
+ {
1195
+ "epoch": 84.0111111111111,
1196
+ "grad_norm": 0.9105129837989807,
1197
+ "learning_rate": 3.4053800632196434e-07,
1198
+ "loss": 0.4133,
1199
+ "step": 170
1200
+ },
1201
+ {
1202
+ "epoch": 85.00555555555556,
1203
+ "grad_norm": 0.966189444065094,
1204
+ "learning_rate": 2.7598505293194855e-07,
1205
+ "loss": 0.7778,
1206
+ "step": 171
1207
+ },
1208
+ {
1209
+ "epoch": 85.0111111111111,
1210
+ "grad_norm": 0.9058474898338318,
1211
+ "learning_rate": 2.1816783257708084e-07,
1212
+ "loss": 0.3874,
1213
+ "step": 172
1214
+ },
1215
+ {
1216
+ "epoch": 86.00555555555556,
1217
+ "grad_norm": 0.8936800956726074,
1218
+ "learning_rate": 1.6710608978514509e-07,
1219
+ "loss": 0.8521,
1220
+ "step": 173
1221
+ },
1222
+ {
1223
+ "epoch": 86.0111111111111,
1224
+ "grad_norm": 0.936042070388794,
1225
+ "learning_rate": 1.2281726209452782e-07,
1226
+ "loss": 0.4156,
1227
+ "step": 174
1228
+ },
1229
+ {
1230
+ "epoch": 87.00555555555556,
1231
+ "grad_norm": 0.8637453317642212,
1232
+ "learning_rate": 8.531647409931065e-08,
1233
+ "loss": 0.8011,
1234
+ "step": 175
1235
+ },
1236
+ {
1237
+ "epoch": 87.0111111111111,
1238
+ "grad_norm": 0.8020222187042236,
1239
+ "learning_rate": 5.4616532284239576e-08,
1240
+ "loss": 0.3535,
1241
+ "step": 176
1242
+ },
1243
+ {
1244
+ "epoch": 88.00555555555556,
1245
+ "grad_norm": 0.8789599537849426,
1246
+ "learning_rate": 3.072792065132113e-08,
1247
+ "loss": 0.7711,
1248
+ "step": 177
1249
+ },
1250
+ {
1251
+ "epoch": 88.0111111111111,
1252
+ "grad_norm": 0.9251433610916138,
1253
+ "learning_rate": 1.365879713954188e-08,
1254
+ "loss": 0.4039,
1255
+ "step": 178
1256
+ },
1257
+ {
1258
+ "epoch": 89.00555555555556,
1259
+ "grad_norm": 0.8520329594612122,
1260
+ "learning_rate": 3.414990838945809e-09,
1261
+ "loss": 0.7674,
1262
+ "step": 179
1263
+ },
1264
+ {
1265
+ "epoch": 89.0111111111111,
1266
+ "grad_norm": 1.1043436527252197,
1267
+ "learning_rate": 0.0,
1268
+ "loss": 0.4059,
1269
+ "step": 180
1270
+ }
1271
+ ],
1272
+ "logging_steps": 1,
1273
+ "max_steps": 180,
1274
+ "num_input_tokens_seen": 0,
1275
+ "num_train_epochs": 9223372036854775807,
1276
+ "save_steps": 200,
1277
+ "stateful_callbacks": {
1278
+ "TrainerControl": {
1279
+ "args": {
1280
+ "should_epoch_stop": false,
1281
+ "should_evaluate": false,
1282
+ "should_log": false,
1283
+ "should_save": true,
1284
+ "should_training_stop": true
1285
+ },
1286
+ "attributes": {}
1287
+ }
1288
+ },
1289
+ "total_flos": 870396069937152.0,
1290
+ "train_batch_size": 2,
1291
+ "trial_name": null,
1292
+ "trial_params": null
1293
+ }
checkpoint-180/training_args.bin ADDED
@@ -0,0 +1,3 @@
 
 
 
 
1
+ version https://git-lfs.github.com/spec/v1
2
+ oid sha256:8358c760de71fead3d43b49d726f9b3f5dec2f2eb4cd58baf1f0c05914e265c2
3
+ size 6712
config.json ADDED
@@ -0,0 +1,30 @@
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
1
+ {
2
+ "_name_or_path": "Jiminiya/tiny_step1",
3
+ "architectures": [
4
+ "LlamaForCausalLM"
5
+ ],
6
+ "attention_bias": false,
7
+ "attention_dropout": 0.0,
8
+ "bos_token_id": 1,
9
+ "eos_token_id": 2,
10
+ "head_dim": 64,
11
+ "hidden_act": "silu",
12
+ "hidden_size": 2048,
13
+ "initializer_range": 0.02,
14
+ "intermediate_size": 5632,
15
+ "max_position_embeddings": 2048,
16
+ "mlp_bias": false,
17
+ "model_type": "llama",
18
+ "num_attention_heads": 32,
19
+ "num_hidden_layers": 22,
20
+ "num_key_value_heads": 4,
21
+ "pretraining_tp": 1,
22
+ "rms_norm_eps": 1e-05,
23
+ "rope_scaling": null,
24
+ "rope_theta": 10000.0,
25
+ "tie_word_embeddings": false,
26
+ "torch_dtype": "float32",
27
+ "transformers_version": "4.48.1",
28
+ "use_cache": false,
29
+ "vocab_size": 32000
30
+ }
generation_config.json ADDED
@@ -0,0 +1,8 @@
 
 
 
 
 
 
 
 
 
1
+ {
2
+ "bos_token_id": 1,
3
+ "do_sample": true,
4
+ "eos_token_id": 2,
5
+ "max_length": 2048,
6
+ "pad_token_id": 0,
7
+ "transformers_version": "4.48.1"
8
+ }
model.safetensors ADDED
@@ -0,0 +1,3 @@
 
 
 
 
1
+ version https://git-lfs.github.com/spec/v1
2
+ oid sha256:bc8a36d88735d6b99d596818ff20cd98dd222f4e4670290782996cce83e8f042
3
+ size 2462268008
special_tokens_map.json ADDED
@@ -0,0 +1,30 @@
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
1
+ {
2
+ "bos_token": {
3
+ "content": "<s>",
4
+ "lstrip": false,
5
+ "normalized": false,
6
+ "rstrip": false,
7
+ "single_word": false
8
+ },
9
+ "eos_token": {
10
+ "content": "</s>",
11
+ "lstrip": false,
12
+ "normalized": false,
13
+ "rstrip": false,
14
+ "single_word": false
15
+ },
16
+ "pad_token": {
17
+ "content": "</s>",
18
+ "lstrip": false,
19
+ "normalized": false,
20
+ "rstrip": false,
21
+ "single_word": false
22
+ },
23
+ "unk_token": {
24
+ "content": "<unk>",
25
+ "lstrip": false,
26
+ "normalized": false,
27
+ "rstrip": false,
28
+ "single_word": false
29
+ }
30
+ }
tokenizer.json ADDED
The diff for this file is too large to render. See raw diff
 
tokenizer.model ADDED
@@ -0,0 +1,3 @@
 
 
 
 
1
+ version https://git-lfs.github.com/spec/v1
2
+ oid sha256:9e556afd44213b6bd1be2b850ebbbd98f5481437a8021afaf58ee7fb1818d347
3
+ size 499723
tokenizer_config.json ADDED
@@ -0,0 +1,43 @@
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
1
+ {
2
+ "add_bos_token": true,
3
+ "add_eos_token": false,
4
+ "add_prefix_space": null,
5
+ "added_tokens_decoder": {
6
+ "0": {
7
+ "content": "<unk>",
8
+ "lstrip": false,
9
+ "normalized": false,
10
+ "rstrip": false,
11
+ "single_word": false,
12
+ "special": true
13
+ },
14
+ "1": {
15
+ "content": "<s>",
16
+ "lstrip": false,
17
+ "normalized": false,
18
+ "rstrip": false,
19
+ "single_word": false,
20
+ "special": true
21
+ },
22
+ "2": {
23
+ "content": "</s>",
24
+ "lstrip": false,
25
+ "normalized": false,
26
+ "rstrip": false,
27
+ "single_word": false,
28
+ "special": true
29
+ }
30
+ },
31
+ "bos_token": "<s>",
32
+ "clean_up_tokenization_spaces": false,
33
+ "eos_token": "</s>",
34
+ "extra_special_tokens": {},
35
+ "legacy": false,
36
+ "model_max_length": 1000000000000000019884624838656,
37
+ "pad_token": "</s>",
38
+ "padding_side": "right",
39
+ "sp_model_kwargs": {},
40
+ "tokenizer_class": "LlamaTokenizer",
41
+ "unk_token": "<unk>",
42
+ "use_default_system_prompt": false
43
+ }
training_args.bin ADDED
@@ -0,0 +1,3 @@
 
 
 
 
1
+ version https://git-lfs.github.com/spec/v1
2
+ oid sha256:8358c760de71fead3d43b49d726f9b3f5dec2f2eb4cd58baf1f0c05914e265c2
3
+ size 6712