Commit ·
9ab95b7
1
Parent(s): 7cfad99
Upload folder using huggingface_hub
Browse files- README.md +78 -0
- checkpoint-29/config.json +27 -0
- checkpoint-29/generation_config.json +7 -0
- checkpoint-29/model.safetensors +3 -0
- checkpoint-29/optimizer.pt +3 -0
- checkpoint-29/rng_state.pth +3 -0
- checkpoint-29/scheduler.pt +3 -0
- checkpoint-29/trainer_state.json +241 -0
- checkpoint-29/training_args.bin +3 -0
- checkpoint-59/config.json +27 -0
- checkpoint-59/generation_config.json +7 -0
- checkpoint-59/model.safetensors +3 -0
- checkpoint-59/optimizer.pt +3 -0
- checkpoint-59/rng_state.pth +3 -0
- checkpoint-59/scheduler.pt +3 -0
- checkpoint-59/trainer_state.json +469 -0
- checkpoint-59/training_args.bin +3 -0
- checkpoint-87/config.json +27 -0
- checkpoint-87/generation_config.json +7 -0
- checkpoint-87/model.safetensors +3 -0
- checkpoint-87/optimizer.pt +3 -0
- checkpoint-87/rng_state.pth +3 -0
- checkpoint-87/scheduler.pt +3 -0
- checkpoint-87/trainer_state.json +685 -0
- checkpoint-87/training_args.bin +3 -0
- config.json +27 -0
- generation_config.json +7 -0
- pytorch_model.bin +3 -0
- special_tokens_map.json +24 -0
- tokenizer.model +3 -0
- tokenizer_config.json +44 -0
README.md
ADDED
|
@@ -0,0 +1,78 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
---
|
| 2 |
+
license: apache-2.0
|
| 3 |
+
base_model: TinyLlama/TinyLlama-1.1B-intermediate-step-955k-token-2T
|
| 4 |
+
tags:
|
| 5 |
+
- generated_from_trainer
|
| 6 |
+
model-index:
|
| 7 |
+
- name: storage/tiny
|
| 8 |
+
results: []
|
| 9 |
+
---
|
| 10 |
+
|
| 11 |
+
<!-- This model card has been generated automatically according to the information the Trainer had access to. You
|
| 12 |
+
should probably proofread and complete it, then remove this comment. -->
|
| 13 |
+
|
| 14 |
+
[<img src="https://raw.githubusercontent.com/OpenAccess-AI-Collective/axolotl/main/image/axolotl-badge-web.png" alt="Built with Axolotl" width="200" height="32"/>](https://github.com/OpenAccess-AI-Collective/axolotl)
|
| 15 |
+
# storage/tiny
|
| 16 |
+
|
| 17 |
+
This model is a fine-tuned version of [TinyLlama/TinyLlama-1.1B-intermediate-step-955k-token-2T](https://huggingface.co/TinyLlama/TinyLlama-1.1B-intermediate-step-955k-token-2T) on the None dataset.
|
| 18 |
+
It achieves the following results on the evaluation set:
|
| 19 |
+
- Loss: 0.0598
|
| 20 |
+
|
| 21 |
+
## Model description
|
| 22 |
+
|
| 23 |
+
More information needed
|
| 24 |
+
|
| 25 |
+
## Intended uses & limitations
|
| 26 |
+
|
| 27 |
+
More information needed
|
| 28 |
+
|
| 29 |
+
## Training and evaluation data
|
| 30 |
+
|
| 31 |
+
More information needed
|
| 32 |
+
|
| 33 |
+
## Training procedure
|
| 34 |
+
|
| 35 |
+
### Training hyperparameters
|
| 36 |
+
|
| 37 |
+
The following hyperparameters were used during training:
|
| 38 |
+
- learning_rate: 0.0002
|
| 39 |
+
- train_batch_size: 8
|
| 40 |
+
- eval_batch_size: 8
|
| 41 |
+
- seed: 42
|
| 42 |
+
- gradient_accumulation_steps: 16
|
| 43 |
+
- total_train_batch_size: 128
|
| 44 |
+
- optimizer: Adam with betas=(0.9,0.999) and epsilon=1e-08
|
| 45 |
+
- lr_scheduler_type: cosine
|
| 46 |
+
- lr_scheduler_warmup_steps: 10
|
| 47 |
+
- num_epochs: 3
|
| 48 |
+
|
| 49 |
+
### Training results
|
| 50 |
+
|
| 51 |
+
| Training Loss | Epoch | Step | Validation Loss |
|
| 52 |
+
|:-------------:|:-----:|:----:|:---------------:|
|
| 53 |
+
| 1.9849 | 0.03 | 1 | 2.0811 |
|
| 54 |
+
| 1.3107 | 0.17 | 5 | 1.1992 |
|
| 55 |
+
| 0.6399 | 0.34 | 10 | 0.6359 |
|
| 56 |
+
| 0.2779 | 0.51 | 15 | 0.2862 |
|
| 57 |
+
| 0.1807 | 0.68 | 20 | 0.1634 |
|
| 58 |
+
| 0.1256 | 0.85 | 25 | 0.1177 |
|
| 59 |
+
| 0.097 | 1.0 | 30 | 0.0891 |
|
| 60 |
+
| 0.1063 | 1.17 | 35 | 0.0734 |
|
| 61 |
+
| 0.0769 | 1.34 | 40 | 0.0723 |
|
| 62 |
+
| 0.0694 | 1.51 | 45 | 0.0633 |
|
| 63 |
+
| 0.0687 | 1.69 | 50 | 0.0624 |
|
| 64 |
+
| 0.0575 | 1.86 | 55 | 0.0622 |
|
| 65 |
+
| 0.0516 | 2.01 | 60 | 0.0609 |
|
| 66 |
+
| 0.0582 | 2.18 | 65 | 0.0603 |
|
| 67 |
+
| 0.0611 | 2.35 | 70 | 0.0600 |
|
| 68 |
+
| 0.0515 | 2.52 | 75 | 0.0598 |
|
| 69 |
+
| 0.0704 | 2.69 | 80 | 0.0598 |
|
| 70 |
+
| 0.0525 | 2.86 | 85 | 0.0598 |
|
| 71 |
+
|
| 72 |
+
|
| 73 |
+
### Framework versions
|
| 74 |
+
|
| 75 |
+
- Transformers 4.35.2
|
| 76 |
+
- Pytorch 2.0.1+cu118
|
| 77 |
+
- Datasets 2.15.0
|
| 78 |
+
- Tokenizers 0.15.0
|
checkpoint-29/config.json
ADDED
|
@@ -0,0 +1,27 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
{
|
| 2 |
+
"_name_or_path": "TinyLlama/TinyLlama-1.1B-intermediate-step-955k-token-2T",
|
| 3 |
+
"architectures": [
|
| 4 |
+
"LlamaForCausalLM"
|
| 5 |
+
],
|
| 6 |
+
"attention_bias": false,
|
| 7 |
+
"bos_token_id": 1,
|
| 8 |
+
"eos_token_id": 2,
|
| 9 |
+
"hidden_act": "silu",
|
| 10 |
+
"hidden_size": 2048,
|
| 11 |
+
"initializer_range": 0.02,
|
| 12 |
+
"intermediate_size": 5632,
|
| 13 |
+
"max_position_embeddings": 8192,
|
| 14 |
+
"model_type": "llama",
|
| 15 |
+
"num_attention_heads": 32,
|
| 16 |
+
"num_hidden_layers": 22,
|
| 17 |
+
"num_key_value_heads": 4,
|
| 18 |
+
"pretraining_tp": 1,
|
| 19 |
+
"rms_norm_eps": 1e-05,
|
| 20 |
+
"rope_scaling": null,
|
| 21 |
+
"rope_theta": 10000.0,
|
| 22 |
+
"tie_word_embeddings": false,
|
| 23 |
+
"torch_dtype": "bfloat16",
|
| 24 |
+
"transformers_version": "4.35.2",
|
| 25 |
+
"use_cache": false,
|
| 26 |
+
"vocab_size": 32000
|
| 27 |
+
}
|
checkpoint-29/generation_config.json
ADDED
|
@@ -0,0 +1,7 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
{
|
| 2 |
+
"bos_token_id": 1,
|
| 3 |
+
"eos_token_id": 2,
|
| 4 |
+
"max_length": 2048,
|
| 5 |
+
"pad_token_id": 0,
|
| 6 |
+
"transformers_version": "4.35.2"
|
| 7 |
+
}
|
checkpoint-29/model.safetensors
ADDED
|
@@ -0,0 +1,3 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
version https://git-lfs.github.com/spec/v1
|
| 2 |
+
oid sha256:ce30d70157c1afd5d187c848ef4bc5cda73e13a011a0e2cc3f2ac9ba7a6a3619
|
| 3 |
+
size 2200119864
|
checkpoint-29/optimizer.pt
ADDED
|
@@ -0,0 +1,3 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
version https://git-lfs.github.com/spec/v1
|
| 2 |
+
oid sha256:5c1d49f8c93f1bdd56d22fdc6f716e08e69f6efff4b62b1006efe4c5319ec0f0
|
| 3 |
+
size 2205157317
|
checkpoint-29/rng_state.pth
ADDED
|
@@ -0,0 +1,3 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
version https://git-lfs.github.com/spec/v1
|
| 2 |
+
oid sha256:8bb3860d98363e44bc4704fb9458deec927fd96997d1d09c80639070db959d22
|
| 3 |
+
size 14575
|
checkpoint-29/scheduler.pt
ADDED
|
@@ -0,0 +1,3 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
version https://git-lfs.github.com/spec/v1
|
| 2 |
+
oid sha256:ffb75497d89dde753ef1288237c3b10689f61361b980e7a4322d13a99124626f
|
| 3 |
+
size 627
|
checkpoint-29/trainer_state.json
ADDED
|
@@ -0,0 +1,241 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
{
|
| 2 |
+
"best_metric": null,
|
| 3 |
+
"best_model_checkpoint": null,
|
| 4 |
+
"epoch": 0.9872340425531915,
|
| 5 |
+
"eval_steps": 5,
|
| 6 |
+
"global_step": 29,
|
| 7 |
+
"is_hyper_param_search": false,
|
| 8 |
+
"is_local_process_zero": true,
|
| 9 |
+
"is_world_process_zero": true,
|
| 10 |
+
"log_history": [
|
| 11 |
+
{
|
| 12 |
+
"epoch": 0.03,
|
| 13 |
+
"learning_rate": 2e-05,
|
| 14 |
+
"loss": 1.9849,
|
| 15 |
+
"step": 1
|
| 16 |
+
},
|
| 17 |
+
{
|
| 18 |
+
"epoch": 0.03,
|
| 19 |
+
"eval_loss": 2.0810964107513428,
|
| 20 |
+
"eval_runtime": 6.2278,
|
| 21 |
+
"eval_samples_per_second": 44.96,
|
| 22 |
+
"eval_steps_per_second": 5.62,
|
| 23 |
+
"step": 1
|
| 24 |
+
},
|
| 25 |
+
{
|
| 26 |
+
"epoch": 0.07,
|
| 27 |
+
"learning_rate": 4e-05,
|
| 28 |
+
"loss": 1.8931,
|
| 29 |
+
"step": 2
|
| 30 |
+
},
|
| 31 |
+
{
|
| 32 |
+
"epoch": 0.1,
|
| 33 |
+
"learning_rate": 6e-05,
|
| 34 |
+
"loss": 1.763,
|
| 35 |
+
"step": 3
|
| 36 |
+
},
|
| 37 |
+
{
|
| 38 |
+
"epoch": 0.14,
|
| 39 |
+
"learning_rate": 8e-05,
|
| 40 |
+
"loss": 1.6688,
|
| 41 |
+
"step": 4
|
| 42 |
+
},
|
| 43 |
+
{
|
| 44 |
+
"epoch": 0.17,
|
| 45 |
+
"learning_rate": 0.0001,
|
| 46 |
+
"loss": 1.3107,
|
| 47 |
+
"step": 5
|
| 48 |
+
},
|
| 49 |
+
{
|
| 50 |
+
"epoch": 0.17,
|
| 51 |
+
"eval_loss": 1.1991726160049438,
|
| 52 |
+
"eval_runtime": 6.3173,
|
| 53 |
+
"eval_samples_per_second": 44.322,
|
| 54 |
+
"eval_steps_per_second": 5.54,
|
| 55 |
+
"step": 5
|
| 56 |
+
},
|
| 57 |
+
{
|
| 58 |
+
"epoch": 0.2,
|
| 59 |
+
"learning_rate": 0.00012,
|
| 60 |
+
"loss": 1.1899,
|
| 61 |
+
"step": 6
|
| 62 |
+
},
|
| 63 |
+
{
|
| 64 |
+
"epoch": 0.24,
|
| 65 |
+
"learning_rate": 0.00014,
|
| 66 |
+
"loss": 1.041,
|
| 67 |
+
"step": 7
|
| 68 |
+
},
|
| 69 |
+
{
|
| 70 |
+
"epoch": 0.27,
|
| 71 |
+
"learning_rate": 0.00016,
|
| 72 |
+
"loss": 0.9779,
|
| 73 |
+
"step": 8
|
| 74 |
+
},
|
| 75 |
+
{
|
| 76 |
+
"epoch": 0.31,
|
| 77 |
+
"learning_rate": 0.00018,
|
| 78 |
+
"loss": 0.7671,
|
| 79 |
+
"step": 9
|
| 80 |
+
},
|
| 81 |
+
{
|
| 82 |
+
"epoch": 0.34,
|
| 83 |
+
"learning_rate": 0.0002,
|
| 84 |
+
"loss": 0.6399,
|
| 85 |
+
"step": 10
|
| 86 |
+
},
|
| 87 |
+
{
|
| 88 |
+
"epoch": 0.34,
|
| 89 |
+
"eval_loss": 0.6359391212463379,
|
| 90 |
+
"eval_runtime": 6.2721,
|
| 91 |
+
"eval_samples_per_second": 44.642,
|
| 92 |
+
"eval_steps_per_second": 5.58,
|
| 93 |
+
"step": 10
|
| 94 |
+
},
|
| 95 |
+
{
|
| 96 |
+
"epoch": 0.37,
|
| 97 |
+
"learning_rate": 0.0001999167799344583,
|
| 98 |
+
"loss": 0.6777,
|
| 99 |
+
"step": 11
|
| 100 |
+
},
|
| 101 |
+
{
|
| 102 |
+
"epoch": 0.41,
|
| 103 |
+
"learning_rate": 0.00019966725824941932,
|
| 104 |
+
"loss": 0.492,
|
| 105 |
+
"step": 12
|
| 106 |
+
},
|
| 107 |
+
{
|
| 108 |
+
"epoch": 0.44,
|
| 109 |
+
"learning_rate": 0.00019925185024910277,
|
| 110 |
+
"loss": 0.3781,
|
| 111 |
+
"step": 13
|
| 112 |
+
},
|
| 113 |
+
{
|
| 114 |
+
"epoch": 0.48,
|
| 115 |
+
"learning_rate": 0.0001986712473391289,
|
| 116 |
+
"loss": 0.4503,
|
| 117 |
+
"step": 14
|
| 118 |
+
},
|
| 119 |
+
{
|
| 120 |
+
"epoch": 0.51,
|
| 121 |
+
"learning_rate": 0.00019792641587574212,
|
| 122 |
+
"loss": 0.2779,
|
| 123 |
+
"step": 15
|
| 124 |
+
},
|
| 125 |
+
{
|
| 126 |
+
"epoch": 0.51,
|
| 127 |
+
"eval_loss": 0.28619876503944397,
|
| 128 |
+
"eval_runtime": 6.3023,
|
| 129 |
+
"eval_samples_per_second": 44.428,
|
| 130 |
+
"eval_steps_per_second": 5.554,
|
| 131 |
+
"step": 15
|
| 132 |
+
},
|
| 133 |
+
{
|
| 134 |
+
"epoch": 0.54,
|
| 135 |
+
"learning_rate": 0.00019701859555740648,
|
| 136 |
+
"loss": 0.2658,
|
| 137 |
+
"step": 16
|
| 138 |
+
},
|
| 139 |
+
{
|
| 140 |
+
"epoch": 0.58,
|
| 141 |
+
"learning_rate": 0.00019594929736144976,
|
| 142 |
+
"loss": 0.2282,
|
| 143 |
+
"step": 17
|
| 144 |
+
},
|
| 145 |
+
{
|
| 146 |
+
"epoch": 0.61,
|
| 147 |
+
"learning_rate": 0.000194720301029191,
|
| 148 |
+
"loss": 0.2149,
|
| 149 |
+
"step": 18
|
| 150 |
+
},
|
| 151 |
+
{
|
| 152 |
+
"epoch": 0.65,
|
| 153 |
+
"learning_rate": 0.0001933336521037367,
|
| 154 |
+
"loss": 0.1792,
|
| 155 |
+
"step": 19
|
| 156 |
+
},
|
| 157 |
+
{
|
| 158 |
+
"epoch": 0.68,
|
| 159 |
+
"learning_rate": 0.00019179165852537596,
|
| 160 |
+
"loss": 0.1807,
|
| 161 |
+
"step": 20
|
| 162 |
+
},
|
| 163 |
+
{
|
| 164 |
+
"epoch": 0.68,
|
| 165 |
+
"eval_loss": 0.1634436696767807,
|
| 166 |
+
"eval_runtime": 6.2724,
|
| 167 |
+
"eval_samples_per_second": 44.64,
|
| 168 |
+
"eval_steps_per_second": 5.58,
|
| 169 |
+
"step": 20
|
| 170 |
+
},
|
| 171 |
+
{
|
| 172 |
+
"epoch": 0.71,
|
| 173 |
+
"learning_rate": 0.0001900968867902419,
|
| 174 |
+
"loss": 0.1425,
|
| 175 |
+
"step": 21
|
| 176 |
+
},
|
| 177 |
+
{
|
| 178 |
+
"epoch": 0.75,
|
| 179 |
+
"learning_rate": 0.00018825215767863214,
|
| 180 |
+
"loss": 0.1377,
|
| 181 |
+
"step": 22
|
| 182 |
+
},
|
| 183 |
+
{
|
| 184 |
+
"epoch": 0.78,
|
| 185 |
+
"learning_rate": 0.00018626054156009806,
|
| 186 |
+
"loss": 0.1173,
|
| 187 |
+
"step": 23
|
| 188 |
+
},
|
| 189 |
+
{
|
| 190 |
+
"epoch": 0.82,
|
| 191 |
+
"learning_rate": 0.00018412535328311814,
|
| 192 |
+
"loss": 0.1338,
|
| 193 |
+
"step": 24
|
| 194 |
+
},
|
| 195 |
+
{
|
| 196 |
+
"epoch": 0.85,
|
| 197 |
+
"learning_rate": 0.00018185014665785936,
|
| 198 |
+
"loss": 0.1256,
|
| 199 |
+
"step": 25
|
| 200 |
+
},
|
| 201 |
+
{
|
| 202 |
+
"epoch": 0.85,
|
| 203 |
+
"eval_loss": 0.1177145466208458,
|
| 204 |
+
"eval_runtime": 6.2558,
|
| 205 |
+
"eval_samples_per_second": 44.758,
|
| 206 |
+
"eval_steps_per_second": 5.595,
|
| 207 |
+
"step": 25
|
| 208 |
+
},
|
| 209 |
+
{
|
| 210 |
+
"epoch": 0.89,
|
| 211 |
+
"learning_rate": 0.00017943870854121124,
|
| 212 |
+
"loss": 0.1234,
|
| 213 |
+
"step": 26
|
| 214 |
+
},
|
| 215 |
+
{
|
| 216 |
+
"epoch": 0.92,
|
| 217 |
+
"learning_rate": 0.0001768950525339362,
|
| 218 |
+
"loss": 0.1164,
|
| 219 |
+
"step": 27
|
| 220 |
+
},
|
| 221 |
+
{
|
| 222 |
+
"epoch": 0.95,
|
| 223 |
+
"learning_rate": 0.000174223412300427,
|
| 224 |
+
"loss": 0.1077,
|
| 225 |
+
"step": 28
|
| 226 |
+
},
|
| 227 |
+
{
|
| 228 |
+
"epoch": 0.99,
|
| 229 |
+
"learning_rate": 0.00017142823452219038,
|
| 230 |
+
"loss": 0.0877,
|
| 231 |
+
"step": 29
|
| 232 |
+
}
|
| 233 |
+
],
|
| 234 |
+
"logging_steps": 1,
|
| 235 |
+
"max_steps": 87,
|
| 236 |
+
"num_train_epochs": 3,
|
| 237 |
+
"save_steps": 500,
|
| 238 |
+
"total_flos": 1.9186990286320435e+17,
|
| 239 |
+
"trial_name": null,
|
| 240 |
+
"trial_params": null
|
| 241 |
+
}
|
checkpoint-29/training_args.bin
ADDED
|
@@ -0,0 +1,3 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
version https://git-lfs.github.com/spec/v1
|
| 2 |
+
oid sha256:5117945171d6dce6bb2ba0c56f263b94b8a6a7b57ffd9e74113faed0e7efe401
|
| 3 |
+
size 4539
|
checkpoint-59/config.json
ADDED
|
@@ -0,0 +1,27 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
{
|
| 2 |
+
"_name_or_path": "TinyLlama/TinyLlama-1.1B-intermediate-step-955k-token-2T",
|
| 3 |
+
"architectures": [
|
| 4 |
+
"LlamaForCausalLM"
|
| 5 |
+
],
|
| 6 |
+
"attention_bias": false,
|
| 7 |
+
"bos_token_id": 1,
|
| 8 |
+
"eos_token_id": 2,
|
| 9 |
+
"hidden_act": "silu",
|
| 10 |
+
"hidden_size": 2048,
|
| 11 |
+
"initializer_range": 0.02,
|
| 12 |
+
"intermediate_size": 5632,
|
| 13 |
+
"max_position_embeddings": 8192,
|
| 14 |
+
"model_type": "llama",
|
| 15 |
+
"num_attention_heads": 32,
|
| 16 |
+
"num_hidden_layers": 22,
|
| 17 |
+
"num_key_value_heads": 4,
|
| 18 |
+
"pretraining_tp": 1,
|
| 19 |
+
"rms_norm_eps": 1e-05,
|
| 20 |
+
"rope_scaling": null,
|
| 21 |
+
"rope_theta": 10000.0,
|
| 22 |
+
"tie_word_embeddings": false,
|
| 23 |
+
"torch_dtype": "bfloat16",
|
| 24 |
+
"transformers_version": "4.35.2",
|
| 25 |
+
"use_cache": false,
|
| 26 |
+
"vocab_size": 32000
|
| 27 |
+
}
|
checkpoint-59/generation_config.json
ADDED
|
@@ -0,0 +1,7 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
{
|
| 2 |
+
"bos_token_id": 1,
|
| 3 |
+
"eos_token_id": 2,
|
| 4 |
+
"max_length": 2048,
|
| 5 |
+
"pad_token_id": 0,
|
| 6 |
+
"transformers_version": "4.35.2"
|
| 7 |
+
}
|
checkpoint-59/model.safetensors
ADDED
|
@@ -0,0 +1,3 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
version https://git-lfs.github.com/spec/v1
|
| 2 |
+
oid sha256:2349616810da5867bf9a9a1c2c239c53ea71828815de61c84d53632b5d405410
|
| 3 |
+
size 2200119864
|
checkpoint-59/optimizer.pt
ADDED
|
@@ -0,0 +1,3 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
version https://git-lfs.github.com/spec/v1
|
| 2 |
+
oid sha256:51a64c1aee4120ba807c02c5f7ea6c6f73e4880d2f0e8e58e9b964904fe20a4a
|
| 3 |
+
size 2205157317
|
checkpoint-59/rng_state.pth
ADDED
|
@@ -0,0 +1,3 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
version https://git-lfs.github.com/spec/v1
|
| 2 |
+
oid sha256:f0e0f30c7754637247f97149558697a4932915f961818f7360cfc62855116ca0
|
| 3 |
+
size 14575
|
checkpoint-59/scheduler.pt
ADDED
|
@@ -0,0 +1,3 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
version https://git-lfs.github.com/spec/v1
|
| 2 |
+
oid sha256:c5f2d8935dc19752cea3bf0086ddc17cf6bfd4ac366c0c8eef9234ff4d799cf3
|
| 3 |
+
size 627
|
checkpoint-59/trainer_state.json
ADDED
|
@@ -0,0 +1,469 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
{
|
| 2 |
+
"best_metric": null,
|
| 3 |
+
"best_model_checkpoint": null,
|
| 4 |
+
"epoch": 1.9914893617021276,
|
| 5 |
+
"eval_steps": 5,
|
| 6 |
+
"global_step": 59,
|
| 7 |
+
"is_hyper_param_search": false,
|
| 8 |
+
"is_local_process_zero": true,
|
| 9 |
+
"is_world_process_zero": true,
|
| 10 |
+
"log_history": [
|
| 11 |
+
{
|
| 12 |
+
"epoch": 0.03,
|
| 13 |
+
"learning_rate": 2e-05,
|
| 14 |
+
"loss": 1.9849,
|
| 15 |
+
"step": 1
|
| 16 |
+
},
|
| 17 |
+
{
|
| 18 |
+
"epoch": 0.03,
|
| 19 |
+
"eval_loss": 2.0810964107513428,
|
| 20 |
+
"eval_runtime": 6.2278,
|
| 21 |
+
"eval_samples_per_second": 44.96,
|
| 22 |
+
"eval_steps_per_second": 5.62,
|
| 23 |
+
"step": 1
|
| 24 |
+
},
|
| 25 |
+
{
|
| 26 |
+
"epoch": 0.07,
|
| 27 |
+
"learning_rate": 4e-05,
|
| 28 |
+
"loss": 1.8931,
|
| 29 |
+
"step": 2
|
| 30 |
+
},
|
| 31 |
+
{
|
| 32 |
+
"epoch": 0.1,
|
| 33 |
+
"learning_rate": 6e-05,
|
| 34 |
+
"loss": 1.763,
|
| 35 |
+
"step": 3
|
| 36 |
+
},
|
| 37 |
+
{
|
| 38 |
+
"epoch": 0.14,
|
| 39 |
+
"learning_rate": 8e-05,
|
| 40 |
+
"loss": 1.6688,
|
| 41 |
+
"step": 4
|
| 42 |
+
},
|
| 43 |
+
{
|
| 44 |
+
"epoch": 0.17,
|
| 45 |
+
"learning_rate": 0.0001,
|
| 46 |
+
"loss": 1.3107,
|
| 47 |
+
"step": 5
|
| 48 |
+
},
|
| 49 |
+
{
|
| 50 |
+
"epoch": 0.17,
|
| 51 |
+
"eval_loss": 1.1991726160049438,
|
| 52 |
+
"eval_runtime": 6.3173,
|
| 53 |
+
"eval_samples_per_second": 44.322,
|
| 54 |
+
"eval_steps_per_second": 5.54,
|
| 55 |
+
"step": 5
|
| 56 |
+
},
|
| 57 |
+
{
|
| 58 |
+
"epoch": 0.2,
|
| 59 |
+
"learning_rate": 0.00012,
|
| 60 |
+
"loss": 1.1899,
|
| 61 |
+
"step": 6
|
| 62 |
+
},
|
| 63 |
+
{
|
| 64 |
+
"epoch": 0.24,
|
| 65 |
+
"learning_rate": 0.00014,
|
| 66 |
+
"loss": 1.041,
|
| 67 |
+
"step": 7
|
| 68 |
+
},
|
| 69 |
+
{
|
| 70 |
+
"epoch": 0.27,
|
| 71 |
+
"learning_rate": 0.00016,
|
| 72 |
+
"loss": 0.9779,
|
| 73 |
+
"step": 8
|
| 74 |
+
},
|
| 75 |
+
{
|
| 76 |
+
"epoch": 0.31,
|
| 77 |
+
"learning_rate": 0.00018,
|
| 78 |
+
"loss": 0.7671,
|
| 79 |
+
"step": 9
|
| 80 |
+
},
|
| 81 |
+
{
|
| 82 |
+
"epoch": 0.34,
|
| 83 |
+
"learning_rate": 0.0002,
|
| 84 |
+
"loss": 0.6399,
|
| 85 |
+
"step": 10
|
| 86 |
+
},
|
| 87 |
+
{
|
| 88 |
+
"epoch": 0.34,
|
| 89 |
+
"eval_loss": 0.6359391212463379,
|
| 90 |
+
"eval_runtime": 6.2721,
|
| 91 |
+
"eval_samples_per_second": 44.642,
|
| 92 |
+
"eval_steps_per_second": 5.58,
|
| 93 |
+
"step": 10
|
| 94 |
+
},
|
| 95 |
+
{
|
| 96 |
+
"epoch": 0.37,
|
| 97 |
+
"learning_rate": 0.0001999167799344583,
|
| 98 |
+
"loss": 0.6777,
|
| 99 |
+
"step": 11
|
| 100 |
+
},
|
| 101 |
+
{
|
| 102 |
+
"epoch": 0.41,
|
| 103 |
+
"learning_rate": 0.00019966725824941932,
|
| 104 |
+
"loss": 0.492,
|
| 105 |
+
"step": 12
|
| 106 |
+
},
|
| 107 |
+
{
|
| 108 |
+
"epoch": 0.44,
|
| 109 |
+
"learning_rate": 0.00019925185024910277,
|
| 110 |
+
"loss": 0.3781,
|
| 111 |
+
"step": 13
|
| 112 |
+
},
|
| 113 |
+
{
|
| 114 |
+
"epoch": 0.48,
|
| 115 |
+
"learning_rate": 0.0001986712473391289,
|
| 116 |
+
"loss": 0.4503,
|
| 117 |
+
"step": 14
|
| 118 |
+
},
|
| 119 |
+
{
|
| 120 |
+
"epoch": 0.51,
|
| 121 |
+
"learning_rate": 0.00019792641587574212,
|
| 122 |
+
"loss": 0.2779,
|
| 123 |
+
"step": 15
|
| 124 |
+
},
|
| 125 |
+
{
|
| 126 |
+
"epoch": 0.51,
|
| 127 |
+
"eval_loss": 0.28619876503944397,
|
| 128 |
+
"eval_runtime": 6.3023,
|
| 129 |
+
"eval_samples_per_second": 44.428,
|
| 130 |
+
"eval_steps_per_second": 5.554,
|
| 131 |
+
"step": 15
|
| 132 |
+
},
|
| 133 |
+
{
|
| 134 |
+
"epoch": 0.54,
|
| 135 |
+
"learning_rate": 0.00019701859555740648,
|
| 136 |
+
"loss": 0.2658,
|
| 137 |
+
"step": 16
|
| 138 |
+
},
|
| 139 |
+
{
|
| 140 |
+
"epoch": 0.58,
|
| 141 |
+
"learning_rate": 0.00019594929736144976,
|
| 142 |
+
"loss": 0.2282,
|
| 143 |
+
"step": 17
|
| 144 |
+
},
|
| 145 |
+
{
|
| 146 |
+
"epoch": 0.61,
|
| 147 |
+
"learning_rate": 0.000194720301029191,
|
| 148 |
+
"loss": 0.2149,
|
| 149 |
+
"step": 18
|
| 150 |
+
},
|
| 151 |
+
{
|
| 152 |
+
"epoch": 0.65,
|
| 153 |
+
"learning_rate": 0.0001933336521037367,
|
| 154 |
+
"loss": 0.1792,
|
| 155 |
+
"step": 19
|
| 156 |
+
},
|
| 157 |
+
{
|
| 158 |
+
"epoch": 0.68,
|
| 159 |
+
"learning_rate": 0.00019179165852537596,
|
| 160 |
+
"loss": 0.1807,
|
| 161 |
+
"step": 20
|
| 162 |
+
},
|
| 163 |
+
{
|
| 164 |
+
"epoch": 0.68,
|
| 165 |
+
"eval_loss": 0.1634436696767807,
|
| 166 |
+
"eval_runtime": 6.2724,
|
| 167 |
+
"eval_samples_per_second": 44.64,
|
| 168 |
+
"eval_steps_per_second": 5.58,
|
| 169 |
+
"step": 20
|
| 170 |
+
},
|
| 171 |
+
{
|
| 172 |
+
"epoch": 0.71,
|
| 173 |
+
"learning_rate": 0.0001900968867902419,
|
| 174 |
+
"loss": 0.1425,
|
| 175 |
+
"step": 21
|
| 176 |
+
},
|
| 177 |
+
{
|
| 178 |
+
"epoch": 0.75,
|
| 179 |
+
"learning_rate": 0.00018825215767863214,
|
| 180 |
+
"loss": 0.1377,
|
| 181 |
+
"step": 22
|
| 182 |
+
},
|
| 183 |
+
{
|
| 184 |
+
"epoch": 0.78,
|
| 185 |
+
"learning_rate": 0.00018626054156009806,
|
| 186 |
+
"loss": 0.1173,
|
| 187 |
+
"step": 23
|
| 188 |
+
},
|
| 189 |
+
{
|
| 190 |
+
"epoch": 0.82,
|
| 191 |
+
"learning_rate": 0.00018412535328311814,
|
| 192 |
+
"loss": 0.1338,
|
| 193 |
+
"step": 24
|
| 194 |
+
},
|
| 195 |
+
{
|
| 196 |
+
"epoch": 0.85,
|
| 197 |
+
"learning_rate": 0.00018185014665785936,
|
| 198 |
+
"loss": 0.1256,
|
| 199 |
+
"step": 25
|
| 200 |
+
},
|
| 201 |
+
{
|
| 202 |
+
"epoch": 0.85,
|
| 203 |
+
"eval_loss": 0.1177145466208458,
|
| 204 |
+
"eval_runtime": 6.2558,
|
| 205 |
+
"eval_samples_per_second": 44.758,
|
| 206 |
+
"eval_steps_per_second": 5.595,
|
| 207 |
+
"step": 25
|
| 208 |
+
},
|
| 209 |
+
{
|
| 210 |
+
"epoch": 0.89,
|
| 211 |
+
"learning_rate": 0.00017943870854121124,
|
| 212 |
+
"loss": 0.1234,
|
| 213 |
+
"step": 26
|
| 214 |
+
},
|
| 215 |
+
{
|
| 216 |
+
"epoch": 0.92,
|
| 217 |
+
"learning_rate": 0.0001768950525339362,
|
| 218 |
+
"loss": 0.1164,
|
| 219 |
+
"step": 27
|
| 220 |
+
},
|
| 221 |
+
{
|
| 222 |
+
"epoch": 0.95,
|
| 223 |
+
"learning_rate": 0.000174223412300427,
|
| 224 |
+
"loss": 0.1077,
|
| 225 |
+
"step": 28
|
| 226 |
+
},
|
| 227 |
+
{
|
| 228 |
+
"epoch": 0.99,
|
| 229 |
+
"learning_rate": 0.00017142823452219038,
|
| 230 |
+
"loss": 0.0877,
|
| 231 |
+
"step": 29
|
| 232 |
+
},
|
| 233 |
+
{
|
| 234 |
+
"epoch": 1.0,
|
| 235 |
+
"learning_rate": 0.00016851417149678444,
|
| 236 |
+
"loss": 0.097,
|
| 237 |
+
"step": 30
|
| 238 |
+
},
|
| 239 |
+
{
|
| 240 |
+
"epoch": 1.0,
|
| 241 |
+
"eval_loss": 0.08908110111951828,
|
| 242 |
+
"eval_runtime": 6.274,
|
| 243 |
+
"eval_samples_per_second": 44.628,
|
| 244 |
+
"eval_steps_per_second": 5.579,
|
| 245 |
+
"step": 30
|
| 246 |
+
},
|
| 247 |
+
{
|
| 248 |
+
"epoch": 1.04,
|
| 249 |
+
"learning_rate": 0.00016548607339452853,
|
| 250 |
+
"loss": 0.0741,
|
| 251 |
+
"step": 31
|
| 252 |
+
},
|
| 253 |
+
{
|
| 254 |
+
"epoch": 1.07,
|
| 255 |
+
"learning_rate": 0.00016234898018587337,
|
| 256 |
+
"loss": 0.0811,
|
| 257 |
+
"step": 32
|
| 258 |
+
},
|
| 259 |
+
{
|
| 260 |
+
"epoch": 1.11,
|
| 261 |
+
"learning_rate": 0.00015910811325286768,
|
| 262 |
+
"loss": 0.0743,
|
| 263 |
+
"step": 33
|
| 264 |
+
},
|
| 265 |
+
{
|
| 266 |
+
"epoch": 1.14,
|
| 267 |
+
"learning_rate": 0.00015576886669868296,
|
| 268 |
+
"loss": 0.0799,
|
| 269 |
+
"step": 34
|
| 270 |
+
},
|
| 271 |
+
{
|
| 272 |
+
"epoch": 1.17,
|
| 273 |
+
"learning_rate": 0.00015233679836966122,
|
| 274 |
+
"loss": 0.1063,
|
| 275 |
+
"step": 35
|
| 276 |
+
},
|
| 277 |
+
{
|
| 278 |
+
"epoch": 1.17,
|
| 279 |
+
"eval_loss": 0.07344743609428406,
|
| 280 |
+
"eval_runtime": 6.3055,
|
| 281 |
+
"eval_samples_per_second": 44.406,
|
| 282 |
+
"eval_steps_per_second": 5.551,
|
| 283 |
+
"step": 35
|
| 284 |
+
},
|
| 285 |
+
{
|
| 286 |
+
"epoch": 1.21,
|
| 287 |
+
"learning_rate": 0.00014881762060482814,
|
| 288 |
+
"loss": 0.0835,
|
| 289 |
+
"step": 36
|
| 290 |
+
},
|
| 291 |
+
{
|
| 292 |
+
"epoch": 1.24,
|
| 293 |
+
"learning_rate": 0.00014521719072826858,
|
| 294 |
+
"loss": 0.0661,
|
| 295 |
+
"step": 37
|
| 296 |
+
},
|
| 297 |
+
{
|
| 298 |
+
"epoch": 1.28,
|
| 299 |
+
"learning_rate": 0.00014154150130018866,
|
| 300 |
+
"loss": 0.0637,
|
| 301 |
+
"step": 38
|
| 302 |
+
},
|
| 303 |
+
{
|
| 304 |
+
"epoch": 1.31,
|
| 305 |
+
"learning_rate": 0.00013779667014289065,
|
| 306 |
+
"loss": 0.0642,
|
| 307 |
+
"step": 39
|
| 308 |
+
},
|
| 309 |
+
{
|
| 310 |
+
"epoch": 1.34,
|
| 311 |
+
"learning_rate": 0.00013398893015826167,
|
| 312 |
+
"loss": 0.0769,
|
| 313 |
+
"step": 40
|
| 314 |
+
},
|
| 315 |
+
{
|
| 316 |
+
"epoch": 1.34,
|
| 317 |
+
"eval_loss": 0.07225509732961655,
|
| 318 |
+
"eval_runtime": 6.2831,
|
| 319 |
+
"eval_samples_per_second": 44.564,
|
| 320 |
+
"eval_steps_per_second": 5.571,
|
| 321 |
+
"step": 40
|
| 322 |
+
},
|
| 323 |
+
{
|
| 324 |
+
"epoch": 1.38,
|
| 325 |
+
"learning_rate": 0.00013012461895372344,
|
| 326 |
+
"loss": 0.0746,
|
| 327 |
+
"step": 41
|
| 328 |
+
},
|
| 329 |
+
{
|
| 330 |
+
"epoch": 1.41,
|
| 331 |
+
"learning_rate": 0.00012621016829391022,
|
| 332 |
+
"loss": 0.0592,
|
| 333 |
+
"step": 42
|
| 334 |
+
},
|
| 335 |
+
{
|
| 336 |
+
"epoch": 1.45,
|
| 337 |
+
"learning_rate": 0.00012225209339563145,
|
| 338 |
+
"loss": 0.078,
|
| 339 |
+
"step": 43
|
| 340 |
+
},
|
| 341 |
+
{
|
| 342 |
+
"epoch": 1.48,
|
| 343 |
+
"learning_rate": 0.00011825698208393619,
|
| 344 |
+
"loss": 0.088,
|
| 345 |
+
"step": 44
|
| 346 |
+
},
|
| 347 |
+
{
|
| 348 |
+
"epoch": 1.51,
|
| 349 |
+
"learning_rate": 0.00011423148382732853,
|
| 350 |
+
"loss": 0.0694,
|
| 351 |
+
"step": 45
|
| 352 |
+
},
|
| 353 |
+
{
|
| 354 |
+
"epoch": 1.51,
|
| 355 |
+
"eval_loss": 0.06333146244287491,
|
| 356 |
+
"eval_runtime": 6.2803,
|
| 357 |
+
"eval_samples_per_second": 44.584,
|
| 358 |
+
"eval_steps_per_second": 5.573,
|
| 359 |
+
"step": 45
|
| 360 |
+
},
|
| 361 |
+
{
|
| 362 |
+
"epoch": 1.55,
|
| 363 |
+
"learning_rate": 0.00011018229867038356,
|
| 364 |
+
"loss": 0.0591,
|
| 365 |
+
"step": 46
|
| 366 |
+
},
|
| 367 |
+
{
|
| 368 |
+
"epoch": 1.58,
|
| 369 |
+
"learning_rate": 0.00010611616608218429,
|
| 370 |
+
"loss": 0.0646,
|
| 371 |
+
"step": 47
|
| 372 |
+
},
|
| 373 |
+
{
|
| 374 |
+
"epoch": 1.62,
|
| 375 |
+
"learning_rate": 0.00010203985373914056,
|
| 376 |
+
"loss": 0.068,
|
| 377 |
+
"step": 48
|
| 378 |
+
},
|
| 379 |
+
{
|
| 380 |
+
"epoch": 1.65,
|
| 381 |
+
"learning_rate": 9.79601462608595e-05,
|
| 382 |
+
"loss": 0.055,
|
| 383 |
+
"step": 49
|
| 384 |
+
},
|
| 385 |
+
{
|
| 386 |
+
"epoch": 1.69,
|
| 387 |
+
"learning_rate": 9.388383391781575e-05,
|
| 388 |
+
"loss": 0.0687,
|
| 389 |
+
"step": 50
|
| 390 |
+
},
|
| 391 |
+
{
|
| 392 |
+
"epoch": 1.69,
|
| 393 |
+
"eval_loss": 0.06235363706946373,
|
| 394 |
+
"eval_runtime": 6.2849,
|
| 395 |
+
"eval_samples_per_second": 44.551,
|
| 396 |
+
"eval_steps_per_second": 5.569,
|
| 397 |
+
"step": 50
|
| 398 |
+
},
|
| 399 |
+
{
|
| 400 |
+
"epoch": 1.72,
|
| 401 |
+
"learning_rate": 8.981770132961649e-05,
|
| 402 |
+
"loss": 0.0643,
|
| 403 |
+
"step": 51
|
| 404 |
+
},
|
| 405 |
+
{
|
| 406 |
+
"epoch": 1.75,
|
| 407 |
+
"learning_rate": 8.57685161726715e-05,
|
| 408 |
+
"loss": 0.0589,
|
| 409 |
+
"step": 52
|
| 410 |
+
},
|
| 411 |
+
{
|
| 412 |
+
"epoch": 1.79,
|
| 413 |
+
"learning_rate": 8.174301791606385e-05,
|
| 414 |
+
"loss": 0.0729,
|
| 415 |
+
"step": 53
|
| 416 |
+
},
|
| 417 |
+
{
|
| 418 |
+
"epoch": 1.82,
|
| 419 |
+
"learning_rate": 7.774790660436858e-05,
|
| 420 |
+
"loss": 0.0553,
|
| 421 |
+
"step": 54
|
| 422 |
+
},
|
| 423 |
+
{
|
| 424 |
+
"epoch": 1.86,
|
| 425 |
+
"learning_rate": 7.378983170608982e-05,
|
| 426 |
+
"loss": 0.0575,
|
| 427 |
+
"step": 55
|
| 428 |
+
},
|
| 429 |
+
{
|
| 430 |
+
"epoch": 1.86,
|
| 431 |
+
"eval_loss": 0.062243152409791946,
|
| 432 |
+
"eval_runtime": 6.2854,
|
| 433 |
+
"eval_samples_per_second": 44.548,
|
| 434 |
+
"eval_steps_per_second": 5.568,
|
| 435 |
+
"step": 55
|
| 436 |
+
},
|
| 437 |
+
{
|
| 438 |
+
"epoch": 1.89,
|
| 439 |
+
"learning_rate": 6.98753810462766e-05,
|
| 440 |
+
"loss": 0.0675,
|
| 441 |
+
"step": 56
|
| 442 |
+
},
|
| 443 |
+
{
|
| 444 |
+
"epoch": 1.92,
|
| 445 |
+
"learning_rate": 6.601106984173835e-05,
|
| 446 |
+
"loss": 0.068,
|
| 447 |
+
"step": 57
|
| 448 |
+
},
|
| 449 |
+
{
|
| 450 |
+
"epoch": 1.96,
|
| 451 |
+
"learning_rate": 6.220332985710936e-05,
|
| 452 |
+
"loss": 0.0601,
|
| 453 |
+
"step": 58
|
| 454 |
+
},
|
| 455 |
+
{
|
| 456 |
+
"epoch": 1.99,
|
| 457 |
+
"learning_rate": 5.845849869981137e-05,
|
| 458 |
+
"loss": 0.0764,
|
| 459 |
+
"step": 59
|
| 460 |
+
}
|
| 461 |
+
],
|
| 462 |
+
"logging_steps": 1,
|
| 463 |
+
"max_steps": 87,
|
| 464 |
+
"num_train_epochs": 3,
|
| 465 |
+
"save_steps": 500,
|
| 466 |
+
"total_flos": 3.83737422209876e+17,
|
| 467 |
+
"trial_name": null,
|
| 468 |
+
"trial_params": null
|
| 469 |
+
}
|
checkpoint-59/training_args.bin
ADDED
|
@@ -0,0 +1,3 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
version https://git-lfs.github.com/spec/v1
|
| 2 |
+
oid sha256:5117945171d6dce6bb2ba0c56f263b94b8a6a7b57ffd9e74113faed0e7efe401
|
| 3 |
+
size 4539
|
checkpoint-87/config.json
ADDED
|
@@ -0,0 +1,27 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
{
|
| 2 |
+
"_name_or_path": "TinyLlama/TinyLlama-1.1B-intermediate-step-955k-token-2T",
|
| 3 |
+
"architectures": [
|
| 4 |
+
"LlamaForCausalLM"
|
| 5 |
+
],
|
| 6 |
+
"attention_bias": false,
|
| 7 |
+
"bos_token_id": 1,
|
| 8 |
+
"eos_token_id": 2,
|
| 9 |
+
"hidden_act": "silu",
|
| 10 |
+
"hidden_size": 2048,
|
| 11 |
+
"initializer_range": 0.02,
|
| 12 |
+
"intermediate_size": 5632,
|
| 13 |
+
"max_position_embeddings": 8192,
|
| 14 |
+
"model_type": "llama",
|
| 15 |
+
"num_attention_heads": 32,
|
| 16 |
+
"num_hidden_layers": 22,
|
| 17 |
+
"num_key_value_heads": 4,
|
| 18 |
+
"pretraining_tp": 1,
|
| 19 |
+
"rms_norm_eps": 1e-05,
|
| 20 |
+
"rope_scaling": null,
|
| 21 |
+
"rope_theta": 10000.0,
|
| 22 |
+
"tie_word_embeddings": false,
|
| 23 |
+
"torch_dtype": "bfloat16",
|
| 24 |
+
"transformers_version": "4.35.2",
|
| 25 |
+
"use_cache": false,
|
| 26 |
+
"vocab_size": 32000
|
| 27 |
+
}
|
checkpoint-87/generation_config.json
ADDED
|
@@ -0,0 +1,7 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
{
|
| 2 |
+
"bos_token_id": 1,
|
| 3 |
+
"eos_token_id": 2,
|
| 4 |
+
"max_length": 2048,
|
| 5 |
+
"pad_token_id": 0,
|
| 6 |
+
"transformers_version": "4.35.2"
|
| 7 |
+
}
|
checkpoint-87/model.safetensors
ADDED
|
@@ -0,0 +1,3 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
version https://git-lfs.github.com/spec/v1
|
| 2 |
+
oid sha256:24ae3d20794693a952ff9d5a3b016c72fbb1f51d7a74df450e0c3b2924bbfb89
|
| 3 |
+
size 2200119864
|
checkpoint-87/optimizer.pt
ADDED
|
@@ -0,0 +1,3 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
version https://git-lfs.github.com/spec/v1
|
| 2 |
+
oid sha256:33949d39a1688b3c72db263f5f826f3edc5f67ea95e3046eefa08218588b6aa4
|
| 3 |
+
size 2205157317
|
checkpoint-87/rng_state.pth
ADDED
|
@@ -0,0 +1,3 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
version https://git-lfs.github.com/spec/v1
|
| 2 |
+
oid sha256:4a2d2d1561481739972beb9da4cda24cd79618e56b2904d90829eda87023314e
|
| 3 |
+
size 14575
|
checkpoint-87/scheduler.pt
ADDED
|
@@ -0,0 +1,3 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
version https://git-lfs.github.com/spec/v1
|
| 2 |
+
oid sha256:094fe83ba6b00779a606e1c315d5190ccde76611e3c53c464d9f758dd4968766
|
| 3 |
+
size 627
|
checkpoint-87/trainer_state.json
ADDED
|
@@ -0,0 +1,685 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
{
|
| 2 |
+
"best_metric": null,
|
| 3 |
+
"best_model_checkpoint": null,
|
| 4 |
+
"epoch": 2.929787234042553,
|
| 5 |
+
"eval_steps": 5,
|
| 6 |
+
"global_step": 87,
|
| 7 |
+
"is_hyper_param_search": false,
|
| 8 |
+
"is_local_process_zero": true,
|
| 9 |
+
"is_world_process_zero": true,
|
| 10 |
+
"log_history": [
|
| 11 |
+
{
|
| 12 |
+
"epoch": 0.03,
|
| 13 |
+
"learning_rate": 2e-05,
|
| 14 |
+
"loss": 1.9849,
|
| 15 |
+
"step": 1
|
| 16 |
+
},
|
| 17 |
+
{
|
| 18 |
+
"epoch": 0.03,
|
| 19 |
+
"eval_loss": 2.0810964107513428,
|
| 20 |
+
"eval_runtime": 6.2278,
|
| 21 |
+
"eval_samples_per_second": 44.96,
|
| 22 |
+
"eval_steps_per_second": 5.62,
|
| 23 |
+
"step": 1
|
| 24 |
+
},
|
| 25 |
+
{
|
| 26 |
+
"epoch": 0.07,
|
| 27 |
+
"learning_rate": 4e-05,
|
| 28 |
+
"loss": 1.8931,
|
| 29 |
+
"step": 2
|
| 30 |
+
},
|
| 31 |
+
{
|
| 32 |
+
"epoch": 0.1,
|
| 33 |
+
"learning_rate": 6e-05,
|
| 34 |
+
"loss": 1.763,
|
| 35 |
+
"step": 3
|
| 36 |
+
},
|
| 37 |
+
{
|
| 38 |
+
"epoch": 0.14,
|
| 39 |
+
"learning_rate": 8e-05,
|
| 40 |
+
"loss": 1.6688,
|
| 41 |
+
"step": 4
|
| 42 |
+
},
|
| 43 |
+
{
|
| 44 |
+
"epoch": 0.17,
|
| 45 |
+
"learning_rate": 0.0001,
|
| 46 |
+
"loss": 1.3107,
|
| 47 |
+
"step": 5
|
| 48 |
+
},
|
| 49 |
+
{
|
| 50 |
+
"epoch": 0.17,
|
| 51 |
+
"eval_loss": 1.1991726160049438,
|
| 52 |
+
"eval_runtime": 6.3173,
|
| 53 |
+
"eval_samples_per_second": 44.322,
|
| 54 |
+
"eval_steps_per_second": 5.54,
|
| 55 |
+
"step": 5
|
| 56 |
+
},
|
| 57 |
+
{
|
| 58 |
+
"epoch": 0.2,
|
| 59 |
+
"learning_rate": 0.00012,
|
| 60 |
+
"loss": 1.1899,
|
| 61 |
+
"step": 6
|
| 62 |
+
},
|
| 63 |
+
{
|
| 64 |
+
"epoch": 0.24,
|
| 65 |
+
"learning_rate": 0.00014,
|
| 66 |
+
"loss": 1.041,
|
| 67 |
+
"step": 7
|
| 68 |
+
},
|
| 69 |
+
{
|
| 70 |
+
"epoch": 0.27,
|
| 71 |
+
"learning_rate": 0.00016,
|
| 72 |
+
"loss": 0.9779,
|
| 73 |
+
"step": 8
|
| 74 |
+
},
|
| 75 |
+
{
|
| 76 |
+
"epoch": 0.31,
|
| 77 |
+
"learning_rate": 0.00018,
|
| 78 |
+
"loss": 0.7671,
|
| 79 |
+
"step": 9
|
| 80 |
+
},
|
| 81 |
+
{
|
| 82 |
+
"epoch": 0.34,
|
| 83 |
+
"learning_rate": 0.0002,
|
| 84 |
+
"loss": 0.6399,
|
| 85 |
+
"step": 10
|
| 86 |
+
},
|
| 87 |
+
{
|
| 88 |
+
"epoch": 0.34,
|
| 89 |
+
"eval_loss": 0.6359391212463379,
|
| 90 |
+
"eval_runtime": 6.2721,
|
| 91 |
+
"eval_samples_per_second": 44.642,
|
| 92 |
+
"eval_steps_per_second": 5.58,
|
| 93 |
+
"step": 10
|
| 94 |
+
},
|
| 95 |
+
{
|
| 96 |
+
"epoch": 0.37,
|
| 97 |
+
"learning_rate": 0.0001999167799344583,
|
| 98 |
+
"loss": 0.6777,
|
| 99 |
+
"step": 11
|
| 100 |
+
},
|
| 101 |
+
{
|
| 102 |
+
"epoch": 0.41,
|
| 103 |
+
"learning_rate": 0.00019966725824941932,
|
| 104 |
+
"loss": 0.492,
|
| 105 |
+
"step": 12
|
| 106 |
+
},
|
| 107 |
+
{
|
| 108 |
+
"epoch": 0.44,
|
| 109 |
+
"learning_rate": 0.00019925185024910277,
|
| 110 |
+
"loss": 0.3781,
|
| 111 |
+
"step": 13
|
| 112 |
+
},
|
| 113 |
+
{
|
| 114 |
+
"epoch": 0.48,
|
| 115 |
+
"learning_rate": 0.0001986712473391289,
|
| 116 |
+
"loss": 0.4503,
|
| 117 |
+
"step": 14
|
| 118 |
+
},
|
| 119 |
+
{
|
| 120 |
+
"epoch": 0.51,
|
| 121 |
+
"learning_rate": 0.00019792641587574212,
|
| 122 |
+
"loss": 0.2779,
|
| 123 |
+
"step": 15
|
| 124 |
+
},
|
| 125 |
+
{
|
| 126 |
+
"epoch": 0.51,
|
| 127 |
+
"eval_loss": 0.28619876503944397,
|
| 128 |
+
"eval_runtime": 6.3023,
|
| 129 |
+
"eval_samples_per_second": 44.428,
|
| 130 |
+
"eval_steps_per_second": 5.554,
|
| 131 |
+
"step": 15
|
| 132 |
+
},
|
| 133 |
+
{
|
| 134 |
+
"epoch": 0.54,
|
| 135 |
+
"learning_rate": 0.00019701859555740648,
|
| 136 |
+
"loss": 0.2658,
|
| 137 |
+
"step": 16
|
| 138 |
+
},
|
| 139 |
+
{
|
| 140 |
+
"epoch": 0.58,
|
| 141 |
+
"learning_rate": 0.00019594929736144976,
|
| 142 |
+
"loss": 0.2282,
|
| 143 |
+
"step": 17
|
| 144 |
+
},
|
| 145 |
+
{
|
| 146 |
+
"epoch": 0.61,
|
| 147 |
+
"learning_rate": 0.000194720301029191,
|
| 148 |
+
"loss": 0.2149,
|
| 149 |
+
"step": 18
|
| 150 |
+
},
|
| 151 |
+
{
|
| 152 |
+
"epoch": 0.65,
|
| 153 |
+
"learning_rate": 0.0001933336521037367,
|
| 154 |
+
"loss": 0.1792,
|
| 155 |
+
"step": 19
|
| 156 |
+
},
|
| 157 |
+
{
|
| 158 |
+
"epoch": 0.68,
|
| 159 |
+
"learning_rate": 0.00019179165852537596,
|
| 160 |
+
"loss": 0.1807,
|
| 161 |
+
"step": 20
|
| 162 |
+
},
|
| 163 |
+
{
|
| 164 |
+
"epoch": 0.68,
|
| 165 |
+
"eval_loss": 0.1634436696767807,
|
| 166 |
+
"eval_runtime": 6.2724,
|
| 167 |
+
"eval_samples_per_second": 44.64,
|
| 168 |
+
"eval_steps_per_second": 5.58,
|
| 169 |
+
"step": 20
|
| 170 |
+
},
|
| 171 |
+
{
|
| 172 |
+
"epoch": 0.71,
|
| 173 |
+
"learning_rate": 0.0001900968867902419,
|
| 174 |
+
"loss": 0.1425,
|
| 175 |
+
"step": 21
|
| 176 |
+
},
|
| 177 |
+
{
|
| 178 |
+
"epoch": 0.75,
|
| 179 |
+
"learning_rate": 0.00018825215767863214,
|
| 180 |
+
"loss": 0.1377,
|
| 181 |
+
"step": 22
|
| 182 |
+
},
|
| 183 |
+
{
|
| 184 |
+
"epoch": 0.78,
|
| 185 |
+
"learning_rate": 0.00018626054156009806,
|
| 186 |
+
"loss": 0.1173,
|
| 187 |
+
"step": 23
|
| 188 |
+
},
|
| 189 |
+
{
|
| 190 |
+
"epoch": 0.82,
|
| 191 |
+
"learning_rate": 0.00018412535328311814,
|
| 192 |
+
"loss": 0.1338,
|
| 193 |
+
"step": 24
|
| 194 |
+
},
|
| 195 |
+
{
|
| 196 |
+
"epoch": 0.85,
|
| 197 |
+
"learning_rate": 0.00018185014665785936,
|
| 198 |
+
"loss": 0.1256,
|
| 199 |
+
"step": 25
|
| 200 |
+
},
|
| 201 |
+
{
|
| 202 |
+
"epoch": 0.85,
|
| 203 |
+
"eval_loss": 0.1177145466208458,
|
| 204 |
+
"eval_runtime": 6.2558,
|
| 205 |
+
"eval_samples_per_second": 44.758,
|
| 206 |
+
"eval_steps_per_second": 5.595,
|
| 207 |
+
"step": 25
|
| 208 |
+
},
|
| 209 |
+
{
|
| 210 |
+
"epoch": 0.89,
|
| 211 |
+
"learning_rate": 0.00017943870854121124,
|
| 212 |
+
"loss": 0.1234,
|
| 213 |
+
"step": 26
|
| 214 |
+
},
|
| 215 |
+
{
|
| 216 |
+
"epoch": 0.92,
|
| 217 |
+
"learning_rate": 0.0001768950525339362,
|
| 218 |
+
"loss": 0.1164,
|
| 219 |
+
"step": 27
|
| 220 |
+
},
|
| 221 |
+
{
|
| 222 |
+
"epoch": 0.95,
|
| 223 |
+
"learning_rate": 0.000174223412300427,
|
| 224 |
+
"loss": 0.1077,
|
| 225 |
+
"step": 28
|
| 226 |
+
},
|
| 227 |
+
{
|
| 228 |
+
"epoch": 0.99,
|
| 229 |
+
"learning_rate": 0.00017142823452219038,
|
| 230 |
+
"loss": 0.0877,
|
| 231 |
+
"step": 29
|
| 232 |
+
},
|
| 233 |
+
{
|
| 234 |
+
"epoch": 1.0,
|
| 235 |
+
"learning_rate": 0.00016851417149678444,
|
| 236 |
+
"loss": 0.097,
|
| 237 |
+
"step": 30
|
| 238 |
+
},
|
| 239 |
+
{
|
| 240 |
+
"epoch": 1.0,
|
| 241 |
+
"eval_loss": 0.08908110111951828,
|
| 242 |
+
"eval_runtime": 6.274,
|
| 243 |
+
"eval_samples_per_second": 44.628,
|
| 244 |
+
"eval_steps_per_second": 5.579,
|
| 245 |
+
"step": 30
|
| 246 |
+
},
|
| 247 |
+
{
|
| 248 |
+
"epoch": 1.04,
|
| 249 |
+
"learning_rate": 0.00016548607339452853,
|
| 250 |
+
"loss": 0.0741,
|
| 251 |
+
"step": 31
|
| 252 |
+
},
|
| 253 |
+
{
|
| 254 |
+
"epoch": 1.07,
|
| 255 |
+
"learning_rate": 0.00016234898018587337,
|
| 256 |
+
"loss": 0.0811,
|
| 257 |
+
"step": 32
|
| 258 |
+
},
|
| 259 |
+
{
|
| 260 |
+
"epoch": 1.11,
|
| 261 |
+
"learning_rate": 0.00015910811325286768,
|
| 262 |
+
"loss": 0.0743,
|
| 263 |
+
"step": 33
|
| 264 |
+
},
|
| 265 |
+
{
|
| 266 |
+
"epoch": 1.14,
|
| 267 |
+
"learning_rate": 0.00015576886669868296,
|
| 268 |
+
"loss": 0.0799,
|
| 269 |
+
"step": 34
|
| 270 |
+
},
|
| 271 |
+
{
|
| 272 |
+
"epoch": 1.17,
|
| 273 |
+
"learning_rate": 0.00015233679836966122,
|
| 274 |
+
"loss": 0.1063,
|
| 275 |
+
"step": 35
|
| 276 |
+
},
|
| 277 |
+
{
|
| 278 |
+
"epoch": 1.17,
|
| 279 |
+
"eval_loss": 0.07344743609428406,
|
| 280 |
+
"eval_runtime": 6.3055,
|
| 281 |
+
"eval_samples_per_second": 44.406,
|
| 282 |
+
"eval_steps_per_second": 5.551,
|
| 283 |
+
"step": 35
|
| 284 |
+
},
|
| 285 |
+
{
|
| 286 |
+
"epoch": 1.21,
|
| 287 |
+
"learning_rate": 0.00014881762060482814,
|
| 288 |
+
"loss": 0.0835,
|
| 289 |
+
"step": 36
|
| 290 |
+
},
|
| 291 |
+
{
|
| 292 |
+
"epoch": 1.24,
|
| 293 |
+
"learning_rate": 0.00014521719072826858,
|
| 294 |
+
"loss": 0.0661,
|
| 295 |
+
"step": 37
|
| 296 |
+
},
|
| 297 |
+
{
|
| 298 |
+
"epoch": 1.28,
|
| 299 |
+
"learning_rate": 0.00014154150130018866,
|
| 300 |
+
"loss": 0.0637,
|
| 301 |
+
"step": 38
|
| 302 |
+
},
|
| 303 |
+
{
|
| 304 |
+
"epoch": 1.31,
|
| 305 |
+
"learning_rate": 0.00013779667014289065,
|
| 306 |
+
"loss": 0.0642,
|
| 307 |
+
"step": 39
|
| 308 |
+
},
|
| 309 |
+
{
|
| 310 |
+
"epoch": 1.34,
|
| 311 |
+
"learning_rate": 0.00013398893015826167,
|
| 312 |
+
"loss": 0.0769,
|
| 313 |
+
"step": 40
|
| 314 |
+
},
|
| 315 |
+
{
|
| 316 |
+
"epoch": 1.34,
|
| 317 |
+
"eval_loss": 0.07225509732961655,
|
| 318 |
+
"eval_runtime": 6.2831,
|
| 319 |
+
"eval_samples_per_second": 44.564,
|
| 320 |
+
"eval_steps_per_second": 5.571,
|
| 321 |
+
"step": 40
|
| 322 |
+
},
|
| 323 |
+
{
|
| 324 |
+
"epoch": 1.38,
|
| 325 |
+
"learning_rate": 0.00013012461895372344,
|
| 326 |
+
"loss": 0.0746,
|
| 327 |
+
"step": 41
|
| 328 |
+
},
|
| 329 |
+
{
|
| 330 |
+
"epoch": 1.41,
|
| 331 |
+
"learning_rate": 0.00012621016829391022,
|
| 332 |
+
"loss": 0.0592,
|
| 333 |
+
"step": 42
|
| 334 |
+
},
|
| 335 |
+
{
|
| 336 |
+
"epoch": 1.45,
|
| 337 |
+
"learning_rate": 0.00012225209339563145,
|
| 338 |
+
"loss": 0.078,
|
| 339 |
+
"step": 43
|
| 340 |
+
},
|
| 341 |
+
{
|
| 342 |
+
"epoch": 1.48,
|
| 343 |
+
"learning_rate": 0.00011825698208393619,
|
| 344 |
+
"loss": 0.088,
|
| 345 |
+
"step": 44
|
| 346 |
+
},
|
| 347 |
+
{
|
| 348 |
+
"epoch": 1.51,
|
| 349 |
+
"learning_rate": 0.00011423148382732853,
|
| 350 |
+
"loss": 0.0694,
|
| 351 |
+
"step": 45
|
| 352 |
+
},
|
| 353 |
+
{
|
| 354 |
+
"epoch": 1.51,
|
| 355 |
+
"eval_loss": 0.06333146244287491,
|
| 356 |
+
"eval_runtime": 6.2803,
|
| 357 |
+
"eval_samples_per_second": 44.584,
|
| 358 |
+
"eval_steps_per_second": 5.573,
|
| 359 |
+
"step": 45
|
| 360 |
+
},
|
| 361 |
+
{
|
| 362 |
+
"epoch": 1.55,
|
| 363 |
+
"learning_rate": 0.00011018229867038356,
|
| 364 |
+
"loss": 0.0591,
|
| 365 |
+
"step": 46
|
| 366 |
+
},
|
| 367 |
+
{
|
| 368 |
+
"epoch": 1.58,
|
| 369 |
+
"learning_rate": 0.00010611616608218429,
|
| 370 |
+
"loss": 0.0646,
|
| 371 |
+
"step": 47
|
| 372 |
+
},
|
| 373 |
+
{
|
| 374 |
+
"epoch": 1.62,
|
| 375 |
+
"learning_rate": 0.00010203985373914056,
|
| 376 |
+
"loss": 0.068,
|
| 377 |
+
"step": 48
|
| 378 |
+
},
|
| 379 |
+
{
|
| 380 |
+
"epoch": 1.65,
|
| 381 |
+
"learning_rate": 9.79601462608595e-05,
|
| 382 |
+
"loss": 0.055,
|
| 383 |
+
"step": 49
|
| 384 |
+
},
|
| 385 |
+
{
|
| 386 |
+
"epoch": 1.69,
|
| 387 |
+
"learning_rate": 9.388383391781575e-05,
|
| 388 |
+
"loss": 0.0687,
|
| 389 |
+
"step": 50
|
| 390 |
+
},
|
| 391 |
+
{
|
| 392 |
+
"epoch": 1.69,
|
| 393 |
+
"eval_loss": 0.06235363706946373,
|
| 394 |
+
"eval_runtime": 6.2849,
|
| 395 |
+
"eval_samples_per_second": 44.551,
|
| 396 |
+
"eval_steps_per_second": 5.569,
|
| 397 |
+
"step": 50
|
| 398 |
+
},
|
| 399 |
+
{
|
| 400 |
+
"epoch": 1.72,
|
| 401 |
+
"learning_rate": 8.981770132961649e-05,
|
| 402 |
+
"loss": 0.0643,
|
| 403 |
+
"step": 51
|
| 404 |
+
},
|
| 405 |
+
{
|
| 406 |
+
"epoch": 1.75,
|
| 407 |
+
"learning_rate": 8.57685161726715e-05,
|
| 408 |
+
"loss": 0.0589,
|
| 409 |
+
"step": 52
|
| 410 |
+
},
|
| 411 |
+
{
|
| 412 |
+
"epoch": 1.79,
|
| 413 |
+
"learning_rate": 8.174301791606385e-05,
|
| 414 |
+
"loss": 0.0729,
|
| 415 |
+
"step": 53
|
| 416 |
+
},
|
| 417 |
+
{
|
| 418 |
+
"epoch": 1.82,
|
| 419 |
+
"learning_rate": 7.774790660436858e-05,
|
| 420 |
+
"loss": 0.0553,
|
| 421 |
+
"step": 54
|
| 422 |
+
},
|
| 423 |
+
{
|
| 424 |
+
"epoch": 1.86,
|
| 425 |
+
"learning_rate": 7.378983170608982e-05,
|
| 426 |
+
"loss": 0.0575,
|
| 427 |
+
"step": 55
|
| 428 |
+
},
|
| 429 |
+
{
|
| 430 |
+
"epoch": 1.86,
|
| 431 |
+
"eval_loss": 0.062243152409791946,
|
| 432 |
+
"eval_runtime": 6.2854,
|
| 433 |
+
"eval_samples_per_second": 44.548,
|
| 434 |
+
"eval_steps_per_second": 5.568,
|
| 435 |
+
"step": 55
|
| 436 |
+
},
|
| 437 |
+
{
|
| 438 |
+
"epoch": 1.89,
|
| 439 |
+
"learning_rate": 6.98753810462766e-05,
|
| 440 |
+
"loss": 0.0675,
|
| 441 |
+
"step": 56
|
| 442 |
+
},
|
| 443 |
+
{
|
| 444 |
+
"epoch": 1.92,
|
| 445 |
+
"learning_rate": 6.601106984173835e-05,
|
| 446 |
+
"loss": 0.068,
|
| 447 |
+
"step": 57
|
| 448 |
+
},
|
| 449 |
+
{
|
| 450 |
+
"epoch": 1.96,
|
| 451 |
+
"learning_rate": 6.220332985710936e-05,
|
| 452 |
+
"loss": 0.0601,
|
| 453 |
+
"step": 58
|
| 454 |
+
},
|
| 455 |
+
{
|
| 456 |
+
"epoch": 1.99,
|
| 457 |
+
"learning_rate": 5.845849869981137e-05,
|
| 458 |
+
"loss": 0.0764,
|
| 459 |
+
"step": 59
|
| 460 |
+
},
|
| 461 |
+
{
|
| 462 |
+
"epoch": 2.01,
|
| 463 |
+
"learning_rate": 5.478280927173145e-05,
|
| 464 |
+
"loss": 0.0516,
|
| 465 |
+
"step": 60
|
| 466 |
+
},
|
| 467 |
+
{
|
| 468 |
+
"epoch": 2.01,
|
| 469 |
+
"eval_loss": 0.060868892818689346,
|
| 470 |
+
"eval_runtime": 5.9359,
|
| 471 |
+
"eval_samples_per_second": 47.171,
|
| 472 |
+
"eval_steps_per_second": 5.896,
|
| 473 |
+
"step": 60
|
| 474 |
+
},
|
| 475 |
+
{
|
| 476 |
+
"epoch": 2.04,
|
| 477 |
+
"learning_rate": 5.11823793951719e-05,
|
| 478 |
+
"loss": 0.0516,
|
| 479 |
+
"step": 61
|
| 480 |
+
},
|
| 481 |
+
{
|
| 482 |
+
"epoch": 2.08,
|
| 483 |
+
"learning_rate": 4.7663201630338816e-05,
|
| 484 |
+
"loss": 0.0543,
|
| 485 |
+
"step": 62
|
| 486 |
+
},
|
| 487 |
+
{
|
| 488 |
+
"epoch": 2.11,
|
| 489 |
+
"learning_rate": 4.423113330131707e-05,
|
| 490 |
+
"loss": 0.0519,
|
| 491 |
+
"step": 63
|
| 492 |
+
},
|
| 493 |
+
{
|
| 494 |
+
"epoch": 2.15,
|
| 495 |
+
"learning_rate": 4.089188674713236e-05,
|
| 496 |
+
"loss": 0.0602,
|
| 497 |
+
"step": 64
|
| 498 |
+
},
|
| 499 |
+
{
|
| 500 |
+
"epoch": 2.18,
|
| 501 |
+
"learning_rate": 3.7651019814126654e-05,
|
| 502 |
+
"loss": 0.0582,
|
| 503 |
+
"step": 65
|
| 504 |
+
},
|
| 505 |
+
{
|
| 506 |
+
"epoch": 2.18,
|
| 507 |
+
"eval_loss": 0.060279980301856995,
|
| 508 |
+
"eval_runtime": 5.9625,
|
| 509 |
+
"eval_samples_per_second": 46.96,
|
| 510 |
+
"eval_steps_per_second": 5.87,
|
| 511 |
+
"step": 65
|
| 512 |
+
},
|
| 513 |
+
{
|
| 514 |
+
"epoch": 2.21,
|
| 515 |
+
"learning_rate": 3.45139266054715e-05,
|
| 516 |
+
"loss": 0.055,
|
| 517 |
+
"step": 66
|
| 518 |
+
},
|
| 519 |
+
{
|
| 520 |
+
"epoch": 2.25,
|
| 521 |
+
"learning_rate": 3.1485828503215585e-05,
|
| 522 |
+
"loss": 0.0515,
|
| 523 |
+
"step": 67
|
| 524 |
+
},
|
| 525 |
+
{
|
| 526 |
+
"epoch": 2.28,
|
| 527 |
+
"learning_rate": 2.8571765477809643e-05,
|
| 528 |
+
"loss": 0.0515,
|
| 529 |
+
"step": 68
|
| 530 |
+
},
|
| 531 |
+
{
|
| 532 |
+
"epoch": 2.32,
|
| 533 |
+
"learning_rate": 2.5776587699573006e-05,
|
| 534 |
+
"loss": 0.0497,
|
| 535 |
+
"step": 69
|
| 536 |
+
},
|
| 537 |
+
{
|
| 538 |
+
"epoch": 2.35,
|
| 539 |
+
"learning_rate": 2.3104947466063787e-05,
|
| 540 |
+
"loss": 0.0611,
|
| 541 |
+
"step": 70
|
| 542 |
+
},
|
| 543 |
+
{
|
| 544 |
+
"epoch": 2.35,
|
| 545 |
+
"eval_loss": 0.05999369919300079,
|
| 546 |
+
"eval_runtime": 5.9591,
|
| 547 |
+
"eval_samples_per_second": 46.987,
|
| 548 |
+
"eval_steps_per_second": 5.873,
|
| 549 |
+
"step": 70
|
| 550 |
+
},
|
| 551 |
+
{
|
| 552 |
+
"epoch": 2.39,
|
| 553 |
+
"learning_rate": 2.0561291458788733e-05,
|
| 554 |
+
"loss": 0.0564,
|
| 555 |
+
"step": 71
|
| 556 |
+
},
|
| 557 |
+
{
|
| 558 |
+
"epoch": 2.42,
|
| 559 |
+
"learning_rate": 1.8149853342140645e-05,
|
| 560 |
+
"loss": 0.0498,
|
| 561 |
+
"step": 72
|
| 562 |
+
},
|
| 563 |
+
{
|
| 564 |
+
"epoch": 2.45,
|
| 565 |
+
"learning_rate": 1.587464671688187e-05,
|
| 566 |
+
"loss": 0.0503,
|
| 567 |
+
"step": 73
|
| 568 |
+
},
|
| 569 |
+
{
|
| 570 |
+
"epoch": 2.49,
|
| 571 |
+
"learning_rate": 1.373945843990192e-05,
|
| 572 |
+
"loss": 0.0494,
|
| 573 |
+
"step": 74
|
| 574 |
+
},
|
| 575 |
+
{
|
| 576 |
+
"epoch": 2.52,
|
| 577 |
+
"learning_rate": 1.1747842321367886e-05,
|
| 578 |
+
"loss": 0.0515,
|
| 579 |
+
"step": 75
|
| 580 |
+
},
|
| 581 |
+
{
|
| 582 |
+
"epoch": 2.52,
|
| 583 |
+
"eval_loss": 0.05982312187552452,
|
| 584 |
+
"eval_runtime": 5.856,
|
| 585 |
+
"eval_samples_per_second": 47.814,
|
| 586 |
+
"eval_steps_per_second": 5.977,
|
| 587 |
+
"step": 75
|
| 588 |
+
},
|
| 589 |
+
{
|
| 590 |
+
"epoch": 2.56,
|
| 591 |
+
"learning_rate": 9.903113209758096e-06,
|
| 592 |
+
"loss": 0.0508,
|
| 593 |
+
"step": 76
|
| 594 |
+
},
|
| 595 |
+
{
|
| 596 |
+
"epoch": 2.59,
|
| 597 |
+
"learning_rate": 8.208341474624071e-06,
|
| 598 |
+
"loss": 0.0596,
|
| 599 |
+
"step": 77
|
| 600 |
+
},
|
| 601 |
+
{
|
| 602 |
+
"epoch": 2.62,
|
| 603 |
+
"learning_rate": 6.666347896263325e-06,
|
| 604 |
+
"loss": 0.0606,
|
| 605 |
+
"step": 78
|
| 606 |
+
},
|
| 607 |
+
{
|
| 608 |
+
"epoch": 2.66,
|
| 609 |
+
"learning_rate": 5.27969897080901e-06,
|
| 610 |
+
"loss": 0.0484,
|
| 611 |
+
"step": 79
|
| 612 |
+
},
|
| 613 |
+
{
|
| 614 |
+
"epoch": 2.69,
|
| 615 |
+
"learning_rate": 4.050702638550275e-06,
|
| 616 |
+
"loss": 0.0704,
|
| 617 |
+
"step": 80
|
| 618 |
+
},
|
| 619 |
+
{
|
| 620 |
+
"epoch": 2.69,
|
| 621 |
+
"eval_loss": 0.05977718159556389,
|
| 622 |
+
"eval_runtime": 5.9454,
|
| 623 |
+
"eval_samples_per_second": 47.095,
|
| 624 |
+
"eval_steps_per_second": 5.887,
|
| 625 |
+
"step": 80
|
| 626 |
+
},
|
| 627 |
+
{
|
| 628 |
+
"epoch": 2.73,
|
| 629 |
+
"learning_rate": 2.9814044425935606e-06,
|
| 630 |
+
"loss": 0.0534,
|
| 631 |
+
"step": 81
|
| 632 |
+
},
|
| 633 |
+
{
|
| 634 |
+
"epoch": 2.76,
|
| 635 |
+
"learning_rate": 2.073584124257899e-06,
|
| 636 |
+
"loss": 0.0765,
|
| 637 |
+
"step": 82
|
| 638 |
+
},
|
| 639 |
+
{
|
| 640 |
+
"epoch": 2.79,
|
| 641 |
+
"learning_rate": 1.3287526608711131e-06,
|
| 642 |
+
"loss": 0.0557,
|
| 643 |
+
"step": 83
|
| 644 |
+
},
|
| 645 |
+
{
|
| 646 |
+
"epoch": 2.83,
|
| 647 |
+
"learning_rate": 7.481497508972312e-07,
|
| 648 |
+
"loss": 0.056,
|
| 649 |
+
"step": 84
|
| 650 |
+
},
|
| 651 |
+
{
|
| 652 |
+
"epoch": 2.86,
|
| 653 |
+
"learning_rate": 3.3274175058067846e-07,
|
| 654 |
+
"loss": 0.0525,
|
| 655 |
+
"step": 85
|
| 656 |
+
},
|
| 657 |
+
{
|
| 658 |
+
"epoch": 2.86,
|
| 659 |
+
"eval_loss": 0.05975683405995369,
|
| 660 |
+
"eval_runtime": 5.9096,
|
| 661 |
+
"eval_samples_per_second": 47.381,
|
| 662 |
+
"eval_steps_per_second": 5.923,
|
| 663 |
+
"step": 85
|
| 664 |
+
},
|
| 665 |
+
{
|
| 666 |
+
"epoch": 2.9,
|
| 667 |
+
"learning_rate": 8.322006554171146e-08,
|
| 668 |
+
"loss": 0.055,
|
| 669 |
+
"step": 86
|
| 670 |
+
},
|
| 671 |
+
{
|
| 672 |
+
"epoch": 2.93,
|
| 673 |
+
"learning_rate": 0.0,
|
| 674 |
+
"loss": 0.0625,
|
| 675 |
+
"step": 87
|
| 676 |
+
}
|
| 677 |
+
],
|
| 678 |
+
"logging_steps": 1,
|
| 679 |
+
"max_steps": 87,
|
| 680 |
+
"num_train_epochs": 3,
|
| 681 |
+
"save_steps": 500,
|
| 682 |
+
"total_flos": 5.5952097199364506e+17,
|
| 683 |
+
"trial_name": null,
|
| 684 |
+
"trial_params": null
|
| 685 |
+
}
|
checkpoint-87/training_args.bin
ADDED
|
@@ -0,0 +1,3 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
version https://git-lfs.github.com/spec/v1
|
| 2 |
+
oid sha256:5117945171d6dce6bb2ba0c56f263b94b8a6a7b57ffd9e74113faed0e7efe401
|
| 3 |
+
size 4539
|
config.json
ADDED
|
@@ -0,0 +1,27 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
{
|
| 2 |
+
"_name_or_path": "TinyLlama/TinyLlama-1.1B-intermediate-step-955k-token-2T",
|
| 3 |
+
"architectures": [
|
| 4 |
+
"LlamaForCausalLM"
|
| 5 |
+
],
|
| 6 |
+
"attention_bias": false,
|
| 7 |
+
"bos_token_id": 1,
|
| 8 |
+
"eos_token_id": 2,
|
| 9 |
+
"hidden_act": "silu",
|
| 10 |
+
"hidden_size": 2048,
|
| 11 |
+
"initializer_range": 0.02,
|
| 12 |
+
"intermediate_size": 5632,
|
| 13 |
+
"max_position_embeddings": 8192,
|
| 14 |
+
"model_type": "llama",
|
| 15 |
+
"num_attention_heads": 32,
|
| 16 |
+
"num_hidden_layers": 22,
|
| 17 |
+
"num_key_value_heads": 4,
|
| 18 |
+
"pretraining_tp": 1,
|
| 19 |
+
"rms_norm_eps": 1e-05,
|
| 20 |
+
"rope_scaling": null,
|
| 21 |
+
"rope_theta": 10000.0,
|
| 22 |
+
"tie_word_embeddings": false,
|
| 23 |
+
"torch_dtype": "bfloat16",
|
| 24 |
+
"transformers_version": "4.35.2",
|
| 25 |
+
"use_cache": false,
|
| 26 |
+
"vocab_size": 32000
|
| 27 |
+
}
|
generation_config.json
ADDED
|
@@ -0,0 +1,7 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
{
|
| 2 |
+
"bos_token_id": 1,
|
| 3 |
+
"eos_token_id": 2,
|
| 4 |
+
"max_length": 2048,
|
| 5 |
+
"pad_token_id": 0,
|
| 6 |
+
"transformers_version": "4.35.2"
|
| 7 |
+
}
|
pytorch_model.bin
ADDED
|
@@ -0,0 +1,3 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
version https://git-lfs.github.com/spec/v1
|
| 2 |
+
oid sha256:3b2b707c5f4166544aec6bac0b4c47b0b07860bcd15ab13e0cc1ed7dec3bb73e
|
| 3 |
+
size 2200164273
|
special_tokens_map.json
ADDED
|
@@ -0,0 +1,24 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 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": "</s>",
|
| 17 |
+
"unk_token": {
|
| 18 |
+
"content": "<unk>",
|
| 19 |
+
"lstrip": false,
|
| 20 |
+
"normalized": false,
|
| 21 |
+
"rstrip": false,
|
| 22 |
+
"single_word": false
|
| 23 |
+
}
|
| 24 |
+
}
|
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 +1,44 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
{
|
| 2 |
+
"add_bos_token": true,
|
| 3 |
+
"add_eos_token": false,
|
| 4 |
+
"added_tokens_decoder": {
|
| 5 |
+
"0": {
|
| 6 |
+
"content": "<unk>",
|
| 7 |
+
"lstrip": false,
|
| 8 |
+
"normalized": false,
|
| 9 |
+
"rstrip": false,
|
| 10 |
+
"single_word": false,
|
| 11 |
+
"special": true
|
| 12 |
+
},
|
| 13 |
+
"1": {
|
| 14 |
+
"content": "<s>",
|
| 15 |
+
"lstrip": false,
|
| 16 |
+
"normalized": false,
|
| 17 |
+
"rstrip": false,
|
| 18 |
+
"single_word": false,
|
| 19 |
+
"special": true
|
| 20 |
+
},
|
| 21 |
+
"2": {
|
| 22 |
+
"content": "</s>",
|
| 23 |
+
"lstrip": false,
|
| 24 |
+
"normalized": false,
|
| 25 |
+
"rstrip": false,
|
| 26 |
+
"single_word": false,
|
| 27 |
+
"special": true
|
| 28 |
+
}
|
| 29 |
+
},
|
| 30 |
+
"bos_token": "<s>",
|
| 31 |
+
"clean_up_tokenization_spaces": false,
|
| 32 |
+
"eos_token": "</s>",
|
| 33 |
+
"legacy": false,
|
| 34 |
+
"model_max_length": 1000000000000000019884624838656,
|
| 35 |
+
"pad_token": "</s>",
|
| 36 |
+
"padding_side": "right",
|
| 37 |
+
"sp_model_kwargs": {},
|
| 38 |
+
"spaces_between_special_tokens": false,
|
| 39 |
+
"tokenizer_class": "LlamaTokenizer",
|
| 40 |
+
"trust_remote_code": false,
|
| 41 |
+
"unk_token": "<unk>",
|
| 42 |
+
"use_default_system_prompt": false,
|
| 43 |
+
"use_fast": true
|
| 44 |
+
}
|