Commit
•
21e3177
1
Parent(s):
e085616
Model save
Browse files- README.md +66 -0
- all_results.json +9 -0
- generation_config.json +7 -0
- model.safetensors +1 -1
- runs/May30_11-56-23_poseidon/events.out.tfevents.1717070217.poseidon.2010350.0 +2 -2
- train_results.json +9 -0
- trainer_state.json +576 -0
README.md
ADDED
@@ -0,0 +1,66 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
1 |
+
---
|
2 |
+
license: apache-2.0
|
3 |
+
base_model: TinyLlama/TinyLlama-1.1B-intermediate-step-1431k-3T
|
4 |
+
tags:
|
5 |
+
- trl
|
6 |
+
- sft
|
7 |
+
- generated_from_trainer
|
8 |
+
datasets:
|
9 |
+
- generator
|
10 |
+
model-index:
|
11 |
+
- name: tinyllama-1.1b-mt-sft-full
|
12 |
+
results: []
|
13 |
+
---
|
14 |
+
|
15 |
+
<!-- This model card has been generated automatically according to the information the Trainer had access to. You
|
16 |
+
should probably proofread and complete it, then remove this comment. -->
|
17 |
+
|
18 |
+
# tinyllama-1.1b-mt-sft-full
|
19 |
+
|
20 |
+
This model is a fine-tuned version of [TinyLlama/TinyLlama-1.1B-intermediate-step-1431k-3T](https://huggingface.co/TinyLlama/TinyLlama-1.1B-intermediate-step-1431k-3T) on the generator dataset.
|
21 |
+
It achieves the following results on the evaluation set:
|
22 |
+
- Loss: 1.6971
|
23 |
+
|
24 |
+
## Model description
|
25 |
+
|
26 |
+
More information needed
|
27 |
+
|
28 |
+
## Intended uses & limitations
|
29 |
+
|
30 |
+
More information needed
|
31 |
+
|
32 |
+
## Training and evaluation data
|
33 |
+
|
34 |
+
More information needed
|
35 |
+
|
36 |
+
## Training procedure
|
37 |
+
|
38 |
+
### Training hyperparameters
|
39 |
+
|
40 |
+
The following hyperparameters were used during training:
|
41 |
+
- learning_rate: 2e-05
|
42 |
+
- train_batch_size: 8
|
43 |
+
- eval_batch_size: 8
|
44 |
+
- seed: 42
|
45 |
+
- distributed_type: multi-GPU
|
46 |
+
- gradient_accumulation_steps: 4
|
47 |
+
- total_train_batch_size: 32
|
48 |
+
- optimizer: Adam with betas=(0.9,0.999) and epsilon=1e-08
|
49 |
+
- lr_scheduler_type: cosine
|
50 |
+
- lr_scheduler_warmup_ratio: 0.1
|
51 |
+
- num_epochs: 2
|
52 |
+
|
53 |
+
### Training results
|
54 |
+
|
55 |
+
| Training Loss | Epoch | Step | Validation Loss |
|
56 |
+
|:-------------:|:------:|:----:|:---------------:|
|
57 |
+
| 1.601 | 0.9959 | 184 | 1.6972 |
|
58 |
+
| 1.5526 | 1.9919 | 368 | 1.6971 |
|
59 |
+
|
60 |
+
|
61 |
+
### Framework versions
|
62 |
+
|
63 |
+
- Transformers 4.41.1
|
64 |
+
- Pytorch 2.1.2
|
65 |
+
- Datasets 2.19.1
|
66 |
+
- Tokenizers 0.19.1
|
all_results.json
ADDED
@@ -0,0 +1,9 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
1 |
+
{
|
2 |
+
"epoch": 1.9918809201623815,
|
3 |
+
"total_flos": 1.4968483831454106e+17,
|
4 |
+
"train_loss": 1.6339908747569374,
|
5 |
+
"train_runtime": 3216.6871,
|
6 |
+
"train_samples": 117404,
|
7 |
+
"train_samples_per_second": 3.675,
|
8 |
+
"train_steps_per_second": 0.114
|
9 |
+
}
|
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": "4.41.1"
|
7 |
+
}
|
model.safetensors
CHANGED
@@ -1,3 +1,3 @@
|
|
1 |
version https://git-lfs.github.com/spec/v1
|
2 |
-
oid sha256:
|
3 |
size 2200119864
|
|
|
1 |
version https://git-lfs.github.com/spec/v1
|
2 |
+
oid sha256:aae421ae34776ad2b8f19065debf6cd110c0cb64cde21e48e76f317a326ef945
|
3 |
size 2200119864
|
runs/May30_11-56-23_poseidon/events.out.tfevents.1717070217.poseidon.2010350.0
CHANGED
@@ -1,3 +1,3 @@
|
|
1 |
version https://git-lfs.github.com/spec/v1
|
2 |
-
oid sha256:
|
3 |
-
size
|
|
|
1 |
version https://git-lfs.github.com/spec/v1
|
2 |
+
oid sha256:73674fec07efc1ef3eafbb3514ddcd8e084120121932b1e20bc225da460429e6
|
3 |
+
size 21433
|
train_results.json
ADDED
@@ -0,0 +1,9 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
1 |
+
{
|
2 |
+
"epoch": 1.9918809201623815,
|
3 |
+
"total_flos": 1.4968483831454106e+17,
|
4 |
+
"train_loss": 1.6339908747569374,
|
5 |
+
"train_runtime": 3216.6871,
|
6 |
+
"train_samples": 117404,
|
7 |
+
"train_samples_per_second": 3.675,
|
8 |
+
"train_steps_per_second": 0.114
|
9 |
+
}
|
trainer_state.json
ADDED
@@ -0,0 +1,576 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
1 |
+
{
|
2 |
+
"best_metric": null,
|
3 |
+
"best_model_checkpoint": null,
|
4 |
+
"epoch": 1.9918809201623815,
|
5 |
+
"eval_steps": 500,
|
6 |
+
"global_step": 368,
|
7 |
+
"is_hyper_param_search": false,
|
8 |
+
"is_local_process_zero": true,
|
9 |
+
"is_world_process_zero": true,
|
10 |
+
"log_history": [
|
11 |
+
{
|
12 |
+
"epoch": 0.005412719891745603,
|
13 |
+
"grad_norm": 23.5,
|
14 |
+
"learning_rate": 5.405405405405406e-07,
|
15 |
+
"loss": 2.2045,
|
16 |
+
"step": 1
|
17 |
+
},
|
18 |
+
{
|
19 |
+
"epoch": 0.02706359945872801,
|
20 |
+
"grad_norm": 19.375,
|
21 |
+
"learning_rate": 2.702702702702703e-06,
|
22 |
+
"loss": 2.2384,
|
23 |
+
"step": 5
|
24 |
+
},
|
25 |
+
{
|
26 |
+
"epoch": 0.05412719891745602,
|
27 |
+
"grad_norm": 6.6875,
|
28 |
+
"learning_rate": 5.405405405405406e-06,
|
29 |
+
"loss": 2.1809,
|
30 |
+
"step": 10
|
31 |
+
},
|
32 |
+
{
|
33 |
+
"epoch": 0.08119079837618404,
|
34 |
+
"grad_norm": 3.921875,
|
35 |
+
"learning_rate": 8.108108108108109e-06,
|
36 |
+
"loss": 2.0942,
|
37 |
+
"step": 15
|
38 |
+
},
|
39 |
+
{
|
40 |
+
"epoch": 0.10825439783491204,
|
41 |
+
"grad_norm": 2.765625,
|
42 |
+
"learning_rate": 1.0810810810810812e-05,
|
43 |
+
"loss": 2.0225,
|
44 |
+
"step": 20
|
45 |
+
},
|
46 |
+
{
|
47 |
+
"epoch": 0.13531799729364005,
|
48 |
+
"grad_norm": 3.40625,
|
49 |
+
"learning_rate": 1.3513513513513515e-05,
|
50 |
+
"loss": 1.9498,
|
51 |
+
"step": 25
|
52 |
+
},
|
53 |
+
{
|
54 |
+
"epoch": 0.16238159675236807,
|
55 |
+
"grad_norm": 2.703125,
|
56 |
+
"learning_rate": 1.6216216216216218e-05,
|
57 |
+
"loss": 1.8531,
|
58 |
+
"step": 30
|
59 |
+
},
|
60 |
+
{
|
61 |
+
"epoch": 0.18944519621109607,
|
62 |
+
"grad_norm": 1.78125,
|
63 |
+
"learning_rate": 1.891891891891892e-05,
|
64 |
+
"loss": 1.7903,
|
65 |
+
"step": 35
|
66 |
+
},
|
67 |
+
{
|
68 |
+
"epoch": 0.2165087956698241,
|
69 |
+
"grad_norm": 1.703125,
|
70 |
+
"learning_rate": 1.9995946530314384e-05,
|
71 |
+
"loss": 1.7538,
|
72 |
+
"step": 40
|
73 |
+
},
|
74 |
+
{
|
75 |
+
"epoch": 0.2435723951285521,
|
76 |
+
"grad_norm": 1.5078125,
|
77 |
+
"learning_rate": 1.9971187226043746e-05,
|
78 |
+
"loss": 1.7264,
|
79 |
+
"step": 45
|
80 |
+
},
|
81 |
+
{
|
82 |
+
"epoch": 0.2706359945872801,
|
83 |
+
"grad_norm": 1.3515625,
|
84 |
+
"learning_rate": 1.9923976226947417e-05,
|
85 |
+
"loss": 1.6995,
|
86 |
+
"step": 50
|
87 |
+
},
|
88 |
+
{
|
89 |
+
"epoch": 0.2976995940460081,
|
90 |
+
"grad_norm": 1.28125,
|
91 |
+
"learning_rate": 1.985441983600819e-05,
|
92 |
+
"loss": 1.6901,
|
93 |
+
"step": 55
|
94 |
+
},
|
95 |
+
{
|
96 |
+
"epoch": 0.32476319350473615,
|
97 |
+
"grad_norm": 1.34375,
|
98 |
+
"learning_rate": 1.9762674670369757e-05,
|
99 |
+
"loss": 1.6784,
|
100 |
+
"step": 60
|
101 |
+
},
|
102 |
+
{
|
103 |
+
"epoch": 0.35182679296346414,
|
104 |
+
"grad_norm": 1.2890625,
|
105 |
+
"learning_rate": 1.9648947308688594e-05,
|
106 |
+
"loss": 1.6738,
|
107 |
+
"step": 65
|
108 |
+
},
|
109 |
+
{
|
110 |
+
"epoch": 0.37889039242219213,
|
111 |
+
"grad_norm": 1.328125,
|
112 |
+
"learning_rate": 1.9513493825989664e-05,
|
113 |
+
"loss": 1.6719,
|
114 |
+
"step": 70
|
115 |
+
},
|
116 |
+
{
|
117 |
+
"epoch": 0.4059539918809202,
|
118 |
+
"grad_norm": 1.2734375,
|
119 |
+
"learning_rate": 1.9356619217073252e-05,
|
120 |
+
"loss": 1.6617,
|
121 |
+
"step": 75
|
122 |
+
},
|
123 |
+
{
|
124 |
+
"epoch": 0.4330175913396482,
|
125 |
+
"grad_norm": 1.2734375,
|
126 |
+
"learning_rate": 1.917867670977126e-05,
|
127 |
+
"loss": 1.6447,
|
128 |
+
"step": 80
|
129 |
+
},
|
130 |
+
{
|
131 |
+
"epoch": 0.46008119079837617,
|
132 |
+
"grad_norm": 1.296875,
|
133 |
+
"learning_rate": 1.8980066969599216e-05,
|
134 |
+
"loss": 1.6337,
|
135 |
+
"step": 85
|
136 |
+
},
|
137 |
+
{
|
138 |
+
"epoch": 0.4871447902571042,
|
139 |
+
"grad_norm": 1.25,
|
140 |
+
"learning_rate": 1.8761237197594945e-05,
|
141 |
+
"loss": 1.6549,
|
142 |
+
"step": 90
|
143 |
+
},
|
144 |
+
{
|
145 |
+
"epoch": 0.5142083897158322,
|
146 |
+
"grad_norm": 1.25,
|
147 |
+
"learning_rate": 1.852268012337514e-05,
|
148 |
+
"loss": 1.6334,
|
149 |
+
"step": 95
|
150 |
+
},
|
151 |
+
{
|
152 |
+
"epoch": 0.5412719891745602,
|
153 |
+
"grad_norm": 1.265625,
|
154 |
+
"learning_rate": 1.8264932895677195e-05,
|
155 |
+
"loss": 1.6276,
|
156 |
+
"step": 100
|
157 |
+
},
|
158 |
+
{
|
159 |
+
"epoch": 0.5683355886332883,
|
160 |
+
"grad_norm": 1.2421875,
|
161 |
+
"learning_rate": 1.798857587288445e-05,
|
162 |
+
"loss": 1.6326,
|
163 |
+
"step": 105
|
164 |
+
},
|
165 |
+
{
|
166 |
+
"epoch": 0.5953991880920162,
|
167 |
+
"grad_norm": 1.203125,
|
168 |
+
"learning_rate": 1.769423131625808e-05,
|
169 |
+
"loss": 1.6334,
|
170 |
+
"step": 110
|
171 |
+
},
|
172 |
+
{
|
173 |
+
"epoch": 0.6224627875507442,
|
174 |
+
"grad_norm": 1.234375,
|
175 |
+
"learning_rate": 1.738256198881809e-05,
|
176 |
+
"loss": 1.6327,
|
177 |
+
"step": 115
|
178 |
+
},
|
179 |
+
{
|
180 |
+
"epoch": 0.6495263870094723,
|
181 |
+
"grad_norm": 1.3125,
|
182 |
+
"learning_rate": 1.7054269663028232e-05,
|
183 |
+
"loss": 1.6271,
|
184 |
+
"step": 120
|
185 |
+
},
|
186 |
+
{
|
187 |
+
"epoch": 0.6765899864682002,
|
188 |
+
"grad_norm": 1.25,
|
189 |
+
"learning_rate": 1.6710093540645056e-05,
|
190 |
+
"loss": 1.6247,
|
191 |
+
"step": 125
|
192 |
+
},
|
193 |
+
{
|
194 |
+
"epoch": 0.7036535859269283,
|
195 |
+
"grad_norm": 1.296875,
|
196 |
+
"learning_rate": 1.6350808588288964e-05,
|
197 |
+
"loss": 1.6255,
|
198 |
+
"step": 130
|
199 |
+
},
|
200 |
+
{
|
201 |
+
"epoch": 0.7307171853856563,
|
202 |
+
"grad_norm": 1.234375,
|
203 |
+
"learning_rate": 1.597722379248512e-05,
|
204 |
+
"loss": 1.6155,
|
205 |
+
"step": 135
|
206 |
+
},
|
207 |
+
{
|
208 |
+
"epoch": 0.7577807848443843,
|
209 |
+
"grad_norm": 1.2421875,
|
210 |
+
"learning_rate": 1.559018033810316e-05,
|
211 |
+
"loss": 1.6162,
|
212 |
+
"step": 140
|
213 |
+
},
|
214 |
+
{
|
215 |
+
"epoch": 0.7848443843031123,
|
216 |
+
"grad_norm": 1.21875,
|
217 |
+
"learning_rate": 1.5190549714297303e-05,
|
218 |
+
"loss": 1.6081,
|
219 |
+
"step": 145
|
220 |
+
},
|
221 |
+
{
|
222 |
+
"epoch": 0.8119079837618404,
|
223 |
+
"grad_norm": 1.21875,
|
224 |
+
"learning_rate": 1.4779231752211546e-05,
|
225 |
+
"loss": 1.6031,
|
226 |
+
"step": 150
|
227 |
+
},
|
228 |
+
{
|
229 |
+
"epoch": 0.8389715832205683,
|
230 |
+
"grad_norm": 1.28125,
|
231 |
+
"learning_rate": 1.4357152598868478e-05,
|
232 |
+
"loss": 1.6155,
|
233 |
+
"step": 155
|
234 |
+
},
|
235 |
+
{
|
236 |
+
"epoch": 0.8660351826792964,
|
237 |
+
"grad_norm": 1.2109375,
|
238 |
+
"learning_rate": 1.3925262631803722e-05,
|
239 |
+
"loss": 1.6039,
|
240 |
+
"step": 160
|
241 |
+
},
|
242 |
+
{
|
243 |
+
"epoch": 0.8930987821380244,
|
244 |
+
"grad_norm": 1.2265625,
|
245 |
+
"learning_rate": 1.3484534319141592e-05,
|
246 |
+
"loss": 1.608,
|
247 |
+
"step": 165
|
248 |
+
},
|
249 |
+
{
|
250 |
+
"epoch": 0.9201623815967523,
|
251 |
+
"grad_norm": 1.1875,
|
252 |
+
"learning_rate": 1.303596002993028e-05,
|
253 |
+
"loss": 1.6036,
|
254 |
+
"step": 170
|
255 |
+
},
|
256 |
+
{
|
257 |
+
"epoch": 0.9472259810554804,
|
258 |
+
"grad_norm": 1.2578125,
|
259 |
+
"learning_rate": 1.2580549799667034e-05,
|
260 |
+
"loss": 1.6157,
|
261 |
+
"step": 175
|
262 |
+
},
|
263 |
+
{
|
264 |
+
"epoch": 0.9742895805142084,
|
265 |
+
"grad_norm": 1.2734375,
|
266 |
+
"learning_rate": 1.2119329056044533e-05,
|
267 |
+
"loss": 1.601,
|
268 |
+
"step": 180
|
269 |
+
},
|
270 |
+
{
|
271 |
+
"epoch": 0.9959404600811907,
|
272 |
+
"eval_loss": 1.6971594095230103,
|
273 |
+
"eval_runtime": 28.8094,
|
274 |
+
"eval_samples_per_second": 15.099,
|
275 |
+
"eval_steps_per_second": 1.909,
|
276 |
+
"step": 184
|
277 |
+
},
|
278 |
+
{
|
279 |
+
"epoch": 1.0013531799729365,
|
280 |
+
"grad_norm": 1.21875,
|
281 |
+
"learning_rate": 1.165333631003928e-05,
|
282 |
+
"loss": 1.5923,
|
283 |
+
"step": 185
|
284 |
+
},
|
285 |
+
{
|
286 |
+
"epoch": 1.0284167794316643,
|
287 |
+
"grad_norm": 1.2109375,
|
288 |
+
"learning_rate": 1.1183620817540985e-05,
|
289 |
+
"loss": 1.5652,
|
290 |
+
"step": 190
|
291 |
+
},
|
292 |
+
{
|
293 |
+
"epoch": 1.0554803788903924,
|
294 |
+
"grad_norm": 1.2265625,
|
295 |
+
"learning_rate": 1.0711240216788036e-05,
|
296 |
+
"loss": 1.5483,
|
297 |
+
"step": 195
|
298 |
+
},
|
299 |
+
{
|
300 |
+
"epoch": 1.0825439783491204,
|
301 |
+
"grad_norm": 1.2578125,
|
302 |
+
"learning_rate": 1.0237258146928849e-05,
|
303 |
+
"loss": 1.5504,
|
304 |
+
"step": 200
|
305 |
+
},
|
306 |
+
{
|
307 |
+
"epoch": 1.1096075778078485,
|
308 |
+
"grad_norm": 1.21875,
|
309 |
+
"learning_rate": 9.762741853071153e-06,
|
310 |
+
"loss": 1.555,
|
311 |
+
"step": 205
|
312 |
+
},
|
313 |
+
{
|
314 |
+
"epoch": 1.1366711772665765,
|
315 |
+
"grad_norm": 1.1796875,
|
316 |
+
"learning_rate": 9.288759783211967e-06,
|
317 |
+
"loss": 1.5659,
|
318 |
+
"step": 210
|
319 |
+
},
|
320 |
+
{
|
321 |
+
"epoch": 1.1637347767253043,
|
322 |
+
"grad_norm": 1.2109375,
|
323 |
+
"learning_rate": 8.81637918245902e-06,
|
324 |
+
"loss": 1.5601,
|
325 |
+
"step": 215
|
326 |
+
},
|
327 |
+
{
|
328 |
+
"epoch": 1.1907983761840324,
|
329 |
+
"grad_norm": 1.1640625,
|
330 |
+
"learning_rate": 8.346663689960724e-06,
|
331 |
+
"loss": 1.5516,
|
332 |
+
"step": 220
|
333 |
+
},
|
334 |
+
{
|
335 |
+
"epoch": 1.2178619756427604,
|
336 |
+
"grad_norm": 1.171875,
|
337 |
+
"learning_rate": 7.880670943955467e-06,
|
338 |
+
"loss": 1.5542,
|
339 |
+
"step": 225
|
340 |
+
},
|
341 |
+
{
|
342 |
+
"epoch": 1.2449255751014885,
|
343 |
+
"grad_norm": 1.203125,
|
344 |
+
"learning_rate": 7.419450200332965e-06,
|
345 |
+
"loss": 1.5491,
|
346 |
+
"step": 230
|
347 |
+
},
|
348 |
+
{
|
349 |
+
"epoch": 1.2719891745602165,
|
350 |
+
"grad_norm": 1.1640625,
|
351 |
+
"learning_rate": 6.964039970069722e-06,
|
352 |
+
"loss": 1.5564,
|
353 |
+
"step": 235
|
354 |
+
},
|
355 |
+
{
|
356 |
+
"epoch": 1.2990527740189446,
|
357 |
+
"grad_norm": 1.1796875,
|
358 |
+
"learning_rate": 6.515465680858412e-06,
|
359 |
+
"loss": 1.5584,
|
360 |
+
"step": 240
|
361 |
+
},
|
362 |
+
{
|
363 |
+
"epoch": 1.3261163734776726,
|
364 |
+
"grad_norm": 1.21875,
|
365 |
+
"learning_rate": 6.074737368196279e-06,
|
366 |
+
"loss": 1.5534,
|
367 |
+
"step": 245
|
368 |
+
},
|
369 |
+
{
|
370 |
+
"epoch": 1.3531799729364005,
|
371 |
+
"grad_norm": 1.2421875,
|
372 |
+
"learning_rate": 5.642847401131526e-06,
|
373 |
+
"loss": 1.5588,
|
374 |
+
"step": 250
|
375 |
+
},
|
376 |
+
{
|
377 |
+
"epoch": 1.3802435723951285,
|
378 |
+
"grad_norm": 1.21875,
|
379 |
+
"learning_rate": 5.220768247788458e-06,
|
380 |
+
"loss": 1.552,
|
381 |
+
"step": 255
|
382 |
+
},
|
383 |
+
{
|
384 |
+
"epoch": 1.4073071718538566,
|
385 |
+
"grad_norm": 1.1875,
|
386 |
+
"learning_rate": 4.809450285702697e-06,
|
387 |
+
"loss": 1.5462,
|
388 |
+
"step": 260
|
389 |
+
},
|
390 |
+
{
|
391 |
+
"epoch": 1.4343707713125846,
|
392 |
+
"grad_norm": 1.203125,
|
393 |
+
"learning_rate": 4.409819661896839e-06,
|
394 |
+
"loss": 1.5623,
|
395 |
+
"step": 265
|
396 |
+
},
|
397 |
+
{
|
398 |
+
"epoch": 1.4614343707713127,
|
399 |
+
"grad_norm": 1.2421875,
|
400 |
+
"learning_rate": 4.022776207514885e-06,
|
401 |
+
"loss": 1.5605,
|
402 |
+
"step": 270
|
403 |
+
},
|
404 |
+
{
|
405 |
+
"epoch": 1.4884979702300405,
|
406 |
+
"grad_norm": 1.2265625,
|
407 |
+
"learning_rate": 3.6491914117110405e-06,
|
408 |
+
"loss": 1.5616,
|
409 |
+
"step": 275
|
410 |
+
},
|
411 |
+
{
|
412 |
+
"epoch": 1.5155615696887685,
|
413 |
+
"grad_norm": 1.1875,
|
414 |
+
"learning_rate": 3.2899064593549477e-06,
|
415 |
+
"loss": 1.5578,
|
416 |
+
"step": 280
|
417 |
+
},
|
418 |
+
{
|
419 |
+
"epoch": 1.5426251691474966,
|
420 |
+
"grad_norm": 1.1953125,
|
421 |
+
"learning_rate": 2.945730336971767e-06,
|
422 |
+
"loss": 1.5482,
|
423 |
+
"step": 285
|
424 |
+
},
|
425 |
+
{
|
426 |
+
"epoch": 1.5696887686062246,
|
427 |
+
"grad_norm": 1.21875,
|
428 |
+
"learning_rate": 2.6174380111819144e-06,
|
429 |
+
"loss": 1.5559,
|
430 |
+
"step": 290
|
431 |
+
},
|
432 |
+
{
|
433 |
+
"epoch": 1.5967523680649527,
|
434 |
+
"grad_norm": 1.21875,
|
435 |
+
"learning_rate": 2.3057686837419246e-06,
|
436 |
+
"loss": 1.5481,
|
437 |
+
"step": 295
|
438 |
+
},
|
439 |
+
{
|
440 |
+
"epoch": 1.6238159675236807,
|
441 |
+
"grad_norm": 1.1640625,
|
442 |
+
"learning_rate": 2.011424127115552e-06,
|
443 |
+
"loss": 1.5412,
|
444 |
+
"step": 300
|
445 |
+
},
|
446 |
+
{
|
447 |
+
"epoch": 1.6508795669824088,
|
448 |
+
"grad_norm": 1.234375,
|
449 |
+
"learning_rate": 1.7350671043228072e-06,
|
450 |
+
"loss": 1.5611,
|
451 |
+
"step": 305
|
452 |
+
},
|
453 |
+
{
|
454 |
+
"epoch": 1.6779431664411368,
|
455 |
+
"grad_norm": 1.203125,
|
456 |
+
"learning_rate": 1.4773198766248642e-06,
|
457 |
+
"loss": 1.5669,
|
458 |
+
"step": 310
|
459 |
+
},
|
460 |
+
{
|
461 |
+
"epoch": 1.7050067658998647,
|
462 |
+
"grad_norm": 1.2109375,
|
463 |
+
"learning_rate": 1.2387628024050557e-06,
|
464 |
+
"loss": 1.5515,
|
465 |
+
"step": 315
|
466 |
+
},
|
467 |
+
{
|
468 |
+
"epoch": 1.7320703653585927,
|
469 |
+
"grad_norm": 1.2109375,
|
470 |
+
"learning_rate": 1.0199330304007858e-06,
|
471 |
+
"loss": 1.5633,
|
472 |
+
"step": 320
|
473 |
+
},
|
474 |
+
{
|
475 |
+
"epoch": 1.7591339648173205,
|
476 |
+
"grad_norm": 1.234375,
|
477 |
+
"learning_rate": 8.213232902287438e-07,
|
478 |
+
"loss": 1.5542,
|
479 |
+
"step": 325
|
480 |
+
},
|
481 |
+
{
|
482 |
+
"epoch": 1.7861975642760486,
|
483 |
+
"grad_norm": 1.2421875,
|
484 |
+
"learning_rate": 6.433807829267491e-07,
|
485 |
+
"loss": 1.5575,
|
486 |
+
"step": 330
|
487 |
+
},
|
488 |
+
{
|
489 |
+
"epoch": 1.8132611637347766,
|
490 |
+
"grad_norm": 1.1953125,
|
491 |
+
"learning_rate": 4.865061740103361e-07,
|
492 |
+
"loss": 1.5532,
|
493 |
+
"step": 335
|
494 |
+
},
|
495 |
+
{
|
496 |
+
"epoch": 1.8403247631935047,
|
497 |
+
"grad_norm": 1.21875,
|
498 |
+
"learning_rate": 3.510526913114065e-07,
|
499 |
+
"loss": 1.5686,
|
500 |
+
"step": 340
|
501 |
+
},
|
502 |
+
{
|
503 |
+
"epoch": 1.8673883626522327,
|
504 |
+
"grad_norm": 1.1953125,
|
505 |
+
"learning_rate": 2.3732532963024468e-07,
|
506 |
+
"loss": 1.5484,
|
507 |
+
"step": 345
|
508 |
+
},
|
509 |
+
{
|
510 |
+
"epoch": 1.8944519621109608,
|
511 |
+
"grad_norm": 1.1953125,
|
512 |
+
"learning_rate": 1.4558016399181086e-07,
|
513 |
+
"loss": 1.56,
|
514 |
+
"step": 350
|
515 |
+
},
|
516 |
+
{
|
517 |
+
"epoch": 1.9215155615696888,
|
518 |
+
"grad_norm": 1.1953125,
|
519 |
+
"learning_rate": 7.602377305258479e-08,
|
520 |
+
"loss": 1.5479,
|
521 |
+
"step": 355
|
522 |
+
},
|
523 |
+
{
|
524 |
+
"epoch": 1.9485791610284169,
|
525 |
+
"grad_norm": 1.1796875,
|
526 |
+
"learning_rate": 2.8812773956256034e-08,
|
527 |
+
"loss": 1.5456,
|
528 |
+
"step": 360
|
529 |
+
},
|
530 |
+
{
|
531 |
+
"epoch": 1.975642760487145,
|
532 |
+
"grad_norm": 1.2265625,
|
533 |
+
"learning_rate": 4.053469685617595e-09,
|
534 |
+
"loss": 1.5526,
|
535 |
+
"step": 365
|
536 |
+
},
|
537 |
+
{
|
538 |
+
"epoch": 1.9918809201623815,
|
539 |
+
"eval_loss": 1.697079062461853,
|
540 |
+
"eval_runtime": 28.912,
|
541 |
+
"eval_samples_per_second": 15.046,
|
542 |
+
"eval_steps_per_second": 1.902,
|
543 |
+
"step": 368
|
544 |
+
},
|
545 |
+
{
|
546 |
+
"epoch": 1.9918809201623815,
|
547 |
+
"step": 368,
|
548 |
+
"total_flos": 1.4968483831454106e+17,
|
549 |
+
"train_loss": 1.6339908747569374,
|
550 |
+
"train_runtime": 3216.6871,
|
551 |
+
"train_samples_per_second": 3.675,
|
552 |
+
"train_steps_per_second": 0.114
|
553 |
+
}
|
554 |
+
],
|
555 |
+
"logging_steps": 5,
|
556 |
+
"max_steps": 368,
|
557 |
+
"num_input_tokens_seen": 0,
|
558 |
+
"num_train_epochs": 2,
|
559 |
+
"save_steps": 100,
|
560 |
+
"stateful_callbacks": {
|
561 |
+
"TrainerControl": {
|
562 |
+
"args": {
|
563 |
+
"should_epoch_stop": false,
|
564 |
+
"should_evaluate": false,
|
565 |
+
"should_log": false,
|
566 |
+
"should_save": true,
|
567 |
+
"should_training_stop": false
|
568 |
+
},
|
569 |
+
"attributes": {}
|
570 |
+
}
|
571 |
+
},
|
572 |
+
"total_flos": 1.4968483831454106e+17,
|
573 |
+
"train_batch_size": 8,
|
574 |
+
"trial_name": null,
|
575 |
+
"trial_params": null
|
576 |
+
}
|