Model save
Browse files
README.md
CHANGED
@@ -2,7 +2,6 @@
|
|
2 |
license: apache-2.0
|
3 |
library_name: peft
|
4 |
tags:
|
5 |
-
- axolotl
|
6 |
- generated_from_trainer
|
7 |
base_model: mistralai/Mistral-7B-Instruct-v0.2
|
8 |
model-index:
|
@@ -60,7 +59,7 @@ wandb_log_model: end
|
|
60 |
|
61 |
gradient_accumulation_steps: 4
|
62 |
micro_batch_size: 2
|
63 |
-
num_epochs:
|
64 |
optimizer: paged_adamw_8bit
|
65 |
lr_scheduler: cosine
|
66 |
learning_rate: 0.001
|
@@ -103,9 +102,7 @@ fsdp_config:
|
|
103 |
|
104 |
# mistral-lora
|
105 |
|
106 |
-
This model is a fine-tuned version of [mistralai/Mistral-7B-Instruct-v0.2](https://huggingface.co/mistralai/Mistral-7B-Instruct-v0.2) on
|
107 |
-
It achieves the following results on the evaluation set:
|
108 |
-
- Loss: 0.1480
|
109 |
|
110 |
## Model description
|
111 |
|
@@ -136,18 +133,7 @@ The following hyperparameters were used during training:
|
|
136 |
- optimizer: Adam with betas=(0.9,0.95) and epsilon=1e-05
|
137 |
- lr_scheduler_type: cosine
|
138 |
- lr_scheduler_warmup_steps: 10
|
139 |
-
- num_epochs:
|
140 |
-
|
141 |
-
### Training results
|
142 |
-
|
143 |
-
| Training Loss | Epoch | Step | Validation Loss |
|
144 |
-
|:-------------:|:-----:|:----:|:---------------:|
|
145 |
-
| 1.3787 | 0.0 | 1 | 1.4156 |
|
146 |
-
| 0.0868 | 0.1 | 31 | 0.1745 |
|
147 |
-
| 0.149 | 0.21 | 62 | 0.1603 |
|
148 |
-
| 0.1328 | 0.31 | 93 | 0.1532 |
|
149 |
-
| 0.1635 | 0.41 | 124 | 0.1480 |
|
150 |
-
|
151 |
|
152 |
### Framework versions
|
153 |
|
|
|
2 |
license: apache-2.0
|
3 |
library_name: peft
|
4 |
tags:
|
|
|
5 |
- generated_from_trainer
|
6 |
base_model: mistralai/Mistral-7B-Instruct-v0.2
|
7 |
model-index:
|
|
|
59 |
|
60 |
gradient_accumulation_steps: 4
|
61 |
micro_batch_size: 2
|
62 |
+
num_epochs: 2
|
63 |
optimizer: paged_adamw_8bit
|
64 |
lr_scheduler: cosine
|
65 |
learning_rate: 0.001
|
|
|
102 |
|
103 |
# mistral-lora
|
104 |
|
105 |
+
This model is a fine-tuned version of [mistralai/Mistral-7B-Instruct-v0.2](https://huggingface.co/mistralai/Mistral-7B-Instruct-v0.2) on an unknown dataset.
|
|
|
|
|
106 |
|
107 |
## Model description
|
108 |
|
|
|
133 |
- optimizer: Adam with betas=(0.9,0.95) and epsilon=1e-05
|
134 |
- lr_scheduler_type: cosine
|
135 |
- lr_scheduler_warmup_steps: 10
|
136 |
+
- num_epochs: 2
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
137 |
|
138 |
### Framework versions
|
139 |
|