Update README.md
Browse files
README.md
CHANGED
@@ -1,95 +1,11 @@
|
|
1 |
-
---
|
2 |
-
|
3 |
-
|
4 |
-
|
5 |
-
tags:
|
6 |
-
- axolotl
|
7 |
-
- generated_from_trainer
|
8 |
-
model-index:
|
9 |
-
- name: EvolCodeLlama-3.1-8B-Instruct
|
10 |
-
results: []
|
11 |
-
---
|
12 |
|
13 |
<!-- This model card has been generated automatically according to the information the Trainer had access to. You
|
14 |
should probably proofread and complete it, then remove this comment. -->
|
15 |
|
16 |
-
[<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)
|
17 |
-
<details><summary>See axolotl config</summary>
|
18 |
-
|
19 |
-
axolotl version: `0.4.1`
|
20 |
-
```yaml
|
21 |
-
base_model: meta-llama/Meta-Llama-3.1-8B-Instruct
|
22 |
-
model_type: LlamaForCausalLM
|
23 |
-
tokenizer_type: AutoTokenizer
|
24 |
-
is_llama_derived_model: true
|
25 |
-
hub_model_id: EvolCodeLlama-3.1-8B-Instruct
|
26 |
-
|
27 |
-
load_in_8bit: false
|
28 |
-
load_in_4bit: true
|
29 |
-
strict: false
|
30 |
-
|
31 |
-
datasets:
|
32 |
-
- path: mlabonne/Evol-Instruct-Python-1k
|
33 |
-
type: alpaca
|
34 |
-
dataset_prepared_path: last_run_prepared
|
35 |
-
val_set_size: 0.02
|
36 |
-
output_dir: ./qlora-out
|
37 |
-
|
38 |
-
adapter: qlora
|
39 |
-
lora_model_dir:
|
40 |
-
|
41 |
-
sequence_len: 2048
|
42 |
-
sample_packing: true
|
43 |
-
|
44 |
-
lora_r: 32
|
45 |
-
lora_alpha: 16
|
46 |
-
lora_dropout: 0.05
|
47 |
-
lora_target_modules:
|
48 |
-
lora_target_linear: true
|
49 |
-
lora_fan_in_fan_out:
|
50 |
-
|
51 |
-
wandb_project: axolotl
|
52 |
-
wandb_entity:
|
53 |
-
wandb_watch:
|
54 |
-
wandb_run_id:
|
55 |
-
wandb_log_model:
|
56 |
-
|
57 |
-
gradient_accumulation_steps: 4
|
58 |
-
micro_batch_size: 2
|
59 |
-
num_epochs: 3
|
60 |
-
optimizer: paged_adamw_32bit
|
61 |
-
lr_scheduler: cosine
|
62 |
-
learning_rate: 0.0002
|
63 |
-
|
64 |
-
train_on_inputs: false
|
65 |
-
group_by_length: false
|
66 |
-
bf16: true
|
67 |
-
fp16: false
|
68 |
-
tf32: false
|
69 |
-
|
70 |
-
gradient_checkpointing: true
|
71 |
-
early_stopping_patience:
|
72 |
-
resume_from_checkpoint:
|
73 |
-
local_rank:
|
74 |
-
logging_steps: 1
|
75 |
-
xformers_attention:
|
76 |
-
flash_attention: true
|
77 |
-
|
78 |
-
warmup_steps: 100
|
79 |
-
eval_steps: 0.01
|
80 |
-
save_strategy: epoch
|
81 |
-
save_steps:
|
82 |
-
debug:
|
83 |
-
deepspeed:
|
84 |
-
weight_decay: 0.0
|
85 |
-
fsdp:
|
86 |
-
fsdp_config:
|
87 |
-
special_tokens:
|
88 |
-
pad_token: "<|end_of_text|>"
|
89 |
-
|
90 |
-
```
|
91 |
-
|
92 |
-
</details><br>
|
93 |
|
94 |
# EvolCodeLlama-3.1-8B-Instruct
|
95 |
|
@@ -99,7 +15,7 @@ It achieves the following results on the evaluation set:
|
|
99 |
|
100 |
## Training:
|
101 |
|
102 |
-
It was trained on an **A40** for more than 1 hour
|
103 |
|
104 |
### Training hyperparameters
|
105 |
|
|
|
1 |
+
---
|
2 |
+
license: apache-2.0
|
3 |
+
base_model: meta-llama/Meta-Llama-3.1-8B-Instruct
|
4 |
+
---
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
5 |
|
6 |
<!-- This model card has been generated automatically according to the information the Trainer had access to. You
|
7 |
should probably proofread and complete it, then remove this comment. -->
|
8 |
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
9 |
|
10 |
# EvolCodeLlama-3.1-8B-Instruct
|
11 |
|
|
|
15 |
|
16 |
## Training:
|
17 |
|
18 |
+
It was trained on an **A40** for more than 1 hour using Axolotl.
|
19 |
|
20 |
### Training hyperparameters
|
21 |
|