Commit
•
52f5343
1
Parent(s):
6e1277f
Upload folder using huggingface_hub
Browse files- .gitattributes +5 -35
- README.md +187 -0
- dolphin-2.9.1-llama-3-8b-Q3_K_L.gguf +3 -0
- dolphin-2.9.1-llama-3-8b-Q4_K_M.gguf +3 -0
- dolphin-2.9.1-llama-3-8b-Q5_K_M.gguf +3 -0
- dolphin-2.9.1-llama-3-8b-Q6_K.gguf +3 -0
- dolphin-2.9.1-llama-3-8b-Q8_0.gguf +3 -0
.gitattributes
CHANGED
@@ -1,35 +1,5 @@
|
|
1 |
-
|
2 |
-
|
3 |
-
|
4 |
-
|
5 |
-
|
6 |
-
*.ftz filter=lfs diff=lfs merge=lfs -text
|
7 |
-
*.gz filter=lfs diff=lfs merge=lfs -text
|
8 |
-
*.h5 filter=lfs diff=lfs merge=lfs -text
|
9 |
-
*.joblib filter=lfs diff=lfs merge=lfs -text
|
10 |
-
*.lfs.* filter=lfs diff=lfs merge=lfs -text
|
11 |
-
*.mlmodel filter=lfs diff=lfs merge=lfs -text
|
12 |
-
*.model filter=lfs diff=lfs merge=lfs -text
|
13 |
-
*.msgpack filter=lfs diff=lfs merge=lfs -text
|
14 |
-
*.npy filter=lfs diff=lfs merge=lfs -text
|
15 |
-
*.npz filter=lfs diff=lfs merge=lfs -text
|
16 |
-
*.onnx filter=lfs diff=lfs merge=lfs -text
|
17 |
-
*.ot filter=lfs diff=lfs merge=lfs -text
|
18 |
-
*.parquet filter=lfs diff=lfs merge=lfs -text
|
19 |
-
*.pb filter=lfs diff=lfs merge=lfs -text
|
20 |
-
*.pickle filter=lfs diff=lfs merge=lfs -text
|
21 |
-
*.pkl filter=lfs diff=lfs merge=lfs -text
|
22 |
-
*.pt filter=lfs diff=lfs merge=lfs -text
|
23 |
-
*.pth filter=lfs diff=lfs merge=lfs -text
|
24 |
-
*.rar filter=lfs diff=lfs merge=lfs -text
|
25 |
-
*.safetensors filter=lfs diff=lfs merge=lfs -text
|
26 |
-
saved_model/**/* filter=lfs diff=lfs merge=lfs -text
|
27 |
-
*.tar.* filter=lfs diff=lfs merge=lfs -text
|
28 |
-
*.tar filter=lfs diff=lfs merge=lfs -text
|
29 |
-
*.tflite filter=lfs diff=lfs merge=lfs -text
|
30 |
-
*.tgz filter=lfs diff=lfs merge=lfs -text
|
31 |
-
*.wasm filter=lfs diff=lfs merge=lfs -text
|
32 |
-
*.xz filter=lfs diff=lfs merge=lfs -text
|
33 |
-
*.zip filter=lfs diff=lfs merge=lfs -text
|
34 |
-
*.zst filter=lfs diff=lfs merge=lfs -text
|
35 |
-
*tfevents* filter=lfs diff=lfs merge=lfs -text
|
|
|
1 |
+
dolphin-2.9.1-llama-3-8b-Q3_K_L.gguf filter=lfs diff=lfs merge=lfs -text
|
2 |
+
dolphin-2.9.1-llama-3-8b-Q4_K_M.gguf filter=lfs diff=lfs merge=lfs -text
|
3 |
+
dolphin-2.9.1-llama-3-8b-Q5_K_M.gguf filter=lfs diff=lfs merge=lfs -text
|
4 |
+
dolphin-2.9.1-llama-3-8b-Q6_K.gguf filter=lfs diff=lfs merge=lfs -text
|
5 |
+
dolphin-2.9.1-llama-3-8b-Q8_0.gguf filter=lfs diff=lfs merge=lfs -text
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
README.md
ADDED
@@ -0,0 +1,187 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
1 |
+
---
|
2 |
+
license: other
|
3 |
+
base_model: meta-llama/Meta-Llama-3-8B
|
4 |
+
tags:
|
5 |
+
- generated_from_trainer
|
6 |
+
- axolotl
|
7 |
+
model-index:
|
8 |
+
- name: out
|
9 |
+
results: []
|
10 |
+
datasets:
|
11 |
+
- cognitivecomputations/Dolphin-2.9
|
12 |
+
- teknium/OpenHermes-2.5
|
13 |
+
- m-a-p/CodeFeedback-Filtered-Instruction
|
14 |
+
- cognitivecomputations/dolphin-coder
|
15 |
+
- cognitivecomputations/samantha-data
|
16 |
+
- microsoft/orca-math-word-problems-200k
|
17 |
+
- Locutusque/function-calling-chatml
|
18 |
+
- internlm/Agent-FLAN
|
19 |
+
---
|
20 |
+
|
21 |
+
# Dolphin 2.9.1 Llama 3 8b 🐬
|
22 |
+
|
23 |
+
Curated and trained by Eric Hartford, Lucas Atkins, and Fernando Fernandes, and Cognitive Computations
|
24 |
+
|
25 |
+
Discord: https://discord.gg/8fbBeC7ZGx
|
26 |
+
|
27 |
+
<img src="https://cdn-uploads.huggingface.co/production/uploads/63111b2d88942700629f5771/ldkN1J0WIDQwU4vutGYiD.png" width="600" />
|
28 |
+
|
29 |
+
We have retrained our LLama-3-8b fine tune to address behavioral issues in the initial 2.9 dataset. Specifically, Systemchat was causing the model to be *too* reliant on the system prompt. Additionally, it had an occasional quirk that would cause the model to overly reference the system prompt. We also found generation length was at times not sufficient for any given task. We identified the culprit as Ultrachat. Accounting for these concerns, we removed systemchat and ultrachat from the dataset. It is otherwise identical to dolphin-2.9.
|
30 |
+
|
31 |
+
Our appreciation for the sponsors of Dolphin 2.9.1:
|
32 |
+
- [Crusoe Cloud](https://crusoe.ai/) - provided excellent on-demand 8xL40S node
|
33 |
+
|
34 |
+
This model is based on Llama-3-8b, and is governed by [META LLAMA 3 COMMUNITY LICENSE AGREEMENT](LICENSE)
|
35 |
+
|
36 |
+
The base model has 8k context, and the full-weight fine-tuning was with 4k sequence length.
|
37 |
+
|
38 |
+
It took 1.5 days on an 8x L40S provided by Crusoe Cloud
|
39 |
+
|
40 |
+
This model was trained FFT on all parameters, using ChatML prompt template format.
|
41 |
+
|
42 |
+
example:
|
43 |
+
|
44 |
+
```
|
45 |
+
<|im_start|>system
|
46 |
+
You are Dolphin, a helpful AI assistant.<|im_end|>
|
47 |
+
<|im_start|>user
|
48 |
+
{prompt}<|im_end|>
|
49 |
+
<|im_start|>assistant
|
50 |
+
|
51 |
+
```
|
52 |
+
|
53 |
+
Dolphin-2.9.1 has a variety of instruction, conversational, and coding skills. It also has initial agentic abilities and supports function calling.
|
54 |
+
|
55 |
+
Dolphin is uncensored. We have filtered the dataset to remove alignment and bias. This makes the model more compliant. You are advised to implement your own alignment layer before exposing the model as a service. It will be highly compliant with any requests, even unethical ones. Please read my blog post about uncensored models. https://erichartford.com/uncensored-models You are responsible for any content you create using this model. Enjoy responsibly.
|
56 |
+
|
57 |
+
Dolphin is licensed according to Meta's Llama license. We grant permission for any use, including commercial, that falls within accordance with Meta's Llama-3 license. Dolphin was trained on data generated from GPT4, among other models.
|
58 |
+
|
59 |
+
## Evals
|
60 |
+
|
61 |
+
![image/png](https://cdn-uploads.huggingface.co/production/uploads/63111b2d88942700629f5771/0pqSc8jsJlhBH8dcgpwE7.png)
|
62 |
+
|
63 |
+
## Training
|
64 |
+
|
65 |
+
[<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)
|
66 |
+
<details><summary>See axolotl config</summary>
|
67 |
+
|
68 |
+
axolotl version: `0.4.0`
|
69 |
+
```yaml
|
70 |
+
base_model: meta-llama/Meta-Llama-3-8B
|
71 |
+
model_type: AutoModelForCausalLM
|
72 |
+
tokenizer_type: AutoTokenizer
|
73 |
+
tokenizer_use_fast: false
|
74 |
+
|
75 |
+
|
76 |
+
load_in_8bit: false
|
77 |
+
load_in_4bit: false
|
78 |
+
strict: false
|
79 |
+
model_config:
|
80 |
+
|
81 |
+
datasets:
|
82 |
+
- path: /workspace/datasets/dolphin-2.9/dolphin201-sharegpt2.jsonl
|
83 |
+
type: sharegpt
|
84 |
+
conversation: chatml
|
85 |
+
- path: /workspace/datasets/dolphin-2.9/dolphin-coder-translate-sharegpt2.jsonl
|
86 |
+
type: sharegpt
|
87 |
+
conversation: chatml
|
88 |
+
- path: /workspace/datasets/dolphin-2.9/dolphin-coder-codegen-sharegpt2.jsonl
|
89 |
+
type: sharegpt
|
90 |
+
conversation: chatml
|
91 |
+
- path: /workspace/datasets/dolphin-2.9/m-a-p_Code-Feedback-sharegpt-unfiltered.jsonl
|
92 |
+
type: sharegpt
|
93 |
+
conversation: chatml
|
94 |
+
- path: /workspace/datasets/dolphin-2.9/m-a-p_CodeFeedback-Filtered-Instruction-sharegpt-unfiltered.jsonl
|
95 |
+
type: sharegpt
|
96 |
+
conversation: chatml
|
97 |
+
- path: /workspace/datasets/dolphin-2.9/not_samantha_norefusals.jsonl
|
98 |
+
type: sharegpt
|
99 |
+
conversation: chatml
|
100 |
+
- path: /workspace/datasets/dolphin-2.9/Orca-Math-resort-unfiltered.jsonl
|
101 |
+
type: sharegpt
|
102 |
+
conversation: chatml
|
103 |
+
- path: /workspace/datasets/dolphin-2.9/agent_instruct_react_unfiltered.jsonl
|
104 |
+
type: sharegpt
|
105 |
+
conversation: chatml
|
106 |
+
- path: /workspace/datasets/dolphin-2.9/toolbench_instruct_j1s1_3k_unfiltered.jsonl
|
107 |
+
type: sharegpt
|
108 |
+
conversation: chatml
|
109 |
+
- path: /workspace/datasets/dolphin-2.9/toolbench_negative_unfiltered.jsonl
|
110 |
+
type: sharegpt
|
111 |
+
conversation: chatml
|
112 |
+
- path: /workspace/datasets/dolphin-2.9/toolbench_react_10p_unfiltered.jsonl
|
113 |
+
type: sharegpt
|
114 |
+
conversation: chatml
|
115 |
+
- path: /workspace/datasets/dolphin-2.9/toolbench_tflan_cot_30p_unfiltered.jsonl
|
116 |
+
type: sharegpt
|
117 |
+
conversation: chatml
|
118 |
+
- path: /workspace/datasets/dolphin-2.9/openhermes200k_unfiltered.jsonl
|
119 |
+
type: sharegpt
|
120 |
+
conversation: chatml
|
121 |
+
|
122 |
+
chat_template: chatml
|
123 |
+
|
124 |
+
|
125 |
+
dataset_prepared_path: /workspace/datasets/dolphin-2.9/thingy
|
126 |
+
val_set_size: 0.0002
|
127 |
+
output_dir: ./out
|
128 |
+
|
129 |
+
sequence_len: 4096
|
130 |
+
sample_packing: true
|
131 |
+
pad_to_sequence_len: true
|
132 |
+
|
133 |
+
gradient_accumulation_steps: 4
|
134 |
+
micro_batch_size: 3
|
135 |
+
num_epochs: 3
|
136 |
+
logging_steps: 1
|
137 |
+
optimizer: adamw_8bit
|
138 |
+
lr_scheduler: cosine
|
139 |
+
learning_rate: 2e-5
|
140 |
+
|
141 |
+
wandb_project: dolphin-2.9-mixtral-8x22b
|
142 |
+
wandb_watch:
|
143 |
+
wandb_run_id:
|
144 |
+
wandb_log_model:
|
145 |
+
|
146 |
+
train_on_inputs: false
|
147 |
+
group_by_length: false
|
148 |
+
bf16: auto
|
149 |
+
fp16:
|
150 |
+
tf32: false
|
151 |
+
|
152 |
+
gradient_checkpointing: true
|
153 |
+
gradient_checkpointing_kwargs:
|
154 |
+
use_reentrant: false
|
155 |
+
early_stopping_patience:
|
156 |
+
resume_from_checkpoint:
|
157 |
+
local_rank:
|
158 |
+
logging_steps: 1
|
159 |
+
xformers_attention:
|
160 |
+
flash_attention: true
|
161 |
+
saves_per_epoch: 4
|
162 |
+
save_total_limit: 2
|
163 |
+
save_steps:
|
164 |
+
evals_per_epoch: 4
|
165 |
+
eval_sample_packing: false
|
166 |
+
debug:
|
167 |
+
deepspeed: deepspeed_configs/zero3_bf16.json
|
168 |
+
weight_decay: 0.05
|
169 |
+
fsdp:
|
170 |
+
fsdp_config:
|
171 |
+
special_tokens:
|
172 |
+
eos_token: "<|im_end|>"
|
173 |
+
pad_token: "<|end_of_text|>"
|
174 |
+
tokens:
|
175 |
+
- "<|im_start|>"
|
176 |
+
- "<|im_end|>"
|
177 |
+
|
178 |
+
```
|
179 |
+
|
180 |
+
</details><br>
|
181 |
+
|
182 |
+
### Framework versions
|
183 |
+
|
184 |
+
- Transformers 4.40.0
|
185 |
+
- Pytorch 2.2.2+cu121
|
186 |
+
- Datasets 2.18.0
|
187 |
+
- Tokenizers 0.19.1
|
dolphin-2.9.1-llama-3-8b-Q3_K_L.gguf
ADDED
@@ -0,0 +1,3 @@
|
|
|
|
|
|
|
|
|
1 |
+
version https://git-lfs.github.com/spec/v1
|
2 |
+
oid sha256:b298425cf39c558a608554ad977c86129ced7d00d17ca46d69c37f6a6bbe0c41
|
3 |
+
size 4321966368
|
dolphin-2.9.1-llama-3-8b-Q4_K_M.gguf
ADDED
@@ -0,0 +1,3 @@
|
|
|
|
|
|
|
|
|
1 |
+
version https://git-lfs.github.com/spec/v1
|
2 |
+
oid sha256:e3d2ab9d4ca41be70c39f00ad3f8b0f890bd32999a6138336628332d23014045
|
3 |
+
size 4920745312
|
dolphin-2.9.1-llama-3-8b-Q5_K_M.gguf
ADDED
@@ -0,0 +1,3 @@
|
|
|
|
|
|
|
|
|
1 |
+
version https://git-lfs.github.com/spec/v1
|
2 |
+
oid sha256:0e4a1637ca10fde28829627b1adb9e7e5611bf8d7fbf5d3a0ea7e2e3ca62a642
|
3 |
+
size 5732999520
|
dolphin-2.9.1-llama-3-8b-Q6_K.gguf
ADDED
@@ -0,0 +1,3 @@
|
|
|
|
|
|
|
|
|
1 |
+
version https://git-lfs.github.com/spec/v1
|
2 |
+
oid sha256:d9e4fd16806d9af4d5cec83dc2153fbaa1b9d9c2e801e19805d110e9c553098a
|
3 |
+
size 6596019616
|
dolphin-2.9.1-llama-3-8b-Q8_0.gguf
ADDED
@@ -0,0 +1,3 @@
|
|
|
|
|
|
|
|
|
1 |
+
version https://git-lfs.github.com/spec/v1
|
2 |
+
oid sha256:4ec41846af26d9d6c2eb544e8047ff17c5bec82c3266b205ed0f7d91a6c16aac
|
3 |
+
size 8540788000
|