RichardErkhov commited on
Commit
617d6e9
1 Parent(s): 77c23b2

uploaded readme

Browse files
Files changed (1) hide show
  1. README.md +251 -0
README.md ADDED
@@ -0,0 +1,251 @@
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
1
+ Quantization made by Richard Erkhov.
2
+
3
+ [Github](https://github.com/RichardErkhov)
4
+
5
+ [Discord](https://discord.gg/pvy7H8DZMG)
6
+
7
+ [Request more models](https://github.com/RichardErkhov/quant_request)
8
+
9
+
10
+ Chunky-Lemon-Cookie-11B - GGUF
11
+ - Model creator: https://huggingface.co/FallenMerick/
12
+ - Original model: https://huggingface.co/FallenMerick/Chunky-Lemon-Cookie-11B/
13
+
14
+
15
+ | Name | Quant method | Size |
16
+ | ---- | ---- | ---- |
17
+ | [Chunky-Lemon-Cookie-11B.Q2_K.gguf](https://huggingface.co/RichardErkhov/FallenMerick_-_Chunky-Lemon-Cookie-11B-gguf/blob/main/Chunky-Lemon-Cookie-11B.Q2_K.gguf) | Q2_K | 3.73GB |
18
+ | [Chunky-Lemon-Cookie-11B.IQ3_XS.gguf](https://huggingface.co/RichardErkhov/FallenMerick_-_Chunky-Lemon-Cookie-11B-gguf/blob/main/Chunky-Lemon-Cookie-11B.IQ3_XS.gguf) | IQ3_XS | 4.14GB |
19
+ | [Chunky-Lemon-Cookie-11B.IQ3_S.gguf](https://huggingface.co/RichardErkhov/FallenMerick_-_Chunky-Lemon-Cookie-11B-gguf/blob/main/Chunky-Lemon-Cookie-11B.IQ3_S.gguf) | IQ3_S | 4.37GB |
20
+ | [Chunky-Lemon-Cookie-11B.Q3_K_S.gguf](https://huggingface.co/RichardErkhov/FallenMerick_-_Chunky-Lemon-Cookie-11B-gguf/blob/main/Chunky-Lemon-Cookie-11B.Q3_K_S.gguf) | Q3_K_S | 4.34GB |
21
+ | [Chunky-Lemon-Cookie-11B.IQ3_M.gguf](https://huggingface.co/RichardErkhov/FallenMerick_-_Chunky-Lemon-Cookie-11B-gguf/blob/main/Chunky-Lemon-Cookie-11B.IQ3_M.gguf) | IQ3_M | 4.51GB |
22
+ | [Chunky-Lemon-Cookie-11B.Q3_K.gguf](https://huggingface.co/RichardErkhov/FallenMerick_-_Chunky-Lemon-Cookie-11B-gguf/blob/main/Chunky-Lemon-Cookie-11B.Q3_K.gguf) | Q3_K | 4.84GB |
23
+ | [Chunky-Lemon-Cookie-11B.Q3_K_M.gguf](https://huggingface.co/RichardErkhov/FallenMerick_-_Chunky-Lemon-Cookie-11B-gguf/blob/main/Chunky-Lemon-Cookie-11B.Q3_K_M.gguf) | Q3_K_M | 4.84GB |
24
+ | [Chunky-Lemon-Cookie-11B.Q3_K_L.gguf](https://huggingface.co/RichardErkhov/FallenMerick_-_Chunky-Lemon-Cookie-11B-gguf/blob/main/Chunky-Lemon-Cookie-11B.Q3_K_L.gguf) | Q3_K_L | 5.26GB |
25
+ | [Chunky-Lemon-Cookie-11B.IQ4_XS.gguf](https://huggingface.co/RichardErkhov/FallenMerick_-_Chunky-Lemon-Cookie-11B-gguf/blob/main/Chunky-Lemon-Cookie-11B.IQ4_XS.gguf) | IQ4_XS | 5.43GB |
26
+ | [Chunky-Lemon-Cookie-11B.Q4_0.gguf](https://huggingface.co/RichardErkhov/FallenMerick_-_Chunky-Lemon-Cookie-11B-gguf/blob/main/Chunky-Lemon-Cookie-11B.Q4_0.gguf) | Q4_0 | 5.66GB |
27
+ | [Chunky-Lemon-Cookie-11B.IQ4_NL.gguf](https://huggingface.co/RichardErkhov/FallenMerick_-_Chunky-Lemon-Cookie-11B-gguf/blob/main/Chunky-Lemon-Cookie-11B.IQ4_NL.gguf) | IQ4_NL | 5.72GB |
28
+ | [Chunky-Lemon-Cookie-11B.Q4_K_S.gguf](https://huggingface.co/RichardErkhov/FallenMerick_-_Chunky-Lemon-Cookie-11B-gguf/blob/main/Chunky-Lemon-Cookie-11B.Q4_K_S.gguf) | Q4_K_S | 5.7GB |
29
+ | [Chunky-Lemon-Cookie-11B.Q4_K.gguf](https://huggingface.co/RichardErkhov/FallenMerick_-_Chunky-Lemon-Cookie-11B-gguf/blob/main/Chunky-Lemon-Cookie-11B.Q4_K.gguf) | Q4_K | 6.02GB |
30
+ | [Chunky-Lemon-Cookie-11B.Q4_K_M.gguf](https://huggingface.co/RichardErkhov/FallenMerick_-_Chunky-Lemon-Cookie-11B-gguf/blob/main/Chunky-Lemon-Cookie-11B.Q4_K_M.gguf) | Q4_K_M | 6.02GB |
31
+ | [Chunky-Lemon-Cookie-11B.Q4_1.gguf](https://huggingface.co/RichardErkhov/FallenMerick_-_Chunky-Lemon-Cookie-11B-gguf/blob/main/Chunky-Lemon-Cookie-11B.Q4_1.gguf) | Q4_1 | 6.27GB |
32
+ | [Chunky-Lemon-Cookie-11B.Q5_0.gguf](https://huggingface.co/RichardErkhov/FallenMerick_-_Chunky-Lemon-Cookie-11B-gguf/blob/main/Chunky-Lemon-Cookie-11B.Q5_0.gguf) | Q5_0 | 6.89GB |
33
+ | [Chunky-Lemon-Cookie-11B.Q5_K_S.gguf](https://huggingface.co/RichardErkhov/FallenMerick_-_Chunky-Lemon-Cookie-11B-gguf/blob/main/Chunky-Lemon-Cookie-11B.Q5_K_S.gguf) | Q5_K_S | 6.89GB |
34
+ | [Chunky-Lemon-Cookie-11B.Q5_K.gguf](https://huggingface.co/RichardErkhov/FallenMerick_-_Chunky-Lemon-Cookie-11B-gguf/blob/main/Chunky-Lemon-Cookie-11B.Q5_K.gguf) | Q5_K | 7.08GB |
35
+ | [Chunky-Lemon-Cookie-11B.Q5_K_M.gguf](https://huggingface.co/RichardErkhov/FallenMerick_-_Chunky-Lemon-Cookie-11B-gguf/blob/main/Chunky-Lemon-Cookie-11B.Q5_K_M.gguf) | Q5_K_M | 7.08GB |
36
+ | [Chunky-Lemon-Cookie-11B.Q5_1.gguf](https://huggingface.co/RichardErkhov/FallenMerick_-_Chunky-Lemon-Cookie-11B-gguf/blob/main/Chunky-Lemon-Cookie-11B.Q5_1.gguf) | Q5_1 | 7.51GB |
37
+ | [Chunky-Lemon-Cookie-11B.Q6_K.gguf](https://huggingface.co/RichardErkhov/FallenMerick_-_Chunky-Lemon-Cookie-11B-gguf/blob/main/Chunky-Lemon-Cookie-11B.Q6_K.gguf) | Q6_K | 8.2GB |
38
+ | [Chunky-Lemon-Cookie-11B.Q8_0.gguf](https://huggingface.co/RichardErkhov/FallenMerick_-_Chunky-Lemon-Cookie-11B-gguf/blob/main/Chunky-Lemon-Cookie-11B.Q8_0.gguf) | Q8_0 | 10.62GB |
39
+
40
+
41
+
42
+
43
+ Original model description:
44
+ ---
45
+ license: cc-by-4.0
46
+ language:
47
+ - en
48
+ base_model:
49
+ - mistralai/Mistral-7B-v0.1
50
+ - SanjiWatsuki/Kunoichi-7B
51
+ - SanjiWatsuki/Silicon-Maid-7B
52
+ - KatyTheCutie/LemonadeRP-4.5.3
53
+ - Sao10K/Fimbulvetr-11B-v2
54
+ library_name: transformers
55
+ tags:
56
+ - mergekit
57
+ - merge
58
+ - mistral
59
+ - text-generation
60
+ - roleplay
61
+ model-index:
62
+ - name: Smart-Lemon-Cookie-7B
63
+ results:
64
+ - task:
65
+ type: text-generation
66
+ name: Text Generation
67
+ dataset:
68
+ name: AI2 Reasoning Challenge (25-Shot)
69
+ type: ai2_arc
70
+ config: ARC-Challenge
71
+ split: test
72
+ args:
73
+ num_few_shot: 25
74
+ metrics:
75
+ - type: acc_norm
76
+ value: 69.62
77
+ name: normalized accuracy
78
+ source:
79
+ url: https://huggingface.co/spaces/HuggingFaceH4/open_llm_leaderboard?query=FallenMerick/Chunky-Lemon-Cookie-11B
80
+ name: Open LLM Leaderboard
81
+ - task:
82
+ type: text-generation
83
+ name: Text Generation
84
+ dataset:
85
+ name: HellaSwag (10-Shot)
86
+ type: hellaswag
87
+ split: validation
88
+ args:
89
+ num_few_shot: 10
90
+ metrics:
91
+ - type: acc_norm
92
+ value: 86.55
93
+ name: normalized accuracy
94
+ source:
95
+ url: https://huggingface.co/spaces/HuggingFaceH4/open_llm_leaderboard?query=FallenMerick/Chunky-Lemon-Cookie-11B
96
+ name: Open LLM Leaderboard
97
+ - task:
98
+ type: text-generation
99
+ name: Text Generation
100
+ dataset:
101
+ name: MMLU (5-Shot)
102
+ type: cais/mmlu
103
+ config: all
104
+ split: test
105
+ args:
106
+ num_few_shot: 5
107
+ metrics:
108
+ - type: acc
109
+ value: 65.35
110
+ name: accuracy
111
+ source:
112
+ url: https://huggingface.co/spaces/HuggingFaceH4/open_llm_leaderboard?query=FallenMerick/Chunky-Lemon-Cookie-11B
113
+ name: Open LLM Leaderboard
114
+ - task:
115
+ type: text-generation
116
+ name: Text Generation
117
+ dataset:
118
+ name: TruthfulQA (0-shot)
119
+ type: truthful_qa
120
+ config: multiple_choice
121
+ split: validation
122
+ args:
123
+ num_few_shot: 0
124
+ metrics:
125
+ - type: mc2
126
+ value: 61.59
127
+ source:
128
+ url: https://huggingface.co/spaces/HuggingFaceH4/open_llm_leaderboard?query=FallenMerick/Chunky-Lemon-Cookie-11B
129
+ name: Open LLM Leaderboard
130
+ - task:
131
+ type: text-generation
132
+ name: Text Generation
133
+ dataset:
134
+ name: Winogrande (5-shot)
135
+ type: winogrande
136
+ config: winogrande_xl
137
+ split: validation
138
+ args:
139
+ num_few_shot: 5
140
+ metrics:
141
+ - type: acc
142
+ value: 79.79
143
+ name: accuracy
144
+ source:
145
+ url: https://huggingface.co/spaces/HuggingFaceH4/open_llm_leaderboard?query=FallenMerick/Chunky-Lemon-Cookie-11B
146
+ name: Open LLM Leaderboard
147
+ - task:
148
+ type: text-generation
149
+ name: Text Generation
150
+ dataset:
151
+ name: GSM8k (5-shot)
152
+ type: gsm8k
153
+ config: main
154
+ split: test
155
+ args:
156
+ num_few_shot: 5
157
+ metrics:
158
+ - type: acc
159
+ value: 58.45
160
+ name: accuracy
161
+ source:
162
+ url: https://huggingface.co/spaces/HuggingFaceH4/open_llm_leaderboard?query=FallenMerick/Chunky-Lemon-Cookie-11B
163
+ name: Open LLM Leaderboard
164
+ ---
165
+
166
+ ![cute](https://huggingface.co/FallenMerick/Chunky-Lemon-Cookie-11B/resolve/main/Chunky-Lemon-Cookie.png)
167
+
168
+ # Chunky-Lemon-Cookie-11B
169
+
170
+ This is a merge of pre-trained language models created using [mergekit](https://github.com/cg123/mergekit).
171
+
172
+ GGUF quants:
173
+ * https://huggingface.co/backyardai/Chunky-Lemon-Cookie-11B-GGUF
174
+ * https://huggingface.co/mradermacher/Chunky-Lemon-Cookie-11B-GGUF
175
+
176
+ ## Merge Details
177
+ ### Merge Method
178
+
179
+ This model was merged using the following methods:
180
+ * passthrough
181
+ * [task arithmetic](https://arxiv.org/abs/2212.04089)
182
+
183
+ ### Models Merged
184
+
185
+ The following models were included in the merge:
186
+ * [SanjiWatsuki/Kunoichi-7B](https://huggingface.co/SanjiWatsuki/Kunoichi-7B)
187
+ * [SanjiWatsuki/Silicon-Maid-7B](https://huggingface.co/SanjiWatsuki/Silicon-Maid-7B)
188
+ * [KatyTheCutie/LemonadeRP-4.5.3](https://huggingface.co/KatyTheCutie/LemonadeRP-4.5.3)
189
+ * [Sao10K/Fimbulvetr-11B-v2](https://huggingface.co/Sao10K/Fimbulvetr-11B-v2)
190
+ * [mistralai/Mistral-7B-v0.1](https://huggingface.co/mistralai/Mistral-7B-v0.1)
191
+
192
+ ### Configuration
193
+
194
+ The following YAML configurations were used to produce this model:
195
+
196
+ ```yaml
197
+ slices:
198
+ - sources:
199
+ - model: mistralai/Mistral-7B-v0.1
200
+ layer_range: [0, 24]
201
+ - sources:
202
+ - model: mistralai/Mistral-7B-v0.1
203
+ layer_range: [8, 32]
204
+ merge_method: passthrough
205
+ dtype: float16
206
+ name: Mistral-11B
207
+
208
+ ---
209
+
210
+ slices:
211
+ - sources:
212
+ - model: SanjiWatsuki/Kunoichi-7B
213
+ layer_range: [0, 24]
214
+ - sources:
215
+ - model: SanjiWatsuki/Silicon-Maid-7B
216
+ layer_range: [8, 24]
217
+ - sources:
218
+ - model: KatyTheCutie/LemonadeRP-4.5.3
219
+ layer_range: [24, 32]
220
+ merge_method: passthrough
221
+ dtype: float16
222
+ name: Big-Lemon-Cookie-11B
223
+
224
+ ---
225
+
226
+ models:
227
+ - model: Big-Lemon-Cookie-11B
228
+ parameters:
229
+ weight: 0.85
230
+ - model: Sao10K/Fimbulvetr-11B-v2
231
+ parameters:
232
+ weight: 0.15
233
+ merge_method: task_arithmetic
234
+ base_model: Mistral-11B
235
+ dtype: float16
236
+ name: Chunky-Lemon-Cookie-11B
237
+ ```
238
+
239
+ # [Open LLM Leaderboard Evaluation Results](https://huggingface.co/spaces/HuggingFaceH4/open_llm_leaderboard)
240
+ Detailed results can be found [here](https://huggingface.co/datasets/open-llm-leaderboard/details_FallenMerick__Chunky-Lemon-Cookie-11B)
241
+
242
+ | Metric |Value|
243
+ |---------------------------------|----:|
244
+ |Avg. |70.23|
245
+ |AI2 Reasoning Challenge (25-Shot)|69.62|
246
+ |HellaSwag (10-Shot) |86.55|
247
+ |MMLU (5-Shot) |65.35|
248
+ |TruthfulQA (0-shot) |61.59|
249
+ |Winogrande (5-shot) |79.79|
250
+ |GSM8k (5-shot) |58.45|
251
+