morriszms commited on
Commit
22f2dac
·
verified ·
1 Parent(s): 5acb52e

Upload folder using huggingface_hub

Browse files
.gitattributes CHANGED
@@ -33,3 +33,15 @@ saved_model/**/* 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
 
 
 
 
 
 
 
 
 
 
 
 
 
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
36
+ Smol-Llama-101M-Chat-v1-Q2_K.gguf filter=lfs diff=lfs merge=lfs -text
37
+ Smol-Llama-101M-Chat-v1-Q3_K_L.gguf filter=lfs diff=lfs merge=lfs -text
38
+ Smol-Llama-101M-Chat-v1-Q3_K_M.gguf filter=lfs diff=lfs merge=lfs -text
39
+ Smol-Llama-101M-Chat-v1-Q3_K_S.gguf filter=lfs diff=lfs merge=lfs -text
40
+ Smol-Llama-101M-Chat-v1-Q4_0.gguf filter=lfs diff=lfs merge=lfs -text
41
+ Smol-Llama-101M-Chat-v1-Q4_K_M.gguf filter=lfs diff=lfs merge=lfs -text
42
+ Smol-Llama-101M-Chat-v1-Q4_K_S.gguf filter=lfs diff=lfs merge=lfs -text
43
+ Smol-Llama-101M-Chat-v1-Q5_0.gguf filter=lfs diff=lfs merge=lfs -text
44
+ Smol-Llama-101M-Chat-v1-Q5_K_M.gguf filter=lfs diff=lfs merge=lfs -text
45
+ Smol-Llama-101M-Chat-v1-Q5_K_S.gguf filter=lfs diff=lfs merge=lfs -text
46
+ Smol-Llama-101M-Chat-v1-Q6_K.gguf filter=lfs diff=lfs merge=lfs -text
47
+ Smol-Llama-101M-Chat-v1-Q8_0.gguf filter=lfs diff=lfs merge=lfs -text
README.md ADDED
@@ -0,0 +1,272 @@
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
1
+ ---
2
+ language:
3
+ - en
4
+ license: apache-2.0
5
+ tags:
6
+ - text-generation
7
+ - TensorBlock
8
+ - GGUF
9
+ base_model: Felladrin/Smol-Llama-101M-Chat-v1
10
+ datasets:
11
+ - Open-Orca/SlimOrca-Dedup
12
+ - VMware/open-instruct
13
+ - LDJnr/Capybara
14
+ - cognitivecomputations/ultrachat-uncensored
15
+ - starfishmedical/webGPT_x_dolly
16
+ - THUDM/webglm-qa
17
+ widget:
18
+ - messages:
19
+ - role: system
20
+ content: You are a helpful assistant who gives creative responses.
21
+ - role: user
22
+ content: Write the background story of a game about wizards and llamas in a sci-fi
23
+ world.
24
+ - messages:
25
+ - role: system
26
+ content: A friendly chat between a user and an assistant.
27
+ - role: user
28
+ content: Got a question for you!
29
+ - role: assistant
30
+ content: Sure! What's it?
31
+ - role: user
32
+ content: I need to build a simple website. Where should I start learning about
33
+ web development?
34
+ - messages:
35
+ - role: system
36
+ content: You are a helpful assistant who provides concise answers to the user's
37
+ questions.
38
+ - role: user
39
+ content: How to become more healthy?
40
+ - messages:
41
+ - role: system
42
+ content: You are a helpful assistant, who always answers with empathy.
43
+ - role: user
44
+ content: List the pros and cons of social media.
45
+ - messages:
46
+ - role: system
47
+ content: You are a helpful assistant, who always answers with empathy.
48
+ - role: user
49
+ content: Hello!
50
+ - role: assistant
51
+ content: Hi! How can I help you today?
52
+ - role: user
53
+ content: 'Take a look at the info below.
54
+
55
+ - The tape inside the VHS cassettes is very delicate and can be easily ruined,
56
+ making them unplayable and unrepairable. The reason the tape deteriorates is
57
+ that the magnetic charge needed for them to work is not permanent, and the magnetic
58
+ particles end up losing their charge in a process known as remanence decay.
59
+ These particles could also become demagnetised via being stored too close to
60
+ a magnetic source.
61
+
62
+ - One of the most significant issues with VHS tapes is that they have moving
63
+ parts, meaning that there are more occasions when something can go wrong, damaging
64
+ your footage or preventing it from playing back. The tape itself is a prominent
65
+ cause of this, and tape slippage can occur. Tapes slippage can be caused when
66
+ the tape loses its tension, or it has become warped. These problems can occur
67
+ in storage due to high temperatures or frequent changes in humidity.
68
+
69
+ - VHS tapes deteriorate over time from infrequent or overuse. Neglect means
70
+ mold and dirt, while overuse can lead to scratches and technical difficulties.
71
+ This is why old VHS tapes inevitably experience malfunctions after a long period
72
+ of time. Usually anywhere between 10 to 25+ years.
73
+
74
+ - Some VHS tapes like newer mini DVs and Digital 8 tapes can suffer from digital
75
+ corruption, meaning that the footage becomes lost and cannot be recovered. These
76
+ tapes were the steppingstone from VHS to the digital age when capturing footage
77
+ straight to digital became the norm. Unfortunately,they are susceptible to digital
78
+ corruption, which causes video pixilation and/or loss of audio.'
79
+ - role: assistant
80
+ content: Alright!
81
+ - role: user
82
+ content: 'Now I''m going to write my question, and if the info above is useful,
83
+ you can use them in your response.
84
+
85
+ Ready?'
86
+ - role: assistant
87
+ content: Ready for your question!
88
+ - role: user
89
+ content: Why do VHS tapes deteriorate over time?
90
+ inference:
91
+ parameters:
92
+ max_new_tokens: 250
93
+ penalty_alpha: 0.5
94
+ top_k: 4
95
+ repetition_penalty: 1.105
96
+ model-index:
97
+ - name: Smol-Llama-101M-Chat-v1
98
+ results:
99
+ - task:
100
+ type: text-generation
101
+ name: Text Generation
102
+ dataset:
103
+ name: AI2 Reasoning Challenge (25-Shot)
104
+ type: ai2_arc
105
+ config: ARC-Challenge
106
+ split: test
107
+ args:
108
+ num_few_shot: 25
109
+ metrics:
110
+ - type: acc_norm
111
+ value: 22.87
112
+ name: normalized accuracy
113
+ source:
114
+ url: https://huggingface.co/spaces/HuggingFaceH4/open_llm_leaderboard?query=Felladrin/Smol-Llama-101M-Chat-v1
115
+ name: Open LLM Leaderboard
116
+ - task:
117
+ type: text-generation
118
+ name: Text Generation
119
+ dataset:
120
+ name: HellaSwag (10-Shot)
121
+ type: hellaswag
122
+ split: validation
123
+ args:
124
+ num_few_shot: 10
125
+ metrics:
126
+ - type: acc_norm
127
+ value: 28.69
128
+ name: normalized accuracy
129
+ source:
130
+ url: https://huggingface.co/spaces/HuggingFaceH4/open_llm_leaderboard?query=Felladrin/Smol-Llama-101M-Chat-v1
131
+ name: Open LLM Leaderboard
132
+ - task:
133
+ type: text-generation
134
+ name: Text Generation
135
+ dataset:
136
+ name: MMLU (5-Shot)
137
+ type: cais/mmlu
138
+ config: all
139
+ split: test
140
+ args:
141
+ num_few_shot: 5
142
+ metrics:
143
+ - type: acc
144
+ value: 24.93
145
+ name: accuracy
146
+ source:
147
+ url: https://huggingface.co/spaces/HuggingFaceH4/open_llm_leaderboard?query=Felladrin/Smol-Llama-101M-Chat-v1
148
+ name: Open LLM Leaderboard
149
+ - task:
150
+ type: text-generation
151
+ name: Text Generation
152
+ dataset:
153
+ name: TruthfulQA (0-shot)
154
+ type: truthful_qa
155
+ config: multiple_choice
156
+ split: validation
157
+ args:
158
+ num_few_shot: 0
159
+ metrics:
160
+ - type: mc2
161
+ value: 45.76
162
+ source:
163
+ url: https://huggingface.co/spaces/HuggingFaceH4/open_llm_leaderboard?query=Felladrin/Smol-Llama-101M-Chat-v1
164
+ name: Open LLM Leaderboard
165
+ - task:
166
+ type: text-generation
167
+ name: Text Generation
168
+ dataset:
169
+ name: Winogrande (5-shot)
170
+ type: winogrande
171
+ config: winogrande_xl
172
+ split: validation
173
+ args:
174
+ num_few_shot: 5
175
+ metrics:
176
+ - type: acc
177
+ value: 50.04
178
+ name: accuracy
179
+ source:
180
+ url: https://huggingface.co/spaces/HuggingFaceH4/open_llm_leaderboard?query=Felladrin/Smol-Llama-101M-Chat-v1
181
+ name: Open LLM Leaderboard
182
+ - task:
183
+ type: text-generation
184
+ name: Text Generation
185
+ dataset:
186
+ name: GSM8k (5-shot)
187
+ type: gsm8k
188
+ config: main
189
+ split: test
190
+ args:
191
+ num_few_shot: 5
192
+ metrics:
193
+ - type: acc
194
+ value: 0.08
195
+ name: accuracy
196
+ source:
197
+ url: https://huggingface.co/spaces/HuggingFaceH4/open_llm_leaderboard?query=Felladrin/Smol-Llama-101M-Chat-v1
198
+ name: Open LLM Leaderboard
199
+ ---
200
+
201
+ <div style="width: auto; margin-left: auto; margin-right: auto">
202
+ <img src="https://i.imgur.com/jC7kdl8.jpeg" alt="TensorBlock" style="width: 100%; min-width: 400px; display: block; margin: auto;">
203
+ </div>
204
+ <div style="display: flex; justify-content: space-between; width: 100%;">
205
+ <div style="display: flex; flex-direction: column; align-items: flex-start;">
206
+ <p style="margin-top: 0.5em; margin-bottom: 0em;">
207
+ Feedback and support: TensorBlock's <a href="https://x.com/tensorblock_aoi">Twitter/X</a>, <a href="https://t.me/TensorBlock">Telegram Group</a> and <a href="https://x.com/tensorblock_aoi">Discord server</a>
208
+ </p>
209
+ </div>
210
+ </div>
211
+
212
+ ## Felladrin/Smol-Llama-101M-Chat-v1 - GGUF
213
+
214
+ This repo contains GGUF format model files for [Felladrin/Smol-Llama-101M-Chat-v1](https://huggingface.co/Felladrin/Smol-Llama-101M-Chat-v1).
215
+
216
+ The files were quantized using machines provided by [TensorBlock](https://tensorblock.co/), and they are compatible with llama.cpp as of [commit b4242](https://github.com/ggerganov/llama.cpp/commit/a6744e43e80f4be6398fc7733a01642c846dce1d).
217
+
218
+ <div style="text-align: left; margin: 20px 0;">
219
+ <a href="https://tensorblock.co/waitlist/client" style="display: inline-block; padding: 10px 20px; background-color: #007bff; color: white; text-decoration: none; border-radius: 5px; font-weight: bold;">
220
+ Run them on the TensorBlock client using your local machine ↗
221
+ </a>
222
+ </div>
223
+
224
+ ## Prompt template
225
+
226
+ ```
227
+ <|im_start|>system
228
+ {system_prompt}<|im_end|>
229
+ <|im_start|>user
230
+ {prompt}<|im_end|>
231
+ <|im_start|>assistant
232
+ ```
233
+
234
+ ## Model file specification
235
+
236
+ | Filename | Quant type | File Size | Description |
237
+ | -------- | ---------- | --------- | ----------- |
238
+ | [Smol-Llama-101M-Chat-v1-Q2_K.gguf](https://huggingface.co/tensorblock/Smol-Llama-101M-Chat-v1-GGUF/blob/main/Smol-Llama-101M-Chat-v1-Q2_K.gguf) | Q2_K | 0.048 GB | smallest, significant quality loss - not recommended for most purposes |
239
+ | [Smol-Llama-101M-Chat-v1-Q3_K_S.gguf](https://huggingface.co/tensorblock/Smol-Llama-101M-Chat-v1-GGUF/blob/main/Smol-Llama-101M-Chat-v1-Q3_K_S.gguf) | Q3_K_S | 0.054 GB | very small, high quality loss |
240
+ | [Smol-Llama-101M-Chat-v1-Q3_K_M.gguf](https://huggingface.co/tensorblock/Smol-Llama-101M-Chat-v1-GGUF/blob/main/Smol-Llama-101M-Chat-v1-Q3_K_M.gguf) | Q3_K_M | 0.056 GB | very small, high quality loss |
241
+ | [Smol-Llama-101M-Chat-v1-Q3_K_L.gguf](https://huggingface.co/tensorblock/Smol-Llama-101M-Chat-v1-GGUF/blob/main/Smol-Llama-101M-Chat-v1-Q3_K_L.gguf) | Q3_K_L | 0.059 GB | small, substantial quality loss |
242
+ | [Smol-Llama-101M-Chat-v1-Q4_0.gguf](https://huggingface.co/tensorblock/Smol-Llama-101M-Chat-v1-GGUF/blob/main/Smol-Llama-101M-Chat-v1-Q4_0.gguf) | Q4_0 | 0.064 GB | legacy; small, very high quality loss - prefer using Q3_K_M |
243
+ | [Smol-Llama-101M-Chat-v1-Q4_K_S.gguf](https://huggingface.co/tensorblock/Smol-Llama-101M-Chat-v1-GGUF/blob/main/Smol-Llama-101M-Chat-v1-Q4_K_S.gguf) | Q4_K_S | 0.064 GB | small, greater quality loss |
244
+ | [Smol-Llama-101M-Chat-v1-Q4_K_M.gguf](https://huggingface.co/tensorblock/Smol-Llama-101M-Chat-v1-GGUF/blob/main/Smol-Llama-101M-Chat-v1-Q4_K_M.gguf) | Q4_K_M | 0.065 GB | medium, balanced quality - recommended |
245
+ | [Smol-Llama-101M-Chat-v1-Q5_0.gguf](https://huggingface.co/tensorblock/Smol-Llama-101M-Chat-v1-GGUF/blob/main/Smol-Llama-101M-Chat-v1-Q5_0.gguf) | Q5_0 | 0.074 GB | legacy; medium, balanced quality - prefer using Q4_K_M |
246
+ | [Smol-Llama-101M-Chat-v1-Q5_K_S.gguf](https://huggingface.co/tensorblock/Smol-Llama-101M-Chat-v1-GGUF/blob/main/Smol-Llama-101M-Chat-v1-Q5_K_S.gguf) | Q5_K_S | 0.074 GB | large, low quality loss - recommended |
247
+ | [Smol-Llama-101M-Chat-v1-Q5_K_M.gguf](https://huggingface.co/tensorblock/Smol-Llama-101M-Chat-v1-GGUF/blob/main/Smol-Llama-101M-Chat-v1-Q5_K_M.gguf) | Q5_K_M | 0.074 GB | large, very low quality loss - recommended |
248
+ | [Smol-Llama-101M-Chat-v1-Q6_K.gguf](https://huggingface.co/tensorblock/Smol-Llama-101M-Chat-v1-GGUF/blob/main/Smol-Llama-101M-Chat-v1-Q6_K.gguf) | Q6_K | 0.084 GB | very large, extremely low quality loss |
249
+ | [Smol-Llama-101M-Chat-v1-Q8_0.gguf](https://huggingface.co/tensorblock/Smol-Llama-101M-Chat-v1-GGUF/blob/main/Smol-Llama-101M-Chat-v1-Q8_0.gguf) | Q8_0 | 0.108 GB | very large, extremely low quality loss - not recommended |
250
+
251
+
252
+ ## Downloading instruction
253
+
254
+ ### Command line
255
+
256
+ Firstly, install Huggingface Client
257
+
258
+ ```shell
259
+ pip install -U "huggingface_hub[cli]"
260
+ ```
261
+
262
+ Then, downoad the individual model file the a local directory
263
+
264
+ ```shell
265
+ huggingface-cli download tensorblock/Smol-Llama-101M-Chat-v1-GGUF --include "Smol-Llama-101M-Chat-v1-Q2_K.gguf" --local-dir MY_LOCAL_DIR
266
+ ```
267
+
268
+ If you wanna download multiple model files with a pattern (e.g., `*Q4_K*gguf`), you can try:
269
+
270
+ ```shell
271
+ huggingface-cli download tensorblock/Smol-Llama-101M-Chat-v1-GGUF --local-dir MY_LOCAL_DIR --local-dir-use-symlinks False --include='*Q4_K*gguf'
272
+ ```
Smol-Llama-101M-Chat-v1-Q2_K.gguf ADDED
@@ -0,0 +1,3 @@
 
 
 
 
1
+ version https://git-lfs.github.com/spec/v1
2
+ oid sha256:0d5bf0716a03bf9cd419337cf672d73f82109eb3e30e1d12dd39da8555114c23
3
+ size 48057728
Smol-Llama-101M-Chat-v1-Q3_K_L.gguf ADDED
@@ -0,0 +1,3 @@
 
 
 
 
1
+ version https://git-lfs.github.com/spec/v1
2
+ oid sha256:c59b575db893fc9933fc5b0a98fa9493a2d37ba512d340bb48e885fa8f9fa234
3
+ size 58784384
Smol-Llama-101M-Chat-v1-Q3_K_M.gguf ADDED
@@ -0,0 +1,3 @@
 
 
 
 
1
+ version https://git-lfs.github.com/spec/v1
2
+ oid sha256:c1383b9f13e1e6942ee21206078cfd14424025988eaaf249b877c78e92b188c4
3
+ size 56474240
Smol-Llama-101M-Chat-v1-Q3_K_S.gguf ADDED
@@ -0,0 +1,3 @@
 
 
 
 
1
+ version https://git-lfs.github.com/spec/v1
2
+ oid sha256:0cb74f11b8b51f55d04dc89759ee914223d851bb12dccb5747d03af4208293c3
3
+ size 53918336
Smol-Llama-101M-Chat-v1-Q4_0.gguf ADDED
@@ -0,0 +1,3 @@
 
 
 
 
1
+ version https://git-lfs.github.com/spec/v1
2
+ oid sha256:e6ced517a4a8acdd794395a62c8341afb5086f84416397cd0931625a3b959be9
3
+ size 64088960
Smol-Llama-101M-Chat-v1-Q4_K_M.gguf ADDED
@@ -0,0 +1,3 @@
 
 
 
 
1
+ version https://git-lfs.github.com/spec/v1
2
+ oid sha256:9d43c1df17539c18c93784f14c6ca95c097515bb8bca38bc32dbd8daa062517d
3
+ size 65406848
Smol-Llama-101M-Chat-v1-Q4_K_S.gguf ADDED
@@ -0,0 +1,3 @@
 
 
 
 
1
+ version https://git-lfs.github.com/spec/v1
2
+ oid sha256:26b1a1780c94eeed2468b9b32acda7853e5e25e2346f403afd51372a62979c2d
3
+ size 64187264
Smol-Llama-101M-Chat-v1-Q5_0.gguf ADDED
@@ -0,0 +1,3 @@
 
 
 
 
1
+ version https://git-lfs.github.com/spec/v1
2
+ oid sha256:f31b18e449827b69034d6103b748fc82d8701809e0de058a9497f2191fe7d38e
3
+ size 73661312
Smol-Llama-101M-Chat-v1-Q5_K_M.gguf ADDED
@@ -0,0 +1,3 @@
 
 
 
 
1
+ version https://git-lfs.github.com/spec/v1
2
+ oid sha256:d29ad16e878d7e6a63cc15aa5be6b9c6fb26269d9b6e67e9632debdc61883e3d
3
+ size 74340224
Smol-Llama-101M-Chat-v1-Q5_K_S.gguf ADDED
@@ -0,0 +1,3 @@
 
 
 
 
1
+ version https://git-lfs.github.com/spec/v1
2
+ oid sha256:ae87d89ce95753380b98809cde7e942819c7eb20c44cd185f19f83e511c26722
3
+ size 73661312
Smol-Llama-101M-Chat-v1-Q6_K.gguf ADDED
@@ -0,0 +1,3 @@
 
 
 
 
1
+ version https://git-lfs.github.com/spec/v1
2
+ oid sha256:673efc73615bebecbc5dd7a775139fbcbe4026f18cc1dc74bc68b2d7a290f110
3
+ size 83831936
Smol-Llama-101M-Chat-v1-Q8_0.gguf ADDED
@@ -0,0 +1,3 @@
 
 
 
 
1
+ version https://git-lfs.github.com/spec/v1
2
+ oid sha256:681cbd470f2f3114320234dcc0930b79b49dc17f7049f1188f04149bf091c167
3
+ size 108354176