afrideva commited on
Commit
79b5c68
1 Parent(s): 20a6f3a

Upload README.md with huggingface_hub

Browse files
Files changed (1) hide show
  1. README.md +153 -0
README.md ADDED
@@ -0,0 +1,153 @@
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
1
+ ---
2
+ base_model: mwitiderrick/open_llama_3b_code_instruct_0.1
3
+ created_by: mwitiderrick
4
+ datasets:
5
+ - mwitiderrick/AlpacaCode
6
+ inference: false
7
+ language:
8
+ - en
9
+ library_name: transformers
10
+ license: apache-2.0
11
+ model-index:
12
+ - name: mwitiderrick/open_llama_3b_instruct_v_0.2
13
+ results:
14
+ - dataset:
15
+ name: hellaswag
16
+ type: hellaswag
17
+ metrics:
18
+ - name: hellaswag(0-Shot)
19
+ type: hellaswag (0-Shot)
20
+ value: 0.6581
21
+ task:
22
+ type: text-generation
23
+ - dataset:
24
+ name: winogrande
25
+ type: winogrande
26
+ metrics:
27
+ - name: winogrande(0-Shot)
28
+ type: winogrande (0-Shot)
29
+ value: 0.6267
30
+ task:
31
+ type: text-generation
32
+ - dataset:
33
+ name: arc_challenge
34
+ type: arc_challenge
35
+ metrics:
36
+ - name: arc_challenge(0-Shot)
37
+ type: arc_challenge (0-Shot)
38
+ value: 0.3712
39
+ source:
40
+ name: open_llama_3b_instruct_v_0.2 model card
41
+ url: https://huggingface.co/mwitiderrick/open_llama_3b_instruct_v_0.2
42
+ task:
43
+ type: text-generation
44
+ model_creator: mwitiderrick
45
+ model_name: open_llama_3b_code_instruct_0.1
46
+ model_type: llama
47
+ pipeline_tag: text-generation
48
+ prompt_template: '### Instruction:\n
49
+
50
+ {prompt}
51
+
52
+ ### Response:
53
+
54
+ '
55
+ quantized_by: afrideva
56
+ tags:
57
+ - transformers
58
+ - gguf
59
+ - ggml
60
+ - quantized
61
+ - q2_k
62
+ - q3_k_m
63
+ - q4_k_m
64
+ - q5_k_m
65
+ - q6_k
66
+ - q8_0
67
+ ---
68
+ # mwitiderrick/open_llama_3b_code_instruct_0.1-GGUF
69
+
70
+ Quantized GGUF model files for [open_llama_3b_code_instruct_0.1](https://huggingface.co/mwitiderrick/open_llama_3b_code_instruct_0.1) from [mwitiderrick](https://huggingface.co/mwitiderrick)
71
+
72
+
73
+ | Name | Quant method | Size |
74
+ | ---- | ---- | ---- |
75
+ | [open_llama_3b_code_instruct_0.1.fp16.gguf](https://huggingface.co/afrideva/open_llama_3b_code_instruct_0.1-GGUF/resolve/main/open_llama_3b_code_instruct_0.1.fp16.gguf) | fp16 | 6.86 GB |
76
+ | [open_llama_3b_code_instruct_0.1.q2_k.gguf](https://huggingface.co/afrideva/open_llama_3b_code_instruct_0.1-GGUF/resolve/main/open_llama_3b_code_instruct_0.1.q2_k.gguf) | q2_k | 2.15 GB |
77
+ | [open_llama_3b_code_instruct_0.1.q3_k_m.gguf](https://huggingface.co/afrideva/open_llama_3b_code_instruct_0.1-GGUF/resolve/main/open_llama_3b_code_instruct_0.1.q3_k_m.gguf) | q3_k_m | 2.27 GB |
78
+ | [open_llama_3b_code_instruct_0.1.q4_k_m.gguf](https://huggingface.co/afrideva/open_llama_3b_code_instruct_0.1-GGUF/resolve/main/open_llama_3b_code_instruct_0.1.q4_k_m.gguf) | q4_k_m | 2.58 GB |
79
+ | [open_llama_3b_code_instruct_0.1.q5_k_m.gguf](https://huggingface.co/afrideva/open_llama_3b_code_instruct_0.1-GGUF/resolve/main/open_llama_3b_code_instruct_0.1.q5_k_m.gguf) | q5_k_m | 2.76 GB |
80
+ | [open_llama_3b_code_instruct_0.1.q6_k.gguf](https://huggingface.co/afrideva/open_llama_3b_code_instruct_0.1-GGUF/resolve/main/open_llama_3b_code_instruct_0.1.q6_k.gguf) | q6_k | 3.64 GB |
81
+ | [open_llama_3b_code_instruct_0.1.q8_0.gguf](https://huggingface.co/afrideva/open_llama_3b_code_instruct_0.1-GGUF/resolve/main/open_llama_3b_code_instruct_0.1.q8_0.gguf) | q8_0 | 3.64 GB |
82
+
83
+
84
+
85
+ ## Original Model Card:
86
+ # OpenLLaMA Code Instruct: An Open Reproduction of LLaMA
87
+
88
+ This is an [OpenLlama model](https://huggingface.co/openlm-research/open_llama_3b) that has been fine-tuned on 1 epoch of the
89
+ [AlpacaCode](https://huggingface.co/datasets/mwitiderrick/AlpacaCode) dataset (122K rows).
90
+
91
+ ## Prompt Template
92
+ ```
93
+ ### Instruction:
94
+
95
+ {query}
96
+
97
+ ### Response:
98
+ <Leave new line for model to respond>
99
+ ```
100
+ ## Usage
101
+ ```python
102
+ from transformers import AutoTokenizer, AutoModelForCausalLM,pipeline
103
+
104
+ tokenizer = AutoTokenizer.from_pretrained("mwitiderrick/open_llama_3b_code_instruct_0.1")
105
+ model = AutoModelForCausalLM.from_pretrained("mwitiderrick/open_llama_3b_code_instruct_0.1")
106
+ query = "Write a quick sort algorithm in Python"
107
+ text_gen = pipeline(task="text-generation", model=model, tokenizer=tokenizer, max_length=200)
108
+ output = text_gen(f"### Instruction:\n{query}\n### Response:\n")
109
+ print(output[0]['generated_text'])
110
+ """
111
+ ### Instruction:
112
+ write a quick sort algorithm in Python
113
+ ### Response:
114
+ def quick_sort(arr):
115
+ if len(arr) <= 1:
116
+ return arr
117
+ else:
118
+ pivot = arr[len(arr) // 2]
119
+ left = [x for x in arr if x < pivot]
120
+ middle = [x for x in arr if x == pivot]
121
+ right = [x for x in arr if x > pivot]
122
+ return quick_sort(left) + middle + quick_sort(right)
123
+
124
+ arr = [5,2,4,3,1]
125
+ print(quick_sort(arr))
126
+ """
127
+ [1, 2, 3, 4, 5]
128
+ """
129
+ ```
130
+ ## Metrics
131
+ [Detailed metrics](https://huggingface.co/datasets/open-llm-leaderboard/details_mwitiderrick__open_llama_3b_code_instruct_0.1)
132
+ ```
133
+ | Tasks |Version|Filter|n-shot|Metric|Value | |Stderr|
134
+ |----------|-------|------|-----:|------|-----:|---|-----:|
135
+ |winogrande|Yaml |none | 0|acc |0.6267|± |0.0136|
136
+ |hellaswag|Yaml |none | 0|acc |0.4962|± |0.0050|
137
+ | | |none | 0|acc_norm|0.6581|± |0.0047|
138
+ |arc_challenge|Yaml |none | 0|acc |0.3481|± |0.0139|
139
+ | | |none | 0|acc_norm|0.3712|± |0.0141|
140
+ |truthfulqa|N/A |none | 0|bleu_max | 24.2580|± |0.5985|
141
+ | | |none | 0|bleu_acc | 0.2876|± |0.0003|
142
+ | | |none | 0|bleu_diff | -8.3685|± |0.6065|
143
+ | | |none | 0|rouge1_max | 49.3907|± |0.7350|
144
+ | | |none | 0|rouge1_acc | 0.2558|± |0.0002|
145
+ | | |none | 0|rouge1_diff|-10.6617|± |0.6450|
146
+ | | |none | 0|rouge2_max | 32.4189|± |0.9587|
147
+ | | |none | 0|rouge2_acc | 0.2142|± |0.0002|
148
+ | | |none | 0|rouge2_diff|-12.9903|± |0.9539|
149
+ | | |none | 0|rougeL_max | 46.2337|± |0.7493|
150
+ | | |none | 0|rougeL_acc | 0.2424|± |0.0002|
151
+ | | |none | 0|rougeL_diff|-11.0285|± |0.6576|
152
+ | | |none | 0|acc | 0.3072|± |0.0405|
153
+ ```