File size: 4,148 Bytes
bbc45e8
 
a209831
 
 
25bd7ee
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
bbc45e8
f0076c4
a209831
f0076c4
4254e49
a209831
 
 
 
4fffffb
a209831
4fffffb
a209831
 
 
4fffffb
a209831
 
 
 
25bd7ee
 
 
 
 
 
 
 
 
 
 
 
 
 
1
2
3
4
5
6
7
8
9
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
25
26
27
28
29
30
31
32
33
34
35
36
37
38
39
40
41
42
43
44
45
46
47
48
49
50
51
52
53
54
55
56
57
58
59
60
61
62
63
64
65
66
67
68
69
70
71
72
73
74
75
76
77
78
79
80
81
82
83
84
85
86
87
88
89
90
91
92
93
94
95
96
97
98
99
100
101
102
103
104
105
106
107
108
109
110
111
112
113
114
115
116
117
118
119
120
121
122
123
124
125
126
127
128
129
130
131
132
133
134
135
136
137
138
139
140
141
142
143
---
license: apache-2.0
tags:
- llama-2
- roleplaying
model-index:
- name: TimeCrystal-l2-13B
  results:
  - task:
      type: text-generation
      name: Text Generation
    dataset:
      name: AI2 Reasoning Challenge (25-Shot)
      type: ai2_arc
      config: ARC-Challenge
      split: test
      args:
        num_few_shot: 25
    metrics:
    - type: acc_norm
      value: 61.18
      name: normalized accuracy
    source:
      url: https://huggingface.co/spaces/HuggingFaceH4/open_llm_leaderboard?query=BlueNipples/TimeCrystal-l2-13B
      name: Open LLM Leaderboard
  - task:
      type: text-generation
      name: Text Generation
    dataset:
      name: HellaSwag (10-Shot)
      type: hellaswag
      split: validation
      args:
        num_few_shot: 10
    metrics:
    - type: acc_norm
      value: 83.71
      name: normalized accuracy
    source:
      url: https://huggingface.co/spaces/HuggingFaceH4/open_llm_leaderboard?query=BlueNipples/TimeCrystal-l2-13B
      name: Open LLM Leaderboard
  - task:
      type: text-generation
      name: Text Generation
    dataset:
      name: MMLU (5-Shot)
      type: cais/mmlu
      config: all
      split: test
      args:
        num_few_shot: 5
    metrics:
    - type: acc
      value: 56.46
      name: accuracy
    source:
      url: https://huggingface.co/spaces/HuggingFaceH4/open_llm_leaderboard?query=BlueNipples/TimeCrystal-l2-13B
      name: Open LLM Leaderboard
  - task:
      type: text-generation
      name: Text Generation
    dataset:
      name: TruthfulQA (0-shot)
      type: truthful_qa
      config: multiple_choice
      split: validation
      args:
        num_few_shot: 0
    metrics:
    - type: mc2
      value: 51.3
    source:
      url: https://huggingface.co/spaces/HuggingFaceH4/open_llm_leaderboard?query=BlueNipples/TimeCrystal-l2-13B
      name: Open LLM Leaderboard
  - task:
      type: text-generation
      name: Text Generation
    dataset:
      name: Winogrande (5-shot)
      type: winogrande
      config: winogrande_xl
      split: validation
      args:
        num_few_shot: 5
    metrics:
    - type: acc
      value: 75.37
      name: accuracy
    source:
      url: https://huggingface.co/spaces/HuggingFaceH4/open_llm_leaderboard?query=BlueNipples/TimeCrystal-l2-13B
      name: Open LLM Leaderboard
  - task:
      type: text-generation
      name: Text Generation
    dataset:
      name: GSM8k (5-shot)
      type: gsm8k
      config: main
      split: test
      args:
        num_few_shot: 5
    metrics:
    - type: acc
      value: 27.52
      name: accuracy
    source:
      url: https://huggingface.co/spaces/HuggingFaceH4/open_llm_leaderboard?query=BlueNipples/TimeCrystal-l2-13B
      name: Open LLM Leaderboard
---
This 13B model, TimeCrystal-l2-13B is built to maximize logic and instruct following, whilst also increasing the vividness of prose found in Chronos based models like Mythomax, over the more romantic prose, hopefully without losing the elegent narrative structure touch of newer models like synthia and xwin. TLDR: Attempt at more clever, better prose. 

Tentative test results: I'm not certain if logic/instruct was improved or not (haven't tested much), but the prose infusion seems to have worked really well.

It is built so:

SLERPS:
Amethyst + Openchat Super = OpenStone

MythoMax + Chronos = ChronoMax

ChronoMax + Amethyst = TimeStone

Gradient Merge:

TimeStone + OpenStone (0.9,0,0) = TimeCrystal 

Props to all the mergers, fine tuners!

All models in Merge: Many, lol.
# [Open LLM Leaderboard Evaluation Results](https://huggingface.co/spaces/HuggingFaceH4/open_llm_leaderboard)
Detailed results can be found [here](https://huggingface.co/datasets/open-llm-leaderboard/details_BlueNipples__TimeCrystal-l2-13B)

|             Metric              |Value|
|---------------------------------|----:|
|Avg.                             |59.26|
|AI2 Reasoning Challenge (25-Shot)|61.18|
|HellaSwag (10-Shot)              |83.71|
|MMLU (5-Shot)                    |56.46|
|TruthfulQA (0-shot)              |51.30|
|Winogrande (5-shot)              |75.37|
|GSM8k (5-shot)                   |27.52|