Hugging Face's logo

RichardErkhov
/
cognitivecomputations_-_TinyDolphin-2.8.1-1.1b-gguf

GGUF
Inference Endpoints
Model card Files Community
RichardErkhov commited on
Commit
8ef6886
•
1 Parent(s): 7492c7e

uploaded readme

Browse files
Files changed (1) hide show
  1. README.md +97 -0
README.md ADDED
@@ -0,0 +1,97 @@
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
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
+ TinyDolphin-2.8.1-1.1b - GGUF
11
+ - Model creator: https://huggingface.co/cognitivecomputations/
12
+ - Original model: https://huggingface.co/cognitivecomputations/TinyDolphin-2.8.1-1.1b/
13
+
14
+
15
+ | Name | Quant method | Size |
16
+ | ---- | ---- | ---- |
17
+ | [TinyDolphin-2.8.1-1.1b.Q2_K.gguf](https://huggingface.co/RichardErkhov/cognitivecomputations_-_TinyDolphin-2.8.1-1.1b-gguf/blob/main/TinyDolphin-2.8.1-1.1b.Q2_K.gguf) | Q2_K | 0.4GB |
18
+ | [TinyDolphin-2.8.1-1.1b.IQ3_XS.gguf](https://huggingface.co/RichardErkhov/cognitivecomputations_-_TinyDolphin-2.8.1-1.1b-gguf/blob/main/TinyDolphin-2.8.1-1.1b.IQ3_XS.gguf) | IQ3_XS | 0.44GB |
19
+ | [TinyDolphin-2.8.1-1.1b.IQ3_S.gguf](https://huggingface.co/RichardErkhov/cognitivecomputations_-_TinyDolphin-2.8.1-1.1b-gguf/blob/main/TinyDolphin-2.8.1-1.1b.IQ3_S.gguf) | IQ3_S | 0.47GB |
20
+ | [TinyDolphin-2.8.1-1.1b.Q3_K_S.gguf](https://huggingface.co/RichardErkhov/cognitivecomputations_-_TinyDolphin-2.8.1-1.1b-gguf/blob/main/TinyDolphin-2.8.1-1.1b.Q3_K_S.gguf) | Q3_K_S | 0.47GB |
21
+ | [TinyDolphin-2.8.1-1.1b.IQ3_M.gguf](https://huggingface.co/RichardErkhov/cognitivecomputations_-_TinyDolphin-2.8.1-1.1b-gguf/blob/main/TinyDolphin-2.8.1-1.1b.IQ3_M.gguf) | IQ3_M | 0.48GB |
22
+ | [TinyDolphin-2.8.1-1.1b.Q3_K.gguf](https://huggingface.co/RichardErkhov/cognitivecomputations_-_TinyDolphin-2.8.1-1.1b-gguf/blob/main/TinyDolphin-2.8.1-1.1b.Q3_K.gguf) | Q3_K | 0.51GB |
23
+ | [TinyDolphin-2.8.1-1.1b.Q3_K_M.gguf](https://huggingface.co/RichardErkhov/cognitivecomputations_-_TinyDolphin-2.8.1-1.1b-gguf/blob/main/TinyDolphin-2.8.1-1.1b.Q3_K_M.gguf) | Q3_K_M | 0.51GB |
24
+ | [TinyDolphin-2.8.1-1.1b.Q3_K_L.gguf](https://huggingface.co/RichardErkhov/cognitivecomputations_-_TinyDolphin-2.8.1-1.1b-gguf/blob/main/TinyDolphin-2.8.1-1.1b.Q3_K_L.gguf) | Q3_K_L | 0.55GB |
25
+ | [TinyDolphin-2.8.1-1.1b.IQ4_XS.gguf](https://huggingface.co/RichardErkhov/cognitivecomputations_-_TinyDolphin-2.8.1-1.1b-gguf/blob/main/TinyDolphin-2.8.1-1.1b.IQ4_XS.gguf) | IQ4_XS | 0.57GB |
26
+ | [TinyDolphin-2.8.1-1.1b.Q4_0.gguf](https://huggingface.co/RichardErkhov/cognitivecomputations_-_TinyDolphin-2.8.1-1.1b-gguf/blob/main/TinyDolphin-2.8.1-1.1b.Q4_0.gguf) | Q4_0 | 0.59GB |
27
+ | [TinyDolphin-2.8.1-1.1b.IQ4_NL.gguf](https://huggingface.co/RichardErkhov/cognitivecomputations_-_TinyDolphin-2.8.1-1.1b-gguf/blob/main/TinyDolphin-2.8.1-1.1b.IQ4_NL.gguf) | IQ4_NL | 0.6GB |
28
+ | [TinyDolphin-2.8.1-1.1b.Q4_K_S.gguf](https://huggingface.co/RichardErkhov/cognitivecomputations_-_TinyDolphin-2.8.1-1.1b-gguf/blob/main/TinyDolphin-2.8.1-1.1b.Q4_K_S.gguf) | Q4_K_S | 0.6GB |
29
+ | [TinyDolphin-2.8.1-1.1b.Q4_K.gguf](https://huggingface.co/RichardErkhov/cognitivecomputations_-_TinyDolphin-2.8.1-1.1b-gguf/blob/main/TinyDolphin-2.8.1-1.1b.Q4_K.gguf) | Q4_K | 0.62GB |
30
+ | [TinyDolphin-2.8.1-1.1b.Q4_K_M.gguf](https://huggingface.co/RichardErkhov/cognitivecomputations_-_TinyDolphin-2.8.1-1.1b-gguf/blob/main/TinyDolphin-2.8.1-1.1b.Q4_K_M.gguf) | Q4_K_M | 0.62GB |
31
+ | [TinyDolphin-2.8.1-1.1b.Q4_1.gguf](https://huggingface.co/RichardErkhov/cognitivecomputations_-_TinyDolphin-2.8.1-1.1b-gguf/blob/main/TinyDolphin-2.8.1-1.1b.Q4_1.gguf) | Q4_1 | 0.65GB |
32
+ | [TinyDolphin-2.8.1-1.1b.Q5_0.gguf](https://huggingface.co/RichardErkhov/cognitivecomputations_-_TinyDolphin-2.8.1-1.1b-gguf/blob/main/TinyDolphin-2.8.1-1.1b.Q5_0.gguf) | Q5_0 | 0.71GB |
33
+ | [TinyDolphin-2.8.1-1.1b.Q5_K_S.gguf](https://huggingface.co/RichardErkhov/cognitivecomputations_-_TinyDolphin-2.8.1-1.1b-gguf/blob/main/TinyDolphin-2.8.1-1.1b.Q5_K_S.gguf) | Q5_K_S | 0.71GB |
34
+ | [TinyDolphin-2.8.1-1.1b.Q5_K.gguf](https://huggingface.co/RichardErkhov/cognitivecomputations_-_TinyDolphin-2.8.1-1.1b-gguf/blob/main/TinyDolphin-2.8.1-1.1b.Q5_K.gguf) | Q5_K | 0.73GB |
35
+ | [TinyDolphin-2.8.1-1.1b.Q5_K_M.gguf](https://huggingface.co/RichardErkhov/cognitivecomputations_-_TinyDolphin-2.8.1-1.1b-gguf/blob/main/TinyDolphin-2.8.1-1.1b.Q5_K_M.gguf) | Q5_K_M | 0.73GB |
36
+ | [TinyDolphin-2.8.1-1.1b.Q5_1.gguf](https://huggingface.co/RichardErkhov/cognitivecomputations_-_TinyDolphin-2.8.1-1.1b-gguf/blob/main/TinyDolphin-2.8.1-1.1b.Q5_1.gguf) | Q5_1 | 0.77GB |
37
+ | [TinyDolphin-2.8.1-1.1b.Q6_K.gguf](https://huggingface.co/RichardErkhov/cognitivecomputations_-_TinyDolphin-2.8.1-1.1b-gguf/blob/main/TinyDolphin-2.8.1-1.1b.Q6_K.gguf) | Q6_K | 0.84GB |
38
+ | [TinyDolphin-2.8.1-1.1b.Q8_0.gguf](https://huggingface.co/RichardErkhov/cognitivecomputations_-_TinyDolphin-2.8.1-1.1b-gguf/blob/main/TinyDolphin-2.8.1-1.1b.Q8_0.gguf) | Q8_0 | 1.09GB |
39
+
40
+
41
+
42
+
43
+ Original model description:
44
+ ---
45
+ license: apache-2.0
46
+ datasets:
47
+ - cerebras/SlimPajama-627B
48
+ - bigcode/starcoderdata
49
+ - teknium/openhermes
50
+ language:
51
+ - en
52
+ ---
53
+
54
+
55
+ # TinyDolphin-2.8.1-1.1b
56
+
57
+ ![image/jpeg](https://cdn-uploads.huggingface.co/production/uploads/655dc641accde1bbc8b41aec/qUCZv8vxXDXi9r-9c0jvP.jpeg)
58
+
59
+ Join Our Discord! https://discord.gg/vT3sktQ3zb
60
+
61
+ This is an version 2 of a model trained on 2 3090's by Kearm on the new Dolphin 2.8 dataset by Eric Hartford https://erichartford.com/dolphin 🐬
62
+
63
+ For this version we increased the epochs as well as refined the datasets used.
64
+
65
+ ## Example Outputs
66
+
67
+ TBD
68
+
69
+ Support my efforts! https://ko-fi.com/kearm
70
+
71
+ # Orignal Model Card Below
72
+
73
+ # TinyLlama-1.1B
74
+ </div>
75
+
76
+ https://github.com/jzhang38/TinyLlama
77
+
78
+ The TinyLlama project aims to **pretrain** a **1.1B Llama model on 3 trillion tokens**. With some proper optimization, we can achieve this within a span of "just" 90 days using 16 A100-40G GPUs 🚀🚀. The training has started on 2023-09-01.
79
+
80
+ We adopted exactly the same architecture and tokenizer as Llama 2. This means TinyLlama can be plugged and played in many open-source projects built upon Llama. Besides, TinyLlama is compact with only 1.1B parameters. This compactness allows it to cater to a multitude of applications demanding a restricted computation and memory footprint.
81
+
82
+ #### This Collection
83
+ This collection contains all checkpoints after the 1T fix. Branch name indicates the step and number of tokens seen.
84
+
85
+ #### Eval
86
+
87
+ | Model | Pretrain Tokens | HellaSwag | Obqa | WinoGrande | ARC_c | ARC_e | boolq | piqa | avg |
88
+ |-------------------------------------------|-----------------|-----------|------|------------|-------|-------|-------|------|-----|
89
+ | Pythia-1.0B | 300B | 47.16 | 31.40| 53.43 | 27.05 | 48.99 | 60.83 | 69.21 | 48.30 |
90
+ | TinyLlama-1.1B-intermediate-step-50K-104b | 103B | 43.50 | 29.80| 53.28 | 24.32 | 44.91 | 59.66 | 67.30 | 46.11|
91
+ | TinyLlama-1.1B-intermediate-step-240k-503b| 503B | 49.56 |31.40 |55.80 |26.54 |48.32 |56.91 |69.42 | 48.28 |
92
+ | TinyLlama-1.1B-intermediate-step-480k-1007B | 1007B | 52.54 | 33.40 | 55.96 | 27.82 | 52.36 | 59.54 | 69.91 | 50.22 |
93
+ | TinyLlama-1.1B-intermediate-step-715k-1.5T | 1.5T | 53.68 | 35.20 | 58.33 | 29.18 | 51.89 | 59.08 | 71.65 | 51.29 |
94
+ | TinyLlama-1.1B-intermediate-step-955k-2T | 2T | 54.63 | 33.40 | 56.83 | 28.07 | 54.67 | 63.21 | 70.67 | 51.64 |
95
+ | TinyLlama-1.1B-intermediate-step-1195k-2.5T | 2.5T | 58.96 | 34.40 | 58.72 | 31.91 | 56.78 | 63.21 | 73.07 | 53.86|
96
+ | TinyLlama-1.1B-intermediate-step-1431k-3T | 3T | 59.20 | 36.00 | 59.12 | 30.12 | 55.25 | 57.83 | 73.29 | 52.99|
97
+