aisuko commited on
Commit
e24abf8
1 Parent(s): 0de92cf

Add gpt-2 gguf model

Browse files

Signed-off-by: Aisuko <urakiny@gmail.com>

.gitattributes CHANGED
@@ -33,3 +33,4 @@ 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
+ *.gguf filter=lfs diff=lfs merge=lfs -text
README.md CHANGED
@@ -1,3 +1,67 @@
1
  ---
2
  license: mit
3
  ---
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
1
  ---
2
  license: mit
3
  ---
4
+
5
+
6
+ # How to run
7
+ ```bash
8
+ git clone https://github.com/ggerganov/llama.cpp
9
+ cd llama.cpp
10
+
11
+ make -j<CPU counts-2>
12
+
13
+ ./llama-cli -m name.gguf -n 256 --repeat_penalty 1.0 --color -i -r "User:" -f prompts/chat-with-bob.txt
14
+ ```
15
+
16
+ # 117M model result is unuseful
17
+
18
+ ```markdown
19
+
20
+ system_info: n_threads = 4 / 8 | AVX = 1 | AVX_VNNI = 1 | AVX2 = 1 | AVX512 = 1 | AVX512_VBMI = 1 | AVX512_VNNI = 1 | AVX512_BF16 = 1 | FMA = 1 | NEON = 0 | SVE = 0 | ARM_FMA = 0 | F16C = 1 | FP16_VA = 0 | WASM_SIMD = 0 | BLAS = 0 | SSE3 = 1 | SSSE3 = 1 | VSX = 0 | MATMUL_INT8 = 0 | LLAMAFILE = 1 |
21
+ main: interactive mode on.
22
+ Reverse prompt: 'User:'
23
+ sampling:
24
+ repeat_last_n = 64, repeat_penalty = 1.000, frequency_penalty = 0.000, presence_penalty = 0.000
25
+ top_k = 40, tfs_z = 1.000, top_p = 0.950, min_p = 0.050, typical_p = 1.000, temp = 0.800
26
+ mirostat = 0, mirostat_lr = 0.100, mirostat_ent = 5.000
27
+ sampling order:
28
+ CFG -> Penalties -> top_k -> tfs_z -> typical_p -> top_p -> min_p -> temperature
29
+ generate: n_ctx = 1024, n_batch = 2048, n_predict = 256, n_keep = 0
30
+
31
+
32
+ == Running in interactive mode. ==
33
+ - Press Ctrl+C to interject at any time.
34
+ - Press Return to return control to the AI.
35
+ - To return control without starting a new line, end your input with '/'.
36
+ - If you want to submit another line, end your input with '\'.
37
+
38
+ Transcript of a dialog, where the User interacts with an Assistant named Bob. Bob is helpful, kind, honest, good at writing, and never fails to answer the User's requests immediately and with precision.
39
+
40
+ User: Hello, Bob.
41
+ Bob: Hello. How may I help you today?
42
+ User: Please tell me the largest city in Europe.
43
+ Bob: Sure. The largest city in Europe is Moscow, the capital of Russia.
44
+ User:What is the largest city in Australia?
45
+ Bob: The biggest city in Australia is New York City.
46
+
47
+ User:New York is a city of US
48
+
49
+ Bob: The US is a city of the US.
50
+
51
+ User:thanks
52
+
53
+ User, you do have a question.
54
+
55
+ User, you have a question.
56
+
57
+ Bob: Alright. You are an early user.
58
+
59
+ User:
60
+
61
+ llama_print_timings: load time = 29.65 ms
62
+ llama_print_timings: sample time = 2.09 ms / 66 runs ( 0.03 ms per token, 31548.76 tokens per second)
63
+ llama_print_timings: prompt eval time = 25528.34 ms / 116 tokens ( 220.07 ms per token, 4.54 tokens per second)
64
+ llama_print_timings: eval time = 212.84 ms / 63 runs ( 3.38 ms per token, 296.00 tokens per second)
65
+ llama_print_timings: total time = 69083.22 ms / 179 tokens
66
+ (llama.cpp-4B8ytfKj-py3.10) ec2-user@ip-10-110-145-102:~/workspace/gguf/llama.cpp$
67
+ ```
ggml-model-Q4_K_M-v2.gguf ADDED
@@ -0,0 +1,3 @@
 
 
 
 
1
+ version https://git-lfs.github.com/spec/v1
2
+ oid sha256:6281d15f9663025df2dffc5f4a4a3850bd833b0d20e1d254bd0dd854f7c722a4
3
+ size 112858624
ggml-model-Q4_K_M.gguf ADDED
@@ -0,0 +1,3 @@
 
 
 
 
1
+ version https://git-lfs.github.com/spec/v1
2
+ oid sha256:6281d15f9663025df2dffc5f4a4a3850bd833b0d20e1d254bd0dd854f7c722a4
3
+ size 112858624
ggml-model-f16.gguf ADDED
@@ -0,0 +1,3 @@
 
 
 
 
1
+ version https://git-lfs.github.com/spec/v1
2
+ oid sha256:1e9d608e30bb653af1fa113bf725b7e41a37d1494a5db836e59bb3a599d5e6fd
3
+ size 329664992