Hugging Face's logo

ISTA-DASLab
/
lossless-lm-dev

Safetensors
Model card Files
xet
 Community
helcig commited on
Commit
c3210e7
·
verified ·
1 Parent(s): f21cc6d

Add files using upload-large-folder tool

Browse files
Files changed (29) hide show
  1. .gitattributes +4 -0
  2. 4bit-asym_8bit-sym/Llama-3.3-70B-Instruct/ll_bsearch_kl0.01_eap0.985_sha_bw6.96_4-8bit_grouped_seed1234/model-00001-of-00020.safetensors +3 -0
  3. 4bit-asym_8bit-sym/Llama-3.3-70B-Instruct/ll_bsearch_kl0.01_eap0.985_sha_bw6.96_4-8bit_grouped_seed1234/model-00002-of-00020.safetensors +3 -0
  4. 4bit-asym_8bit-sym/Llama-3.3-70B-Instruct/ll_bsearch_kl0.01_eap0.985_sha_bw7.01_4-8bit_grouped_seed42/model-00001-of-00020.safetensors +3 -0
  5. 4bit-asym_8bit-sym/Llama-3.3-70B-Instruct/ll_bsearch_kl0.01_eap0.985_sha_bw7.01_4-8bit_grouped_seed42/model-00002-of-00020.safetensors +3 -0
  6. 4bit-asym_8bit-sym/Llama-3.3-70B-Instruct/ll_bsearch_kl0.01_eap0.99_sha_bw7.60_4-8bit_grouped_seed1234/model-00001-of-00020.safetensors +3 -0
  7. 4bit-asym_8bit-sym/Llama-3.3-70B-Instruct/ll_bsearch_kl0.01_eap0.99_sha_bw7.60_4-8bit_grouped_seed1234/model-00002-of-00020.safetensors +3 -0
  8. 4bit-asym_8bit-sym/Qwen3-8B/ll_bsearch_kl0.01_eap0.985_sha_bw6.96_4-8bit_grouped_seed1234/model-00001-of-00004.safetensors +3 -0
  9. 4bit-asym_8bit-sym/Qwen3-8B/ll_bsearch_kl0.01_eap0.985_sha_bw6.96_4-8bit_grouped_seed1234/model-00002-of-00004.safetensors +3 -0
  10. 4bit-asym_8bit-sym/Qwen3-8B/ll_bsearch_kl0.01_eap0.985_sha_bw6.96_4-8bit_grouped_seed1234/model-00003-of-00004.safetensors +3 -0
  11. 4bit-asym_8bit-sym/Qwen3-8B/ll_bsearch_kl0.01_eap0.985_sha_bw6.96_4-8bit_grouped_seed1234/model-00004-of-00004.safetensors +3 -0
  12. 4bit-asym_8bit-sym/Qwen3-8B/ll_bsearch_kl0.01_eap0.985_sha_bw6.96_4-8bit_grouped_seed1234/tokenizer.json +3 -0
  13. 4bit-asym_8bit-sym/Qwen3-8B/ll_bsearch_kl0.01_eap0.985_sha_bw7.01_4-8bit_grouped_seed42/model-00001-of-00004.safetensors +3 -0
  14. 4bit-asym_8bit-sym/Qwen3-8B/ll_bsearch_kl0.01_eap0.985_sha_bw7.01_4-8bit_grouped_seed42/model-00002-of-00004.safetensors +3 -0
  15. 4bit-asym_8bit-sym/Qwen3-8B/ll_bsearch_kl0.01_eap0.985_sha_bw7.01_4-8bit_grouped_seed42/model-00003-of-00004.safetensors +3 -0
  16. 4bit-asym_8bit-sym/Qwen3-8B/ll_bsearch_kl0.01_eap0.985_sha_bw7.01_4-8bit_grouped_seed42/model-00004-of-00004.safetensors +3 -0
  17. 4bit-asym_8bit-sym/Qwen3-8B/ll_bsearch_kl0.01_eap0.985_sha_bw7.01_4-8bit_grouped_seed42/tokenizer.json +3 -0
  18. 4bit-asym_8bit-sym/Qwen3-8B/ll_bsearch_kl0.01_eap0.99_sha_bw7.60_4-8bit_grouped_seed1234/model-00001-of-00004.safetensors +3 -0
  19. 4bit-asym_8bit-sym/Qwen3-8B/ll_bsearch_kl0.01_eap0.99_sha_bw7.60_4-8bit_grouped_seed1234/model-00002-of-00004.safetensors +3 -0
  20. 4bit-asym_8bit-sym/Qwen3-8B/ll_bsearch_kl0.01_eap0.99_sha_bw7.60_4-8bit_grouped_seed1234/model-00003-of-00004.safetensors +3 -0
  21. 4bit-asym_8bit-sym/Qwen3-8B/ll_bsearch_kl0.01_eap0.99_sha_bw7.60_4-8bit_grouped_seed1234/model-00004-of-00004.safetensors +3 -0
  22. 4bit-asym_8bit-sym/Qwen3-8B/ll_bsearch_kl0.01_eap0.99_sha_bw7.60_4-8bit_grouped_seed1234/tokenizer.json +3 -0
  23. 4bit-asym_8bit-sym/Qwen3-8B/ll_bsearch_kl0.01_eap0.99_sha_bw7.61_4-8bit_grouped_seed42/README.md +18 -0
  24. 4bit-asym_8bit-sym/Qwen3-8B/ll_bsearch_kl0.01_eap0.99_sha_bw7.61_4-8bit_grouped_seed42/config.json +68 -0
  25. 4bit-asym_8bit-sym/Qwen3-8B/ll_bsearch_kl0.01_eap0.99_sha_bw7.61_4-8bit_grouped_seed42/model-00001-of-00004.safetensors +3 -0
  26. 4bit-asym_8bit-sym/Qwen3-8B/ll_bsearch_kl0.01_eap0.99_sha_bw7.61_4-8bit_grouped_seed42/model-00002-of-00004.safetensors +3 -0
  27. 4bit-asym_8bit-sym/Qwen3-8B/ll_bsearch_kl0.01_eap0.99_sha_bw7.61_4-8bit_grouped_seed42/model-00003-of-00004.safetensors +3 -0
  28. 4bit-asym_8bit-sym/Qwen3-8B/ll_bsearch_kl0.01_eap0.99_sha_bw7.61_4-8bit_grouped_seed42/model-00004-of-00004.safetensors +3 -0
  29. 4bit-asym_8bit-sym/Qwen3-8B/ll_bsearch_kl0.01_eap0.99_sha_bw7.61_4-8bit_grouped_seed42/tokenizer.json +3 -0
.gitattributes CHANGED
@@ -104,3 +104,7 @@ Llama-3.3-70B-Instruct/ll_bsearch_kl0.005_eap0.985_sha_bw6.93_4-8bit_seed42/toke
104
  Llama-3.3-70B-Instruct/ll_bsearch_kl0.01_eap0.985_sha_bw6.93_4-8bit_seed1234/tokenizer.json filter=lfs diff=lfs merge=lfs -text
105
  Llama-3.3-70B-Instruct/ll_bsearch_kl0.005_eap0.985_sha_bw6.93_4-8bit_seed1234/tokenizer.json filter=lfs diff=lfs merge=lfs -text
106
  qtip/Llama-3.1-8B-Instruct/4bit/tokenizer.json filter=lfs diff=lfs merge=lfs -text
 
 
 
 
 
104
  Llama-3.3-70B-Instruct/ll_bsearch_kl0.01_eap0.985_sha_bw6.93_4-8bit_seed1234/tokenizer.json filter=lfs diff=lfs merge=lfs -text
105
  Llama-3.3-70B-Instruct/ll_bsearch_kl0.005_eap0.985_sha_bw6.93_4-8bit_seed1234/tokenizer.json filter=lfs diff=lfs merge=lfs -text
106
  qtip/Llama-3.1-8B-Instruct/4bit/tokenizer.json filter=lfs diff=lfs merge=lfs -text
107
+ 4bit-asym_8bit-sym/Qwen3-8B/ll_bsearch_kl0.01_eap0.99_sha_bw7.60_4-8bit_grouped_seed1234/tokenizer.json filter=lfs diff=lfs merge=lfs -text
108
+ 4bit-asym_8bit-sym/Qwen3-8B/ll_bsearch_kl0.01_eap0.985_sha_bw7.01_4-8bit_grouped_seed42/tokenizer.json filter=lfs diff=lfs merge=lfs -text
109
+ 4bit-asym_8bit-sym/Qwen3-8B/ll_bsearch_kl0.01_eap0.985_sha_bw6.96_4-8bit_grouped_seed1234/tokenizer.json filter=lfs diff=lfs merge=lfs -text
110
+ 4bit-asym_8bit-sym/Qwen3-8B/ll_bsearch_kl0.01_eap0.99_sha_bw7.61_4-8bit_grouped_seed42/tokenizer.json filter=lfs diff=lfs merge=lfs -text
4bit-asym_8bit-sym/Llama-3.3-70B-Instruct/ll_bsearch_kl0.01_eap0.985_sha_bw6.96_4-8bit_grouped_seed1234/model-00001-of-00020.safetensors ADDED
@@ -0,0 +1,3 @@
 
 
 
 
1
+ version https://git-lfs.github.com/spec/v1
2
+ oid sha256:0d9fba19ed80d0470e6e719afb7e9d88fa4732e0768f49e953fd41bcb2e5fee5
3
+ size 4987264784
4bit-asym_8bit-sym/Llama-3.3-70B-Instruct/ll_bsearch_kl0.01_eap0.985_sha_bw6.96_4-8bit_grouped_seed1234/model-00002-of-00020.safetensors ADDED
@@ -0,0 +1,3 @@
 
 
 
 
1
+ version https://git-lfs.github.com/spec/v1
2
+ oid sha256:4225fce600836f7cbbe4bc12713b27c217bfdb32f7a91cd2f67b9035e2d122fe
3
+ size 4916163664
4bit-asym_8bit-sym/Llama-3.3-70B-Instruct/ll_bsearch_kl0.01_eap0.985_sha_bw7.01_4-8bit_grouped_seed42/model-00001-of-00020.safetensors ADDED
@@ -0,0 +1,3 @@
 
 
 
 
1
+ version https://git-lfs.github.com/spec/v1
2
+ oid sha256:4336fae3d8ac8799482b125f003e11561737b131a38128cd0ba6de83e8bef248
3
+ size 4987264784
4bit-asym_8bit-sym/Llama-3.3-70B-Instruct/ll_bsearch_kl0.01_eap0.985_sha_bw7.01_4-8bit_grouped_seed42/model-00002-of-00020.safetensors ADDED
@@ -0,0 +1,3 @@
 
 
 
 
1
+ version https://git-lfs.github.com/spec/v1
2
+ oid sha256:4225fce600836f7cbbe4bc12713b27c217bfdb32f7a91cd2f67b9035e2d122fe
3
+ size 4916163664
4bit-asym_8bit-sym/Llama-3.3-70B-Instruct/ll_bsearch_kl0.01_eap0.99_sha_bw7.60_4-8bit_grouped_seed1234/model-00001-of-00020.safetensors ADDED
@@ -0,0 +1,3 @@
 
 
 
 
1
+ version https://git-lfs.github.com/spec/v1
2
+ oid sha256:23c71f49dbf3d21363a845f1a628271fd3a597334fa1a5b547d9f7ca09e004f6
3
+ size 4987264784
4bit-asym_8bit-sym/Llama-3.3-70B-Instruct/ll_bsearch_kl0.01_eap0.99_sha_bw7.60_4-8bit_grouped_seed1234/model-00002-of-00020.safetensors ADDED
@@ -0,0 +1,3 @@
 
 
 
 
1
+ version https://git-lfs.github.com/spec/v1
2
+ oid sha256:d4abbf92a25bf210129e16512f57a1361d59b5a5e353e88bf4ee83fe7c44b757
3
+ size 4916163664
4bit-asym_8bit-sym/Qwen3-8B/ll_bsearch_kl0.01_eap0.985_sha_bw6.96_4-8bit_grouped_seed1234/model-00001-of-00004.safetensors ADDED
@@ -0,0 +1,3 @@
 
 
 
 
1
+ version https://git-lfs.github.com/spec/v1
2
+ oid sha256:645c4bdb56f604f6d1832d46831ac978d8905b342ecbb7a5077824d98f501447
3
+ size 4902257592
4bit-asym_8bit-sym/Qwen3-8B/ll_bsearch_kl0.01_eap0.985_sha_bw6.96_4-8bit_grouped_seed1234/model-00002-of-00004.safetensors ADDED
@@ -0,0 +1,3 @@
 
 
 
 
1
+ version https://git-lfs.github.com/spec/v1
2
+ oid sha256:bec898c30936f66505c9742799965e1ce640086dcb66557a865ef4db1c6b982f
3
+ size 4915960224
4bit-asym_8bit-sym/Qwen3-8B/ll_bsearch_kl0.01_eap0.985_sha_bw6.96_4-8bit_grouped_seed1234/model-00003-of-00004.safetensors ADDED
@@ -0,0 +1,3 @@
 
 
 
 
1
+ version https://git-lfs.github.com/spec/v1
2
+ oid sha256:eab73409f60086f72732402e427aaa14b4b7b34b6bc419a9bfb8ca859182292d
3
+ size 4983068360
4bit-asym_8bit-sym/Qwen3-8B/ll_bsearch_kl0.01_eap0.985_sha_bw6.96_4-8bit_grouped_seed1234/model-00004-of-00004.safetensors ADDED
@@ -0,0 +1,3 @@
 
 
 
 
1
+ version https://git-lfs.github.com/spec/v1
2
+ oid sha256:993a90f396a23b1a77734ab696608ad83a1c8468e34238734f1cd6203c0c0e9a
3
+ size 1580230248
4bit-asym_8bit-sym/Qwen3-8B/ll_bsearch_kl0.01_eap0.985_sha_bw6.96_4-8bit_grouped_seed1234/tokenizer.json ADDED
@@ -0,0 +1,3 @@
 
 
 
 
1
+ version https://git-lfs.github.com/spec/v1
2
+ oid sha256:aeb13307a71acd8fe81861d94ad54ab689df773318809eed3cbe794b4492dae4
3
+ size 11422654
4bit-asym_8bit-sym/Qwen3-8B/ll_bsearch_kl0.01_eap0.985_sha_bw7.01_4-8bit_grouped_seed42/model-00001-of-00004.safetensors ADDED
@@ -0,0 +1,3 @@
 
 
 
 
1
+ version https://git-lfs.github.com/spec/v1
2
+ oid sha256:758c2bf05cc905a1de0e059405873270f55db2989537d84a004d451fbfbcfd8a
3
+ size 4902257592
4bit-asym_8bit-sym/Qwen3-8B/ll_bsearch_kl0.01_eap0.985_sha_bw7.01_4-8bit_grouped_seed42/model-00002-of-00004.safetensors ADDED
@@ -0,0 +1,3 @@
 
 
 
 
1
+ version https://git-lfs.github.com/spec/v1
2
+ oid sha256:ef73b924778bb0eb0bc09f38ae583c48dad6602c0fb6ed35d7927ead1f9dee5f
3
+ size 4915960224
4bit-asym_8bit-sym/Qwen3-8B/ll_bsearch_kl0.01_eap0.985_sha_bw7.01_4-8bit_grouped_seed42/model-00003-of-00004.safetensors ADDED
@@ -0,0 +1,3 @@
 
 
 
 
1
+ version https://git-lfs.github.com/spec/v1
2
+ oid sha256:eab73409f60086f72732402e427aaa14b4b7b34b6bc419a9bfb8ca859182292d
3
+ size 4983068360
4bit-asym_8bit-sym/Qwen3-8B/ll_bsearch_kl0.01_eap0.985_sha_bw7.01_4-8bit_grouped_seed42/model-00004-of-00004.safetensors ADDED
@@ -0,0 +1,3 @@
 
 
 
 
1
+ version https://git-lfs.github.com/spec/v1
2
+ oid sha256:993a90f396a23b1a77734ab696608ad83a1c8468e34238734f1cd6203c0c0e9a
3
+ size 1580230248
4bit-asym_8bit-sym/Qwen3-8B/ll_bsearch_kl0.01_eap0.985_sha_bw7.01_4-8bit_grouped_seed42/tokenizer.json ADDED
@@ -0,0 +1,3 @@
 
 
 
 
1
+ version https://git-lfs.github.com/spec/v1
2
+ oid sha256:aeb13307a71acd8fe81861d94ad54ab689df773318809eed3cbe794b4492dae4
3
+ size 11422654
4bit-asym_8bit-sym/Qwen3-8B/ll_bsearch_kl0.01_eap0.99_sha_bw7.60_4-8bit_grouped_seed1234/model-00001-of-00004.safetensors ADDED
@@ -0,0 +1,3 @@
 
 
 
 
1
+ version https://git-lfs.github.com/spec/v1
2
+ oid sha256:95c3db24457fce943a6d489664d3c0902a3cb292c49c198c58e0724c0ea78cd8
3
+ size 4902257592
4bit-asym_8bit-sym/Qwen3-8B/ll_bsearch_kl0.01_eap0.99_sha_bw7.60_4-8bit_grouped_seed1234/model-00002-of-00004.safetensors ADDED
@@ -0,0 +1,3 @@
 
 
 
 
1
+ version https://git-lfs.github.com/spec/v1
2
+ oid sha256:b01700f0ed8b37c3393da4988f163638bbd136b0470e39c4c2afc83f00af6ca3
3
+ size 4915960224
4bit-asym_8bit-sym/Qwen3-8B/ll_bsearch_kl0.01_eap0.99_sha_bw7.60_4-8bit_grouped_seed1234/model-00003-of-00004.safetensors ADDED
@@ -0,0 +1,3 @@
 
 
 
 
1
+ version https://git-lfs.github.com/spec/v1
2
+ oid sha256:9e5b3bf736bcd1d3d165e2cfb282ef3c2151c726dbe008cbe2be7b8096594b2c
3
+ size 4983068360
4bit-asym_8bit-sym/Qwen3-8B/ll_bsearch_kl0.01_eap0.99_sha_bw7.60_4-8bit_grouped_seed1234/model-00004-of-00004.safetensors ADDED
@@ -0,0 +1,3 @@
 
 
 
 
1
+ version https://git-lfs.github.com/spec/v1
2
+ oid sha256:993a90f396a23b1a77734ab696608ad83a1c8468e34238734f1cd6203c0c0e9a
3
+ size 1580230248
4bit-asym_8bit-sym/Qwen3-8B/ll_bsearch_kl0.01_eap0.99_sha_bw7.60_4-8bit_grouped_seed1234/tokenizer.json ADDED
@@ -0,0 +1,3 @@
 
 
 
 
1
+ version https://git-lfs.github.com/spec/v1
2
+ oid sha256:aeb13307a71acd8fe81861d94ad54ab689df773318809eed3cbe794b4492dae4
3
+ size 11422654
4bit-asym_8bit-sym/Qwen3-8B/ll_bsearch_kl0.01_eap0.99_sha_bw7.61_4-8bit_grouped_seed42/README.md ADDED
@@ -0,0 +1,18 @@
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
1
+ # Quantized Model Checkpoint
2
+
3
+ **Base model:** Qwen/Qwen3-8B
4
+
5
+ **Average bitwidth:** 7.6053
6
+
7
+ **Sensitivity method:** shapley
8
+
9
+ **Constraints:**
10
+ - max_kl: 0.01
11
+ - min_eap: 0.99
12
+
13
+ **Metrics:**
14
+ - final_kl: 0.001590
15
+ - final_eap: 0.990007
16
+ - final_etl: 0.009993
17
+
18
+ See `quantization_config.txt` for full configuration details.
4bit-asym_8bit-sym/Qwen3-8B/ll_bsearch_kl0.01_eap0.99_sha_bw7.61_4-8bit_grouped_seed42/config.json ADDED
@@ -0,0 +1,68 @@
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
1
+ {
2
+ "architectures": [
3
+ "Qwen3ForCausalLM"
4
+ ],
5
+ "attention_bias": false,
6
+ "attention_dropout": 0.0,
7
+ "bos_token_id": 151643,
8
+ "dtype": "float16",
9
+ "eos_token_id": 151645,
10
+ "head_dim": 128,
11
+ "hidden_act": "silu",
12
+ "hidden_size": 4096,
13
+ "initializer_range": 0.02,
14
+ "intermediate_size": 12288,
15
+ "layer_types": [
16
+ "full_attention",
17
+ "full_attention",
18
+ "full_attention",
19
+ "full_attention",
20
+ "full_attention",
21
+ "full_attention",
22
+ "full_attention",
23
+ "full_attention",
24
+ "full_attention",
25
+ "full_attention",
26
+ "full_attention",
27
+ "full_attention",
28
+ "full_attention",
29
+ "full_attention",
30
+ "full_attention",
31
+ "full_attention",
32
+ "full_attention",
33
+ "full_attention",
34
+ "full_attention",
35
+ "full_attention",
36
+ "full_attention",
37
+ "full_attention",
38
+ "full_attention",
39
+ "full_attention",
40
+ "full_attention",
41
+ "full_attention",
42
+ "full_attention",
43
+ "full_attention",
44
+ "full_attention",
45
+ "full_attention",
46
+ "full_attention",
47
+ "full_attention",
48
+ "full_attention",
49
+ "full_attention",
50
+ "full_attention",
51
+ "full_attention"
52
+ ],
53
+ "max_position_embeddings": 40960,
54
+ "max_window_layers": 36,
55
+ "model_type": "qwen3",
56
+ "num_attention_heads": 32,
57
+ "num_hidden_layers": 36,
58
+ "num_key_value_heads": 8,
59
+ "rms_norm_eps": 1e-06,
60
+ "rope_scaling": null,
61
+ "rope_theta": 1000000,
62
+ "sliding_window": null,
63
+ "tie_word_embeddings": false,
64
+ "transformers_version": "4.57.3",
65
+ "use_cache": false,
66
+ "use_sliding_window": false,
67
+ "vocab_size": 151936
68
+ }
4bit-asym_8bit-sym/Qwen3-8B/ll_bsearch_kl0.01_eap0.99_sha_bw7.61_4-8bit_grouped_seed42/model-00001-of-00004.safetensors ADDED
@@ -0,0 +1,3 @@
 
 
 
 
1
+ version https://git-lfs.github.com/spec/v1
2
+ oid sha256:ef91c25af1827e3d3bae901256f2d3e671b086bed10ffb32f453ae566746ade3
3
+ size 4902257592
4bit-asym_8bit-sym/Qwen3-8B/ll_bsearch_kl0.01_eap0.99_sha_bw7.61_4-8bit_grouped_seed42/model-00002-of-00004.safetensors ADDED
@@ -0,0 +1,3 @@
 
 
 
 
1
+ version https://git-lfs.github.com/spec/v1
2
+ oid sha256:b01700f0ed8b37c3393da4988f163638bbd136b0470e39c4c2afc83f00af6ca3
3
+ size 4915960224
4bit-asym_8bit-sym/Qwen3-8B/ll_bsearch_kl0.01_eap0.99_sha_bw7.61_4-8bit_grouped_seed42/model-00003-of-00004.safetensors ADDED
@@ -0,0 +1,3 @@
 
 
 
 
1
+ version https://git-lfs.github.com/spec/v1
2
+ oid sha256:9e5b3bf736bcd1d3d165e2cfb282ef3c2151c726dbe008cbe2be7b8096594b2c
3
+ size 4983068360
4bit-asym_8bit-sym/Qwen3-8B/ll_bsearch_kl0.01_eap0.99_sha_bw7.61_4-8bit_grouped_seed42/model-00004-of-00004.safetensors ADDED
@@ -0,0 +1,3 @@
 
 
 
 
1
+ version https://git-lfs.github.com/spec/v1
2
+ oid sha256:993a90f396a23b1a77734ab696608ad83a1c8468e34238734f1cd6203c0c0e9a
3
+ size 1580230248
4bit-asym_8bit-sym/Qwen3-8B/ll_bsearch_kl0.01_eap0.99_sha_bw7.61_4-8bit_grouped_seed42/tokenizer.json ADDED
@@ -0,0 +1,3 @@
 
 
 
 
1
+ version https://git-lfs.github.com/spec/v1
2
+ oid sha256:aeb13307a71acd8fe81861d94ad54ab689df773318809eed3cbe794b4492dae4
3
+ size 11422654