Add files using upload-large-folder tool
Browse files- README.md +16 -4
- model-00001-of-00019.safetensors +1 -1
- model-00002-of-00019.safetensors +1 -1
- model-00003-of-00019.safetensors +1 -1
- model-00004-of-00019.safetensors +1 -1
- model-00005-of-00019.safetensors +1 -1
- model-00006-of-00019.safetensors +1 -1
- model-00007-of-00019.safetensors +1 -1
- model-00008-of-00019.safetensors +1 -1
- model-00009-of-00019.safetensors +1 -1
- model-00010-of-00019.safetensors +1 -1
- model-00011-of-00019.safetensors +1 -1
- model-00012-of-00019.safetensors +1 -1
- model-00013-of-00019.safetensors +1 -1
- model-00014-of-00019.safetensors +1 -1
- model-00015-of-00019.safetensors +1 -1
- model-00016-of-00019.safetensors +1 -1
- model-00017-of-00019.safetensors +1 -1
- model-00018-of-00019.safetensors +1 -1
- model-00019-of-00019.safetensors +1 -1
README.md
CHANGED
|
@@ -9,8 +9,20 @@ tags:
|
|
| 9 |
|
| 10 |
# catalystsec/MiniMax-M2-3bit-DWQ
|
| 11 |
|
|
|
|
|
|
|
| 12 |
This model was quantized to 3-bit using DWQ with mlx-lm version **0.28.4**.
|
| 13 |
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 14 |
## Use with mlx
|
| 15 |
|
| 16 |
```bash
|
|
@@ -21,14 +33,14 @@ pip install mlx-lm
|
|
| 21 |
from mlx_lm import load, generate
|
| 22 |
|
| 23 |
model, tokenizer = load("catalystsec/MiniMax-M2-3bit-DWQ")
|
| 24 |
-
|
| 25 |
prompt = "hello"
|
| 26 |
|
| 27 |
if tokenizer.chat_template is not None:
|
| 28 |
-
messages = [{"role": "user", "content": prompt}]
|
| 29 |
prompt = tokenizer.apply_chat_template(
|
| 30 |
-
|
|
|
|
| 31 |
)
|
| 32 |
|
| 33 |
response = generate(model, tokenizer, prompt=prompt, verbose=True)
|
| 34 |
-
|
|
|
|
|
|
| 9 |
|
| 10 |
# catalystsec/MiniMax-M2-3bit-DWQ
|
| 11 |
|
| 12 |
+
**NOTE:** Currently re-uploading fixed version...
|
| 13 |
+
|
| 14 |
This model was quantized to 3-bit using DWQ with mlx-lm version **0.28.4**.
|
| 15 |
|
| 16 |
+
| Parameter | Value |
|
| 17 |
+
|---------------------------|--------------------------------|
|
| 18 |
+
| DWQ learning rate | 3e-7 |
|
| 19 |
+
| Batch size | 1 |
|
| 20 |
+
| Dataset | `allenai/tulu-3-sft-mixture` |
|
| 21 |
+
| Initial validation loss | 0.146 |
|
| 22 |
+
| Final validation loss | 0.088 |
|
| 23 |
+
| Relative KL reduction | ≈40 % |
|
| 24 |
+
| Tokens processed | ≈1.09 M |
|
| 25 |
+
|
| 26 |
## Use with mlx
|
| 27 |
|
| 28 |
```bash
|
|
|
|
| 33 |
from mlx_lm import load, generate
|
| 34 |
|
| 35 |
model, tokenizer = load("catalystsec/MiniMax-M2-3bit-DWQ")
|
|
|
|
| 36 |
prompt = "hello"
|
| 37 |
|
| 38 |
if tokenizer.chat_template is not None:
|
|
|
|
| 39 |
prompt = tokenizer.apply_chat_template(
|
| 40 |
+
[{"role": "user", "content": prompt}],
|
| 41 |
+
add_generation_prompt=True,
|
| 42 |
)
|
| 43 |
|
| 44 |
response = generate(model, tokenizer, prompt=prompt, verbose=True)
|
| 45 |
+
print(response)
|
| 46 |
+
```
|
model-00001-of-00019.safetensors
CHANGED
|
@@ -1,3 +1,3 @@
|
|
| 1 |
version https://git-lfs.github.com/spec/v1
|
| 2 |
-
oid sha256:
|
| 3 |
size 5105944657
|
|
|
|
| 1 |
version https://git-lfs.github.com/spec/v1
|
| 2 |
+
oid sha256:c5bfe6dde3aaeab86c7008e3beb094dbc0c77a97376fb7b06531185a0c364049
|
| 3 |
size 5105944657
|
model-00002-of-00019.safetensors
CHANGED
|
@@ -1,3 +1,3 @@
|
|
| 1 |
version https://git-lfs.github.com/spec/v1
|
| 2 |
-
oid sha256:
|
| 3 |
size 5345372518
|
|
|
|
| 1 |
version https://git-lfs.github.com/spec/v1
|
| 2 |
+
oid sha256:67f3716144cfcc1a158509abd3def029bb021b69fc4b1d96c205319bfd75bda4
|
| 3 |
size 5345372518
|
model-00003-of-00019.safetensors
CHANGED
|
@@ -1,3 +1,3 @@
|
|
| 1 |
version https://git-lfs.github.com/spec/v1
|
| 2 |
-
oid sha256:
|
| 3 |
size 5345372540
|
|
|
|
| 1 |
version https://git-lfs.github.com/spec/v1
|
| 2 |
+
oid sha256:be736a45db71b80a48f7e2ac681dff4a9b13a76c2325846b921eea84af68c119
|
| 3 |
size 5345372540
|
model-00004-of-00019.safetensors
CHANGED
|
@@ -1,3 +1,3 @@
|
|
| 1 |
version https://git-lfs.github.com/spec/v1
|
| 2 |
-
oid sha256:
|
| 3 |
size 5365554338
|
|
|
|
| 1 |
version https://git-lfs.github.com/spec/v1
|
| 2 |
+
oid sha256:bf49b11e0a5a2c07c89b6526220d5c8ca10c0761203f61cd4cd49bee88dd29d0
|
| 3 |
size 5365554338
|
model-00005-of-00019.safetensors
CHANGED
|
@@ -1,3 +1,3 @@
|
|
| 1 |
version https://git-lfs.github.com/spec/v1
|
| 2 |
-
oid sha256:
|
| 3 |
size 5345372592
|
|
|
|
| 1 |
version https://git-lfs.github.com/spec/v1
|
| 2 |
+
oid sha256:7d9d36395622551f6f20ca9112d00dfd89b4b2427e8eafa262764b7b19120717
|
| 3 |
size 5345372592
|
model-00006-of-00019.safetensors
CHANGED
|
@@ -1,3 +1,3 @@
|
|
| 1 |
version https://git-lfs.github.com/spec/v1
|
| 2 |
-
oid sha256:
|
| 3 |
size 5345372628
|
|
|
|
| 1 |
version https://git-lfs.github.com/spec/v1
|
| 2 |
+
oid sha256:2a52fc61640981f9ef296ea62f5d2d01f62cb1b7cd6f9b77c566f0a596266060
|
| 3 |
size 5345372628
|
model-00007-of-00019.safetensors
CHANGED
|
@@ -1,3 +1,3 @@
|
|
| 1 |
version https://git-lfs.github.com/spec/v1
|
| 2 |
-
oid sha256:
|
| 3 |
size 5365554314
|
|
|
|
| 1 |
version https://git-lfs.github.com/spec/v1
|
| 2 |
+
oid sha256:e6a332004cdb0ccf967af8b2d5629731499e23534aa5e7e5a4577d09080864b5
|
| 3 |
size 5365554314
|
model-00008-of-00019.safetensors
CHANGED
|
@@ -1,3 +1,3 @@
|
|
| 1 |
version https://git-lfs.github.com/spec/v1
|
| 2 |
-
oid sha256:
|
| 3 |
size 5345372624
|
|
|
|
| 1 |
version https://git-lfs.github.com/spec/v1
|
| 2 |
+
oid sha256:9655113ee876190505e266fb862b70787c654e704a8e3eb259086617d11f93b1
|
| 3 |
size 5345372624
|
model-00009-of-00019.safetensors
CHANGED
|
@@ -1,3 +1,3 @@
|
|
| 1 |
version https://git-lfs.github.com/spec/v1
|
| 2 |
-
oid sha256:
|
| 3 |
size 5345372622
|
|
|
|
| 1 |
version https://git-lfs.github.com/spec/v1
|
| 2 |
+
oid sha256:e81b0452eeddbeee05a4898b5a5242abb18c1e07673a124b54ba11618eb61aee
|
| 3 |
size 5345372622
|
model-00010-of-00019.safetensors
CHANGED
|
@@ -1,3 +1,3 @@
|
|
| 1 |
version https://git-lfs.github.com/spec/v1
|
| 2 |
-
oid sha256:
|
| 3 |
size 5365554368
|
|
|
|
| 1 |
version https://git-lfs.github.com/spec/v1
|
| 2 |
+
oid sha256:60c7aa16626bbbe48ea8d30ac78c0fc3673c8311f6bdd82423518afc4786bca9
|
| 3 |
size 5365554368
|
model-00011-of-00019.safetensors
CHANGED
|
@@ -1,3 +1,3 @@
|
|
| 1 |
version https://git-lfs.github.com/spec/v1
|
| 2 |
-
oid sha256:
|
| 3 |
size 5345372602
|
|
|
|
| 1 |
version https://git-lfs.github.com/spec/v1
|
| 2 |
+
oid sha256:0209d6938bfb5f6b8639d3e03984c4f7185ebbb3945aee5c64f49723b5bd8365
|
| 3 |
size 5345372602
|
model-00012-of-00019.safetensors
CHANGED
|
@@ -1,3 +1,3 @@
|
|
| 1 |
version https://git-lfs.github.com/spec/v1
|
| 2 |
-
oid sha256:
|
| 3 |
size 5345372650
|
|
|
|
| 1 |
version https://git-lfs.github.com/spec/v1
|
| 2 |
+
oid sha256:51cf4956341ecf1614fba6636bf92dd651300a12b47e9cb9a64c42c0fbd84fbb
|
| 3 |
size 5345372650
|
model-00013-of-00019.safetensors
CHANGED
|
@@ -1,3 +1,3 @@
|
|
| 1 |
version https://git-lfs.github.com/spec/v1
|
| 2 |
-
oid sha256:
|
| 3 |
size 5365554336
|
|
|
|
| 1 |
version https://git-lfs.github.com/spec/v1
|
| 2 |
+
oid sha256:0f13e49c1f3a0830e357a73f74d1392d99604649cfe51b7f35f807550d52cd69
|
| 3 |
size 5365554336
|
model-00014-of-00019.safetensors
CHANGED
|
@@ -1,3 +1,3 @@
|
|
| 1 |
version https://git-lfs.github.com/spec/v1
|
| 2 |
-
oid sha256:
|
| 3 |
size 5345372566
|
|
|
|
| 1 |
version https://git-lfs.github.com/spec/v1
|
| 2 |
+
oid sha256:d6bffb84a8078929a7e8da3fe689c501c19a00e1547534f996e623284963f69b
|
| 3 |
size 5345372566
|
model-00015-of-00019.safetensors
CHANGED
|
@@ -1,3 +1,3 @@
|
|
| 1 |
version https://git-lfs.github.com/spec/v1
|
| 2 |
-
oid sha256:
|
| 3 |
size 5345372634
|
|
|
|
| 1 |
version https://git-lfs.github.com/spec/v1
|
| 2 |
+
oid sha256:e94923c7907cfda4845affa200cf02df8244cb64b757c414b5319b56ea4b3d76
|
| 3 |
size 5345372634
|
model-00016-of-00019.safetensors
CHANGED
|
@@ -1,3 +1,3 @@
|
|
| 1 |
version https://git-lfs.github.com/spec/v1
|
| 2 |
-
oid sha256:
|
| 3 |
size 5365554352
|
|
|
|
| 1 |
version https://git-lfs.github.com/spec/v1
|
| 2 |
+
oid sha256:a2ea32e4b04cbb285bb7df35726292dcf06cb4dc45b9efb7383f718b47f36956
|
| 3 |
size 5365554352
|
model-00017-of-00019.safetensors
CHANGED
|
@@ -1,3 +1,3 @@
|
|
| 1 |
version https://git-lfs.github.com/spec/v1
|
| 2 |
-
oid sha256:
|
| 3 |
size 5345372620
|
|
|
|
| 1 |
version https://git-lfs.github.com/spec/v1
|
| 2 |
+
oid sha256:9078bb09764de95c98c1041bb4b8210a9f414c92ddfd634b20a8f6b8085484fd
|
| 3 |
size 5345372620
|
model-00018-of-00019.safetensors
CHANGED
|
@@ -1,3 +1,3 @@
|
|
| 1 |
version https://git-lfs.github.com/spec/v1
|
| 2 |
-
oid sha256:
|
| 3 |
size 5345372626
|
|
|
|
| 1 |
version https://git-lfs.github.com/spec/v1
|
| 2 |
+
oid sha256:171690828eb83e1611370bd2f71ba84a5910e19f2d3f311c0ce192a13c2267a1
|
| 3 |
size 5345372626
|
model-00019-of-00019.safetensors
CHANGED
|
@@ -1,3 +1,3 @@
|
|
| 1 |
version https://git-lfs.github.com/spec/v1
|
| 2 |
-
oid sha256:
|
| 3 |
size 4008648347
|
|
|
|
| 1 |
version https://git-lfs.github.com/spec/v1
|
| 2 |
+
oid sha256:bbfacce3a9f0a79710faa474be49d74cc6679e57428dcf76cf8e015ec0a71579
|
| 3 |
size 4008648347
|