Text Generation
Transformers
Safetensors
MLX
English
mistral3
image-to-text
shining-valiant
shining-valiant-3
valiant
valiant-labs
mistral
mistral-common
ministral-3-14b
ministral
reasoning
code
code-reasoning
science
science-reasoning
physics
biology
chemistry
earth-science
astronomy
machine-learning
artificial-intelligence
compsci
computer-science
information-theory
ML-Ops
math
cuda
deep-learning
agentic
LLM
neuromorphic
self-improvement
complex-systems
cognition
linguistics
philosophy
logic
epistemology
simulation
game-theory
knowledge-management
creativity
problem-solving
architect
engineer
developer
creative
analytical
expert
rationality
conversational
chat
instruct
mlx-my-repo
4-bit precision
Upload config.json with huggingface_hub
Browse files- config.json +53 -0
config.json
ADDED
|
@@ -0,0 +1,53 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
{
|
| 2 |
+
"architectures": [
|
| 3 |
+
"Mistral3ForConditionalGeneration"
|
| 4 |
+
],
|
| 5 |
+
"dtype": "bfloat16",
|
| 6 |
+
"image_token_index": 10,
|
| 7 |
+
"model_type": "mistral3",
|
| 8 |
+
"multimodal_projector_bias": false,
|
| 9 |
+
"projector_hidden_act": "gelu",
|
| 10 |
+
"quantization": {
|
| 11 |
+
"group_size": 64,
|
| 12 |
+
"bits": 4,
|
| 13 |
+
"mode": "affine"
|
| 14 |
+
},
|
| 15 |
+
"quantization_config": {
|
| 16 |
+
"group_size": 64,
|
| 17 |
+
"bits": 4,
|
| 18 |
+
"mode": "affine"
|
| 19 |
+
},
|
| 20 |
+
"spatial_merge_size": 2,
|
| 21 |
+
"text_config": {
|
| 22 |
+
"attention_dropout": 0.0,
|
| 23 |
+
"head_dim": 128,
|
| 24 |
+
"hidden_act": "silu",
|
| 25 |
+
"hidden_size": 5120,
|
| 26 |
+
"initializer_range": 0.02,
|
| 27 |
+
"intermediate_size": 16384,
|
| 28 |
+
"max_position_embeddings": 262144,
|
| 29 |
+
"model_type": "ministral3",
|
| 30 |
+
"num_attention_heads": 32,
|
| 31 |
+
"num_hidden_layers": 40,
|
| 32 |
+
"num_key_value_heads": 8,
|
| 33 |
+
"rms_norm_eps": 1e-05,
|
| 34 |
+
"rope_parameters": {
|
| 35 |
+
"beta_fast": 32.0,
|
| 36 |
+
"beta_slow": 1.0,
|
| 37 |
+
"factor": 16.0,
|
| 38 |
+
"llama_4_scaling_beta": 0.1,
|
| 39 |
+
"mscale": 1.0,
|
| 40 |
+
"mscale_all_dim": 1.0,
|
| 41 |
+
"original_max_position_embeddings": 16384,
|
| 42 |
+
"rope_theta": 1000000000.0,
|
| 43 |
+
"rope_type": "yarn",
|
| 44 |
+
"type": "yarn"
|
| 45 |
+
},
|
| 46 |
+
"sliding_window": null,
|
| 47 |
+
"use_cache": true,
|
| 48 |
+
"vocab_size": 131072,
|
| 49 |
+
"tie_word_embeddings": false
|
| 50 |
+
},
|
| 51 |
+
"transformers_version": "5.0.0.dev0",
|
| 52 |
+
"vision_feature_layer": -1
|
| 53 |
+
}
|