AxionLab-official commited on
Commit
f3846ce
·
verified ·
1 Parent(s): bc9bf49

Update README.md

Browse files
Files changed (1) hide show
  1. README.md +6 -6
README.md CHANGED
@@ -31,7 +31,7 @@ A functional “blank mind” before reasoning specialization
31
 
32
  What this model is not
33
 
34
- ❌ Not a chatbot
35
 
36
  ❌ Not instruction-tuned
37
 
@@ -58,7 +58,7 @@ A clear separation between language competence and reasoning behavior
58
  Many projects hide their base models.
59
  Axion does the opposite.
60
 
61
- Intended use
62
 
63
  Research and experimentation
64
 
@@ -68,7 +68,7 @@ Studying the effects of reasoning-oriented datasets
68
 
69
  Serving as a backbone for Axion1.5-Reasoning variants
70
 
71
- Limitations
72
 
73
  Because this model is trained only for next-token prediction:
74
 
@@ -82,7 +82,7 @@ It may hallucinate or contradict itself
82
 
83
  These limitations are expected and acknowledged.
84
 
85
- Future work
86
 
87
  This release is part of a broader project:
88
 
@@ -94,7 +94,7 @@ Experiments with short, verifiable reasoning traces
94
 
95
  The base model will remain unchanged to preserve its value as a reference.
96
 
97
- Philosophy
98
 
99
  Scale is not intelligence.
100
  Structure matters.
@@ -104,6 +104,6 @@ Axion explores whether smaller models, trained with the right constraints, can d
104
  This is an experiment.
105
  And experiments are allowed to fail.
106
 
107
- Acknowledgements
108
 
109
  Created as an independent research project focused on understanding how reasoning emerges in language models.
 
31
 
32
  What this model is not
33
 
34
+ **Not a chatbot**
35
 
36
  ❌ Not instruction-tuned
37
 
 
58
  Many projects hide their base models.
59
  Axion does the opposite.
60
 
61
+ **Intended use**
62
 
63
  Research and experimentation
64
 
 
68
 
69
  Serving as a backbone for Axion1.5-Reasoning variants
70
 
71
+ **Limitations**
72
 
73
  Because this model is trained only for next-token prediction:
74
 
 
82
 
83
  These limitations are expected and acknowledged.
84
 
85
+ ****Future work**
86
 
87
  This release is part of a broader project:
88
 
 
94
 
95
  The base model will remain unchanged to preserve its value as a reference.
96
 
97
+ **Philosophy**
98
 
99
  Scale is not intelligence.
100
  Structure matters.
 
104
  This is an experiment.
105
  And experiments are allowed to fail.
106
 
107
+ **Acknowledgements**
108
 
109
  Created as an independent research project focused on understanding how reasoning emerges in language models.