Download transitions.json from DrJohnCode/NebulaText: direct link, hf CLI and curl.
- Browser
- Download file 5.02 kB
-
https://huggingface.co/DrJohnCode/NebulaText/resolve/main/transitions.json
- Command line
-
hf download hf://DrJohnCode/NebulaText/transitions.json
-
curl -L -o transitions.json https://huggingface.co/DrJohnCode/NebulaText/resolve/main/transitions.json
5.02 kB
| { | |
| ".": { | |
| "<END>": 20 | |
| }, | |
| "<START>": { | |
| "The": 3, | |
| "AI": 2, | |
| "Small": 2, | |
| "Researchers": 2, | |
| "A": 3, | |
| "GPU": 1, | |
| "Efficient": 1, | |
| "NebulaText": 1, | |
| "This": 1, | |
| "Reproducible": 1, | |
| "Language": 1, | |
| "An": 1, | |
| "Transparent": 1 | |
| }, | |
| "A": { | |
| "synthetic": 1, | |
| "model": 1, | |
| "Markov": 1 | |
| }, | |
| "AI": { | |
| "depends": 1, | |
| "includes": 1, | |
| "systems": 2 | |
| }, | |
| "An": { | |
| "experimental": 1 | |
| }, | |
| "Efficient": { | |
| "inference": 1 | |
| }, | |
| "GPU": { | |
| "acceleration": 1 | |
| }, | |
| "Language": { | |
| "modeling": 1 | |
| }, | |
| "Markov": { | |
| "chain": 1 | |
| }, | |
| "NebulaText": { | |
| "is": 1 | |
| }, | |
| "Reproducible": { | |
| "experiments": 1 | |
| }, | |
| "Researchers": { | |
| "explore": 1, | |
| "test": 1 | |
| }, | |
| "Small": { | |
| "models": 2 | |
| }, | |
| "The": { | |
| "future": 2, | |
| "demonstration": 1 | |
| }, | |
| "This": { | |
| "repository": 1 | |
| }, | |
| "Transparent": { | |
| "documentation": 1 | |
| }, | |
| "a": { | |
| "substitute": 1, | |
| "fictional": 1, | |
| "small": 1, | |
| "table": 1 | |
| }, | |
| "acceleration": { | |
| "can": 1 | |
| }, | |
| "and": { | |
| "clear": 1, | |
| "responsible": 1, | |
| "summarize": 1, | |
| "limitations": 1 | |
| }, | |
| "architectures": { | |
| ".": 1 | |
| }, | |
| "are": { | |
| "useful": 1 | |
| }, | |
| "baselines": { | |
| ".": 1 | |
| }, | |
| "basics": { | |
| "of": 1 | |
| }, | |
| "be": { | |
| "introduced": 1 | |
| }, | |
| "can": { | |
| "help": 1, | |
| "process": 1, | |
| "demonstrate": 1, | |
| "support": 1, | |
| "be": 1 | |
| }, | |
| "card": { | |
| "should": 1 | |
| }, | |
| "careful": { | |
| "research": 1, | |
| "measurement": 1 | |
| }, | |
| "chain": { | |
| "predicts": 1 | |
| }, | |
| "claim": { | |
| "unmeasured": 1 | |
| }, | |
| "clear": { | |
| "evaluation": 1 | |
| }, | |
| "concept": { | |
| ".": 1 | |
| }, | |
| "concepts": { | |
| "from": 1 | |
| }, | |
| "contains": { | |
| "a": 1 | |
| }, | |
| "creative": { | |
| "ideas": 1 | |
| }, | |
| "data": { | |
| "and": 1 | |
| }, | |
| "dataset": { | |
| "is": 1 | |
| }, | |
| "demo": { | |
| ".": 1 | |
| }, | |
| "demonstrate": { | |
| "the": 1 | |
| }, | |
| "demonstration": { | |
| "samples": 1 | |
| }, | |
| "depends": { | |
| "on": 1 | |
| }, | |
| "deployment": { | |
| ".": 1 | |
| }, | |
| "documentation": { | |
| "makes": 1 | |
| }, | |
| "easier": { | |
| "to": 1 | |
| }, | |
| "educational": { | |
| "demo": 1 | |
| }, | |
| "efficient": { | |
| "models": 1 | |
| }, | |
| "evaluate": { | |
| ".": 1 | |
| }, | |
| "evaluation": { | |
| ".": 2 | |
| }, | |
| "experimental": { | |
| "prototype": 1 | |
| }, | |
| "experiments": { | |
| ".": 1, | |
| "help": 1 | |
| }, | |
| "explain": { | |
| "training": 1 | |
| }, | |
| "explore": { | |
| "creative": 1, | |
| "memory": 1 | |
| }, | |
| "fictional": { | |
| "research": 1 | |
| }, | |
| "for": { | |
| "quick": 1, | |
| "real": 1 | |
| }, | |
| "from": { | |
| "results": 1, | |
| "the": 1, | |
| "a": 1 | |
| }, | |
| "future": { | |
| "of": 2 | |
| }, | |
| "generation": { | |
| ".": 1 | |
| }, | |
| "help": { | |
| "people": 1, | |
| "separate": 1 | |
| }, | |
| "hypothetical": { | |
| "architectures": 1 | |
| }, | |
| "ideas": { | |
| ".": 1, | |
| "with": 1 | |
| }, | |
| "in": { | |
| "hypothetical": 1 | |
| }, | |
| "includes": { | |
| "efficient": 1 | |
| }, | |
| "inference": { | |
| "requires": 1 | |
| }, | |
| "introduced": { | |
| "with": 1 | |
| }, | |
| "is": { | |
| "not": 1, | |
| "a": 1 | |
| }, | |
| "large": { | |
| "neural": 1 | |
| }, | |
| "limitations": { | |
| ".": 1 | |
| }, | |
| "makes": { | |
| "software": 1 | |
| }, | |
| "measurement": { | |
| ".": 1 | |
| }, | |
| "memory": { | |
| "routing": 1 | |
| }, | |
| "model": { | |
| "card": 1 | |
| }, | |
| "modeling": { | |
| "can": 1 | |
| }, | |
| "models": { | |
| "and": 1, | |
| "are": 1, | |
| "can": 1 | |
| }, | |
| "networks": { | |
| ".": 1 | |
| }, | |
| "neural": { | |
| "networks": 1 | |
| }, | |
| "next": { | |
| "word": 1 | |
| }, | |
| "not": { | |
| "a": 1, | |
| "claim": 1 | |
| }, | |
| "observed": { | |
| "transitions": 1 | |
| }, | |
| "of": { | |
| "AI": 2, | |
| "text": 1, | |
| "observed": 1 | |
| }, | |
| "on": { | |
| "careful": 1 | |
| }, | |
| "patterns": { | |
| ".": 1 | |
| }, | |
| "people": { | |
| "explore": 1 | |
| }, | |
| "performance": { | |
| ".": 1 | |
| }, | |
| "predicts": { | |
| "the": 1 | |
| }, | |
| "previous": { | |
| "state": 1 | |
| }, | |
| "process": { | |
| "text": 1 | |
| }, | |
| "prototype": { | |
| "should": 1 | |
| }, | |
| "quick": { | |
| "experiments": 1 | |
| }, | |
| "real": { | |
| "evaluation": 1 | |
| }, | |
| "repository": { | |
| "contains": 1 | |
| }, | |
| "requires": { | |
| "careful": 1 | |
| }, | |
| "research": { | |
| "and": 1, | |
| "concept": 1 | |
| }, | |
| "responsible": { | |
| "deployment": 1 | |
| }, | |
| "results": { | |
| ".": 1 | |
| }, | |
| "routing": { | |
| "in": 1 | |
| }, | |
| "samples": { | |
| "from": 1 | |
| }, | |
| "separate": { | |
| "concepts": 1 | |
| }, | |
| "should": { | |
| "explain": 1, | |
| "not": 1 | |
| }, | |
| "simple": { | |
| "patterns": 1 | |
| }, | |
| "small": { | |
| "educational": 1 | |
| }, | |
| "software": { | |
| "easier": 1 | |
| }, | |
| "state": { | |
| ".": 1 | |
| }, | |
| "substitute": { | |
| "for": 1 | |
| }, | |
| "summarize": { | |
| "simple": 1 | |
| }, | |
| "support": { | |
| "large": 1 | |
| }, | |
| "synthetic": { | |
| "dataset": 1 | |
| }, | |
| "systems": { | |
| "can": 2 | |
| }, | |
| "table": { | |
| "of": 1 | |
| }, | |
| "test": { | |
| "ideas": 1 | |
| }, | |
| "text": { | |
| "and": 1, | |
| "generation": 1 | |
| }, | |
| "the": { | |
| "basics": 1, | |
| "next": 1, | |
| "previous": 1 | |
| }, | |
| "to": { | |
| "evaluate": 1 | |
| }, | |
| "training": { | |
| "data": 1 | |
| }, | |
| "transitions": { | |
| ".": 2 | |
| }, | |
| "transparent": { | |
| "baselines": 1 | |
| }, | |
| "unmeasured": { | |
| "performance": 1 | |
| }, | |
| "useful": { | |
| "for": 1 | |
| }, | |
| "with": { | |
| "transparent": 1, | |
| "word": 1 | |
| }, | |
| "word": { | |
| "transitions": 1, | |
| "from": 1 | |
| } | |
| } | |