Pclanglais commited on
Commit
4e34fb1
1 Parent(s): 687a6ca

Update README.md

Browse files
Files changed (1) hide show
  1. README.md +26 -4
README.md CHANGED
@@ -8,11 +8,13 @@ pipeline_tag: text-generation
8
 
9
  <div style="text-align: right;font-size:.7em;margin-left:50%"><em>Per un attimo Brahe cercò le parole, le immagini, le analogie; pensò perfino i gesti della mano e delle dita, come un attore che si prepari a rendere fisico un sentimento. Ma appena cominciò a dire "come", a dare solidità a ciò che non aveva, a rendere visibile ciò che non lo era, a collocare, nello spazio ciò che era pura probabilità, e a cercare una qualsiasi cosa tra le forme del mondo cui paragonarlo, Epstein lo interruppe.</em><br>Daniele del Giudice, <em>Atlante occidentale</em></div>
10
 
11
- *Brahe* is an analytical LLM for English literature fine-tuned from llama-13B. Given any text, Brahe will generate a list of potentially twenty annotations. Brahe is intended to be used by computational humanities project, similarly to BookNLP.
12
 
13
- *Brahe* has been trained on 4,000 excerpts of English or English translated literature in the public domain and on a set of synthetic and manual annotations.
14
 
15
- *Brahe* is the reversed companion of *[Epstein](https://huggingface.co/Pclanglais/epstein)*, a generative AI model to create new literary texts by submitting annotated prompts. Both models are named after the protagonists of the philosophical novel of Daniele del Giudice, *Atlante occidentale*. Brahe is a scientist working at the CERN on quantum physics, Epstein is a novelist and they both confront their different views of reality.
 
 
16
 
17
 
18
  ## Annotations
@@ -84,4 +86,24 @@ Output of Brahe:
84
  > Active character: The narrator, the aunt<br>
85
  > Time setting: Childhood<br>
86
  > Absolute place: Combray<br>
87
- > Fuzzy place: Combray
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
8
 
9
  <div style="text-align: right;font-size:.7em;margin-left:50%"><em>Per un attimo Brahe cercò le parole, le immagini, le analogie; pensò perfino i gesti della mano e delle dita, come un attore che si prepari a rendere fisico un sentimento. Ma appena cominciò a dire "come", a dare solidità a ciò che non aveva, a rendere visibile ciò che non lo era, a collocare, nello spazio ciò che era pura probabilità, e a cercare una qualsiasi cosa tra le forme del mondo cui paragonarlo, Epstein lo interruppe.</em><br>Daniele del Giudice, <em>Atlante occidentale</em></div>
10
 
11
+ *Brahe* is an analytical LLM for multilingual literature fine-tuned from llama-13B. Given any text, Brahe will generate a list of potentially twenty annotations. Brahe is intended to be used by computational humanities project, similarly to BookNLP.
12
 
13
+ *Brahe* has been trained on 8,000 excerpts of literature in the public domain and on a set of synthetic and manual annotations. Half of the excerpts are in English and half in other languages (mostly French, German, Italian…).
14
 
15
+ Thanks to the native multilinual capacity of llama-13B, *Brahe* has been proven to work on languages that were not part of its original corpus, such as Occitant.
16
+
17
+ *Brahe* is a reversed companion of *[Epstein](https://huggingface.co/Pclanglais/epstein)*, a generative AI model to create new literary texts by submitting annotated prompts (for now in English-only). Both models are named after the protagonists of the philosophical novel of Daniele del Giudice, *Atlante occidentale*. Brahe is a scientist working at the CERN on quantum physics, Epstein is a novelist and they both confront their different views of reality.
18
 
19
 
20
  ## Annotations
 
86
  > Active character: The narrator, the aunt<br>
87
  > Time setting: Childhood<br>
88
  > Absolute place: Combray<br>
89
+ > Fuzzy place: Combray
90
+
91
+ With an incipit of Nueit de junh by Joan Francés Blanc, in Gascon dialect of Occitan (a language not part of *Brahe* training corpus):
92
+
93
+ > Que voleré morí's. Non pas tant entà pèrder la vita, l'alen, tà deishà's escapar l'arma. Mes tà disaparéisher. Tà desbrembà's l'amor vadut aquiu, e tant viste perdut. Que voleré cercar solaç dens quauque ostau amic. Mes l'amic, lo sol, qui n'avèn pas hrèita de parlà's tà compréne's, que moriscó. E adara que vira a casa las pajas deus líbers shens legi'us. Que minja shens trobar lo gost ni brembà's lo nom deus plats. Que's passeja shens véder las caras deu monde qui tròba suu camin. No's bremba pas tanpauc deu tribalh, qui'u hè passar totun bèths moments en tot ocupar l'esperit a causas bestiaments materiaus : quin empachar los eslhèves de dromí's. Quin despertar la soa intelligéncia, que n'an ua, mes dens aqueth vegarau de París, n'ei pas aquò qui ajuda a vàder respectat... Òras e òras passadas (pergudas ?) en talhèrs de pedagogia. Sajar de dar idèas, de préner iniciativas, entà cargar un pauc mei l'aso de tribalh. Tà qui posqui pas tròp pensar a l'ivèrn d'avants. Decémer. Lo hred que cadè d'un còp sus la vila. Que passè de dètz degrès a mensh cinc dens la nueit. Que vedó lo lendoman lo mond qui's torran qui s'an desbrembat d'escotar la ràdio. Vita shens nat chepic que lo tribalh quotidian. Puish lo telefòne, dens la nueit, d'islas de luenh, tà díser la mort de l'amic. Que partiva tau tribalh, dens l'autò crompada dab quauques milèrs de quilomètres. E los travaders qui ne'u vòlen pas créder. Lo camion en fàcia. Lo néguer. La hemna qui quitè de vàder hòla. Òras au telefòne. Amics mei disponibles, qui pòden deishar lo tribalh tà gahar lo purmèr avion. Nau òras dinc a Pointe-à-Pitre. E eth, esperar lo ser, cada ser, avants o après lo telefòne, tà plorà's. Shens masclor. O dab la masclor deu gojat qui plorèva dab la casa sancèra aus enterraments. Preparar lo viatge. Esperar un mes tàd'aquò. Aver lo temps de crompar un disc de Lluís Llach. Gira la roda del temps... i si no fós absurd el mes enllà... Que pensa a la soa mort. Non se vei víver shens aquera preséncia de l'amic. Qu'èra partit luenh, sovent. Lille, a bisa. E après, qu'èra partit eth, tà París. Mes adara, qu'ei un partir definitiu. A maugrat la fe.
94
+
95
+ Output of Brahe:
96
+
97
+ > Summary: The text is a series of thoughts and reflections on death, friendship, and the passage of time.
98
+ > Trope: The passage of time, the fleeting nature of life
99
+ > Narrative arc: Reflective, introspective
100
+ > Enunciation: First-person narrative
101
+ > Tone: Reflective, melancholic
102
+ > Genre: Poetry
103
+ > Intertextuality: Poetry
104
+ > Speech standard: Poetic, elevated language
105
+ > Literary form: Stream of consciousness
106
+ > Literary movement: Existentialism
107
+ > Time setting: Contemporary
108
+ > Fuzzy place: Unnamed village, Paris, Pointe-à-Pitre
109
+