mradermacher/Llama-3.1-Discord-Short-8B-i1-GGUF
8B • Updated • 93
id stringlengths 1 4 | conversations listlengths 126 523 |
|---|---|
1 | [
{
"from": "human",
"value": "Hagane no Moonsault: speakerphone\nspeakerphone\nspeakerphone\nuserphone\nMoshi mosh\n?\n?\nAre you drunk perhaps?\nuserphone\nuserphone\nuserphone\nuserphone\nHello there\nuserphone\nHello there\nSir/Ma'am?\nANYONE\nUSERPHONE\nuserphone\nuserphone\nHello\nAw\nBye gn\nIts 12 her... |
2 | [
{
"from": "gpt",
"value": "Cookeyjar: @Velvet! food poisoning/general illness"
},
{
"from": "human",
"value": "Velvet!: ah I see"
},
{
"from": "gpt",
"value": "Hagane no Moonsault: I got newwwsss"
},
{
"from": "human",
"value": "Cookeyjar: Oooh\nLet's hear it\n@Hagane no ... |
3 | [
{
"from": "human",
"value": "Woofles: The Y must stand for Yaoi"
},
{
"from": "gpt",
"value": "Nii!: Thats free yall"
},
{
"from": "human",
"value": "Woofles: Oh\nDisappointing"
},
{
"from": "gpt",
"value": "Nii!: not a yaoi anime"
},
{
"from": "human",
"value... |
4 | [
{
"from": "gpt",
"value": "Zael: I also have an Elesis I'd like to have in the guild, but not in game rn, we'll do that later"
},
{
"from": "human",
"value": "Hagane no Moonsault: ok, just tell me when\nwb\nstill loud af here\noh?\nyea move it"
},
{
"from": "gpt",
"value": "Nii!: mov... |
5 | [
{
"from": "gpt",
"value": "Velvet!: like they dont sting me\nThey just fly at me\nits fuckin scary\nand then my friends think I'm a little bitch for jumping when I see the color yellow"
},
{
"from": "human",
"value": "Hagane no Moonsault: Any kind of yellow?\nOr just like if something had a litt... |
6 | [
{
"from": "gpt",
"value": "Shizumu: What's up"
},
{
"from": "human",
"value": "Lalvie: I'm fine.\nhbu?"
},
{
"from": "gpt",
"value": "Shizumu: Dying.\nbut good\nand spamming"
},
{
"from": "human",
"value": "Lalvie: I have a question"
},
{
"from": "gpt",
"value... |
7 | [
{
"from": "gpt",
"value": "Astawyr: Every single new girl that come in to work is i. There teens bro like 15 to 16"
},
{
"from": "human",
"value": "Hagane no Moonsault: Dude"
},
{
"from": "gpt",
"value": "Velvet!: at least you can look at memes?"
},
{
"from": "human",
"va... |
8 | [{"from":"gpt","value":"Hagane no Moonsault: w.claim hildegard"},{"from":"human","value":"Velvet!: w(...TRUNCATED) |
9 | [{"from":"gpt","value":"Lalvie: I don't think I will get allowed to.."},{"from":"human","value":"Vel(...TRUNCATED) |
10 | [{"from":"human","value":"Lalvie: God\nKami"},{"from":"gpt","value":"Velvet!: im neither kami nor pi(...TRUNCATED) |
Single channel's conversations turned into ShareGPT format from Discord-Data. It has also been optimized for Llama 3.1 tokenizer with each conversation being max 8192 tokens.
Since I'm using Unsloth I had to add another adjustment, for some reason it adds a 28 token length system prompt to each conversation so I also need to account for that in this dataset, and Llama 3.1 format also uses 7 tokens per message just for the formatting.