The first question is perhaps a question of two parts - First, who is creating stories about the gods? Second, how are they preserved and propagated? Naturally, this question is a bit tricky because all we have are the fragments of oral tradition that were recorded several hundred years later. The exact nature of this relationship has been the subject of perpetual lively debate,

Part the First: Who is telling stories.

Pre-Christian Norse religions were, at least until the back half of the 10th century, fairly decentralized. Local communities were led by their chieftain, who is also the person in charge of ritual practices - the Old Norse word goði originally likely had a religious component, though it later came to refer to purely secular chieftains in Iceland. These chieftains would often hire poets, or skálds, to write poetry about their successes. These poems are sometimes explicitly mythological, but more often include oblique references to various myths through kennings, poetic formulae that refer to a mundane object via analogy (e.g. Tree of Weapons = warrior, Burden of Sigyn's Arms = Loki). Additionally, there were traveling religious leaders, like the Völur, who are often assigned the role of cosmological knowledge and therefore the telling of stories, divination, etc. However, these practices likely occurred in such a way that a very strong oral tradition developed. Simon Nygaard has read the poem Hakonarmal, telling of the arrival of a potentially-legendary Norwegian king Hakon to Valholl, as a participatory ritual, where the community would move into and out of the hall for various scenes in the poem. I know Skirnismal, one of the poems in the Poetic Edda, has also been read that way, and this community participation would absolutely help cement the story in people's minds.

Furthermore, if we postulate an ancient origin to the Icelandic Kvoldvaka, a wintertime storytelling tradition, it seems that lots of people in the community, but especially women, were responsible for the preservation and dissemination of stories not only for ritual purposes but for entertainment as well. Certainly, this seems to be the case even centuries after Christianization: Snorri Sturluson's father, Hvamm-Sturla, is said to have regularly invited people to his farmstead to tell him stories of the ancient gods (leading one of his enemies, a priest's wife, to shout "You should look like Odin, since you like him so much" while trying to stab out his eye. The attack missed.)

So, I think we can very safely conclude that pre-Christian Norse religions represented a very robust mnemonic tradition that a large part of society was involved in. What stories were being told would have absolutely depended on the community, and which beings they venerated - most stories that likely existed are lost to us, but in every case they are likely very strong traditions indeed.

Part the Second - How were these stories preserved?

If you've ever tried to remember a story, you know it's not always easy. Now, modern society doesn't train memory in the way oral societies do, but we happen to know quite well 3 primary mechanisms that assisted Norse people in preserving these stories.

Meter: Most material we have that we are confident is genuinely composed pre-Christianization is in the form of poetry, as mentioned above. There are numerous different meters, grouped into Eddic meters (i.e. those that the mythological poems of the Poetic Edda are composed in) and Skaldic meters (those that shorter, often secular, often praise-poems are composed in). These are incredibly complex, particularly the skaldic meter dróttkvætt (praise-meter), featuring potentially end-rhyme, alliteration, rigid stress patterns, internal rhyme, and semantic requirements for each half-stanza. The absurd complexity of these poems appears to have fossilized them for hundreds of years. Many of these poems were composed by specific poets, or skálds, in the court of a petty king (or at the end of the period, the kings of Norway/Denmark/Sweden or powerful earls like the Earls of Lade). These poems are often incorporated into prose saga texts, suggesting that perhaps the poetry was used as "supports" for larger stories that contextualize their composition (though we should not assume that the sagas are accurate in the context provided).

Runestones: One of our most helpful early sources are carved stones that were created throughout the Viking Age. From the 7th century Gutnish picture stones to 11th and 12th century runestones (with special shout-outs to hybrid pagan-Christian materials like the Gosforth Cross or Manx picture-crosses), it is extremely common to find mythological scenes pictorally represented. While we cannot always identify what story is being told, it is clear that they absolutely had known significances to the community that interacted with them. Runestones were expensive and a form of dominion over the landscape, and so the frequent depictions of Sigurdr killing the dragon Fafnir, for example, would have served as a reminder of those stories to the community. Why certain stories are shown on certain runestones is often a bit mysterious - the inscriptions generally have very little to do with the pictures that they exist in conjunction with, but their value as mnemonic markers is not diminished by that.

Landscapes: The last major mnemonic tool is found in the landscapes of Scandinavia and Iceland itself. Whether it's Lokasenna referring to the island of Samsø, or the long trend in sagas of having a sequence of "So-and-so did this-and-that here, and that's why it's called such-and-such", the landscape itself is a dominant mnemonic tool. By looking at a particular local feature, a story could be recalled about or around that feature, and therefore preserved. This could also be geologic events, too - Völuspá uses the features of a volcanic eruption to code the sequences of Ragnarök. It is, in the estimate of many scholars (such as Emily Lethbridge and Pernille Hermann) an indispensible tool for the oral traditions that emerge out of the Viking Age.

​

Now, an addendum for the second part: there simply isn't enough of the oral traditions preserved in order to say anything widespread about taboos or negative portrayals of features. There are secret rituals, such as the alfablot recorded in the skaldic poem Austarfararvisur, which are practiced in such a way as to prevent malicious spirits from entering the ritual space (the author of the poem calls down trolls onto the community who refuses to let him in, a perfect example of one such malicious entity). As to Loki themself - there are no placenames that suggest a worship-site for Loki, and little archaeological evidence for them. This could indicate that he was, until quite late in the period, taboo, or simply that he wasn't even counted as a god to be worshipped. Unfortunately, not enough survives to say anything about any community with surety. Certainly, by the 1200s, the ones who are still considered important enough to be addressed significantly in the Eddas have fairly robust and complex representations that are neither wholly positive nor wholly negative.